{
  "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": 40.29560017448091,
      "language_loss": 2.5798173,
      "learning_rate": 0.0,
      "loss": 1.90189219,
      "num_input_tokens_seen": 20375,
      "step": 1,
      "time_per_iteration": 13.533031463623047
    },
    {
      "auxiliary_loss_clip": 0.03096462,
      "auxiliary_loss_mlp": 0.016449,
      "balance_loss_clip": 1.65451014,
      "balance_loss_mlp": 1.32103169,
      "epoch": 0.00024048578127818193,
      "flos": 30664624377600.0,
      "grad_norm": 55.94489656313446,
      "language_loss": 1.89316106,
      "learning_rate": 5.021476677069823e-07,
      "loss": 1.94057465,
      "num_input_tokens_seen": 39035,
      "step": 2,
      "time_per_iteration": 2.4680161476135254
    },
    {
      "auxiliary_loss_clip": 0.03069692,
      "auxiliary_loss_mlp": 0.01669571,
      "balance_loss_clip": 1.65272069,
      "balance_loss_mlp": 1.34818316,
      "epoch": 0.0003607286719172729,
      "flos": 19026227969280.0,
      "grad_norm": 40.107296391464416,
      "language_loss": 1.61673808,
      "learning_rate": 7.958852231401551e-07,
      "loss": 1.66413069,
      "num_input_tokens_seen": 57600,
      "step": 3,
      "time_per_iteration": 2.34382700920105
    },
    {
      "auxiliary_loss_clip": 0.03077994,
      "auxiliary_loss_mlp": 0.0173602,
      "balance_loss_clip": 1.65171075,
      "balance_loss_mlp": 1.41310573,
      "epoch": 0.00048097156255636386,
      "flos": 19316314206720.0,
      "grad_norm": 36.87690991080161,
      "language_loss": 1.64510322,
      "learning_rate": 1.0042953354139647e-06,
      "loss": 1.69324338,
      "num_input_tokens_seen": 76465,
      "step": 4,
      "time_per_iteration": 2.4014604091644287
    },
    {
      "auxiliary_loss_clip": 0.03073864,
      "auxiliary_loss_mlp": 0.01667524,
      "balance_loss_clip": 1.65356755,
      "balance_loss_mlp": 1.35071325,
      "epoch": 0.0006012144531954548,
      "flos": 13991264893440.0,
      "grad_norm": 55.24854794788945,
      "language_loss": 1.93569183,
      "learning_rate": 1.1659507774310057e-06,
      "loss": 1.98310578,
      "num_input_tokens_seen": 94350,
      "step": 5,
      "time_per_iteration": 2.6521339416503906
    },
    {
      "auxiliary_loss_clip": 0.03084522,
      "auxiliary_loss_mlp": 0.01682753,
      "balance_loss_clip": 1.65810752,
      "balance_loss_mlp": 1.36117351,
      "epoch": 0.0007214573438345458,
      "flos": 23148988225920.0,
      "grad_norm": 45.194418366608836,
      "language_loss": 1.61075056,
      "learning_rate": 1.2980328908471373e-06,
      "loss": 1.6584233,
      "num_input_tokens_seen": 114595,
      "step": 6,
      "time_per_iteration": 2.793560028076172
    },
    {
      "auxiliary_loss_clip": 0.03138458,
      "auxiliary_loss_mlp": 0.01606507,
      "balance_loss_clip": 1.79562807,
      "balance_loss_mlp": 1.40661681,
      "epoch": 0.0008417002344736367,
      "flos": 67663246170240.0,
      "grad_norm": 4.613852836749538,
      "language_loss": 0.81504482,
      "learning_rate": 1.4097067265369432e-06,
      "loss": 0.86249447,
      "num_input_tokens_seen": 179590,
      "step": 7,
      "time_per_iteration": 3.0909554958343506
    },
    {
      "auxiliary_loss_clip": 0.0305004,
      "auxiliary_loss_mlp": 0.0170907,
      "balance_loss_clip": 1.64211285,
      "balance_loss_mlp": 1.39588332,
      "epoch": 0.0009619431251127277,
      "flos": 21281381504640.0,
      "grad_norm": 49.659725106120916,
      "language_loss": 1.582201,
      "learning_rate": 1.506443003120947e-06,
      "loss": 1.62979209,
      "num_input_tokens_seen": 195090,
      "step": 8,
      "time_per_iteration": 2.709663152694702
    },
    {
      "auxiliary_loss_clip": 0.03055894,
      "auxiliary_loss_mlp": 0.01697121,
      "balance_loss_clip": 1.64996243,
      "balance_loss_mlp": 1.37897575,
      "epoch": 0.0010821860157518186,
      "flos": 23331342597120.0,
      "grad_norm": 17.508890825397188,
      "language_loss": 1.47852767,
      "learning_rate": 1.5917704462803102e-06,
      "loss": 1.52605772,
      "num_input_tokens_seen": 211635,
      "step": 9,
      "time_per_iteration": 2.653956174850464
    },
    {
      "auxiliary_loss_clip": 0.03042046,
      "auxiliary_loss_mlp": 0.01654968,
      "balance_loss_clip": 1.64881778,
      "balance_loss_mlp": 1.3379668,
      "epoch": 0.0012024289063909096,
      "flos": 17010166337280.0,
      "grad_norm": 13.44495182275048,
      "language_loss": 1.52919805,
      "learning_rate": 1.6680984451379884e-06,
      "loss": 1.57616818,
      "num_input_tokens_seen": 224705,
      "step": 10,
      "time_per_iteration": 2.69167423248291
    },
    {
      "auxiliary_loss_clip": 0.03049854,
      "auxiliary_loss_mlp": 0.01682941,
      "balance_loss_clip": 1.64465737,
      "balance_loss_mlp": 1.37852824,
      "epoch": 0.0013226717970300007,
      "flos": 21288133261440.0,
      "grad_norm": 14.412222420014784,
      "language_loss": 1.3266654,
      "learning_rate": 1.7371455188905097e-06,
      "loss": 1.3739934,
      "num_input_tokens_seen": 244635,
      "step": 11,
      "time_per_iteration": 2.680786371231079
    },
    {
      "auxiliary_loss_clip": 0.03064559,
      "auxiliary_loss_mlp": 0.0170214,
      "balance_loss_clip": 1.64962065,
      "balance_loss_mlp": 1.37445784,
      "epoch": 0.0014429146876690916,
      "flos": 27237884935680.0,
      "grad_norm": 11.478639317387843,
      "language_loss": 1.25377536,
      "learning_rate": 1.8001805585541196e-06,
      "loss": 1.30144238,
      "num_input_tokens_seen": 265765,
      "step": 12,
      "time_per_iteration": 2.7420272827148438
    },
    {
      "auxiliary_loss_clip": 0.03044011,
      "auxiliary_loss_mlp": 0.01663496,
      "balance_loss_clip": 1.64189076,
      "balance_loss_mlp": 1.35603178,
      "epoch": 0.0015631575783081825,
      "flos": 19062174504960.0,
      "grad_norm": 6.593714778584316,
      "language_loss": 1.29324901,
      "learning_rate": 1.8581671739548328e-06,
      "loss": 1.34032416,
      "num_input_tokens_seen": 283500,
      "step": 13,
      "time_per_iteration": 2.669037103652954
    },
    {
      "auxiliary_loss_clip": 0.03039403,
      "auxiliary_loss_mlp": 0.01618412,
      "balance_loss_clip": 1.63795567,
      "balance_loss_mlp": 1.30617929,
      "epoch": 0.0016834004689472734,
      "flos": 48139473985920.0,
      "grad_norm": 6.260191131609957,
      "language_loss": 1.1350081,
      "learning_rate": 1.9118543942439254e-06,
      "loss": 1.18158627,
      "num_input_tokens_seen": 305685,
      "step": 14,
      "time_per_iteration": 2.8571760654449463
    },
    {
      "auxiliary_loss_clip": 0.03015065,
      "auxiliary_loss_mlp": 0.01677578,
      "balance_loss_clip": 1.6330477,
      "balance_loss_mlp": 1.36076784,
      "epoch": 0.0018036433595863645,
      "flos": 34970026314240.0,
      "grad_norm": 5.78064015765431,
      "language_loss": 1.12634933,
      "learning_rate": 1.961836000571161e-06,
      "loss": 1.17327571,
      "num_input_tokens_seen": 327340,
      "step": 15,
      "time_per_iteration": 2.760812997817993
    },
    {
      "auxiliary_loss_clip": 0.03027068,
      "auxiliary_loss_mlp": 0.01439522,
      "balance_loss_clip": 1.77060544,
      "balance_loss_mlp": 1.25031328,
      "epoch": 0.0019238862502254555,
      "flos": 59768284440960.0,
      "grad_norm": 3.756658059109441,
      "language_loss": 0.64692342,
      "learning_rate": 2.0085906708279293e-06,
      "loss": 0.69158936,
      "num_input_tokens_seen": 382710,
      "step": 16,
      "time_per_iteration": 5.50650429725647
    },
    {
      "auxiliary_loss_clip": 0.03001511,
      "auxiliary_loss_mlp": 0.01635746,
      "balance_loss_clip": 1.63701475,
      "balance_loss_mlp": 1.32484829,
      "epoch": 0.0020441291408645466,
      "flos": 20814543417600.0,
      "grad_norm": 4.349171996444234,
      "language_loss": 1.160671,
      "learning_rate": 2.0525099325728135e-06,
      "loss": 1.20704341,
      "num_input_tokens_seen": 400890,
      "step": 17,
      "time_per_iteration": 2.6450772285461426
    },
    {
      "auxiliary_loss_clip": 0.02990377,
      "auxiliary_loss_mlp": 0.01400776,
      "balance_loss_clip": 1.76002371,
      "balance_loss_mlp": 1.21461892,
      "epoch": 0.0021643720315036373,
      "flos": 63857001582720.0,
      "grad_norm": 3.5101266357191325,
      "language_loss": 0.72124314,
      "learning_rate": 2.0939181139872922e-06,
      "loss": 0.76515466,
      "num_input_tokens_seen": 462605,
      "step": 18,
      "time_per_iteration": 3.085275173187256
    },
    {
      "auxiliary_loss_clip": 0.02970693,
      "auxiliary_loss_mlp": 0.01583358,
      "balance_loss_clip": 1.63045931,
      "balance_loss_mlp": 1.28276038,
      "epoch": 0.0022846149221427284,
      "flos": 31284981192960.0,
      "grad_norm": 4.7102787974699085,
      "language_loss": 1.01558769,
      "learning_rate": 2.1330868934640175e-06,
      "loss": 1.06112814,
      "num_input_tokens_seen": 483280,
      "step": 19,
      "time_per_iteration": 2.696502208709717
    },
    {
      "auxiliary_loss_clip": 0.02941308,
      "auxiliary_loss_mlp": 0.01372513,
      "balance_loss_clip": 1.74952602,
      "balance_loss_mlp": 1.19017053,
      "epoch": 0.002404857812781819,
      "flos": 51083648161920.0,
      "grad_norm": 3.555214763055384,
      "language_loss": 0.76406908,
      "learning_rate": 2.170246112844971e-06,
      "loss": 0.80720729,
      "num_input_tokens_seen": 537620,
      "step": 20,
      "time_per_iteration": 2.8700852394104004
    },
    {
      "auxiliary_loss_clip": 0.02914774,
      "auxiliary_loss_mlp": 0.01538777,
      "balance_loss_clip": 1.61850083,
      "balance_loss_mlp": 1.23951364,
      "epoch": 0.0025251007034209102,
      "flos": 15815347309440.0,
      "grad_norm": 4.735304505590101,
      "language_loss": 1.01388359,
      "learning_rate": 2.2055919496770983e-06,
      "loss": 1.05841899,
      "num_input_tokens_seen": 555760,
      "step": 21,
      "time_per_iteration": 2.647648811340332
    },
    {
      "auxiliary_loss_clip": 0.02901449,
      "auxiliary_loss_mlp": 0.01523067,
      "balance_loss_clip": 1.61415291,
      "balance_loss_mlp": 1.22533035,
      "epoch": 0.0026453435940600014,
      "flos": 37851857458560.0,
      "grad_norm": 5.2540440972606985,
      "language_loss": 0.8961674,
      "learning_rate": 2.2392931865974923e-06,
      "loss": 0.94041252,
      "num_input_tokens_seen": 578450,
      "step": 22,
      "time_per_iteration": 2.8207290172576904
    },
    {
      "auxiliary_loss_clip": 0.02862227,
      "auxiliary_loss_mlp": 0.01506672,
      "balance_loss_clip": 1.60537946,
      "balance_loss_mlp": 1.21026993,
      "epoch": 0.002765586484699092,
      "flos": 21141976821120.0,
      "grad_norm": 4.620206115742967,
      "language_loss": 1.01830506,
      "learning_rate": 2.271496085962064e-06,
      "loss": 1.06199408,
      "num_input_tokens_seen": 596145,
      "step": 23,
      "time_per_iteration": 2.69580078125
    },
    {
      "auxiliary_loss_clip": 0.02834849,
      "auxiliary_loss_mlp": 0.01483805,
      "balance_loss_clip": 1.59239292,
      "balance_loss_mlp": 1.19102716,
      "epoch": 0.002885829375338183,
      "flos": 20667381396480.0,
      "grad_norm": 3.2009452444972335,
      "language_loss": 1.02576804,
      "learning_rate": 2.3023282262611022e-06,
      "loss": 1.06895471,
      "num_input_tokens_seen": 614920,
      "step": 24,
      "time_per_iteration": 2.6719298362731934
    },
    {
      "auxiliary_loss_clip": 0.02845174,
      "auxiliary_loss_mlp": 0.01484913,
      "balance_loss_clip": 1.60035145,
      "balance_loss_mlp": 1.20014632,
      "epoch": 0.003006072265977274,
      "flos": 34823869873920.0,
      "grad_norm": 3.8772769835335987,
      "language_loss": 0.92433882,
      "learning_rate": 2.3319015548620114e-06,
      "loss": 0.96763968,
      "num_input_tokens_seen": 636060,
      "step": 25,
      "time_per_iteration": 2.8007843494415283
    },
    {
      "auxiliary_loss_clip": 0.02802626,
      "auxiliary_loss_mlp": 0.01451476,
      "balance_loss_clip": 1.58807039,
      "balance_loss_mlp": 1.17510104,
      "epoch": 0.003126315156616365,
      "flos": 24422021118720.0,
      "grad_norm": 2.271328937384169,
      "language_loss": 0.92861021,
      "learning_rate": 2.3603148416618152e-06,
      "loss": 0.97115123,
      "num_input_tokens_seen": 655575,
      "step": 26,
      "time_per_iteration": 2.705730438232422
    },
    {
      "auxiliary_loss_clip": 0.02808334,
      "auxiliary_loss_mlp": 0.01435545,
      "balance_loss_clip": 1.58833385,
      "balance_loss_mlp": 1.16241288,
      "epoch": 0.003246558047255456,
      "flos": 23622326674560.0,
      "grad_norm": 2.294819173954996,
      "language_loss": 1.00859427,
      "learning_rate": 2.3876556694204647e-06,
      "loss": 1.05103302,
      "num_input_tokens_seen": 675730,
      "step": 27,
      "time_per_iteration": 2.707465410232544
    },
    {
      "auxiliary_loss_clip": 0.02768539,
      "auxiliary_loss_mlp": 0.01436921,
      "balance_loss_clip": 1.58165073,
      "balance_loss_mlp": 1.15062785,
      "epoch": 0.003366800937894547,
      "flos": 17820275725440.0,
      "grad_norm": 2.5769542258057805,
      "language_loss": 0.90632999,
      "learning_rate": 2.414002061950908e-06,
      "loss": 0.94838458,
      "num_input_tokens_seen": 694605,
      "step": 28,
      "time_per_iteration": 2.656536340713501
    },
    {
      "auxiliary_loss_clip": 0.0275307,
      "auxiliary_loss_mlp": 0.01410387,
      "balance_loss_clip": 1.5744226,
      "balance_loss_mlp": 1.14259565,
      "epoch": 0.003487043828533638,
      "flos": 24426115269120.0,
      "grad_norm": 2.3808094420721275,
      "language_loss": 0.9981423,
      "learning_rate": 2.4394238264681557e-06,
      "loss": 1.0397768,
      "num_input_tokens_seen": 714340,
      "step": 29,
      "time_per_iteration": 2.68894624710083
    },
    {
      "auxiliary_loss_clip": 0.02727111,
      "auxiliary_loss_mlp": 0.01410366,
      "balance_loss_clip": 1.56732988,
      "balance_loss_mlp": 1.1368525,
      "epoch": 0.003607286719172729,
      "flos": 26140311002880.0,
      "grad_norm": 2.0666836282045256,
      "language_loss": 0.99509072,
      "learning_rate": 2.4639836682781433e-06,
      "loss": 1.03646541,
      "num_input_tokens_seen": 734470,
      "step": 30,
      "time_per_iteration": 2.7190186977386475
    },
    {
      "auxiliary_loss_clip": 0.02741221,
      "auxiliary_loss_mlp": 0.01397796,
      "balance_loss_clip": 1.5814749,
      "balance_loss_mlp": 1.11760616,
      "epoch": 0.00372752960981182,
      "flos": 20593082113920.0,
      "grad_norm": 3.2524685202632178,
      "language_loss": 1.00112808,
      "learning_rate": 2.487738122623307e-06,
      "loss": 1.04251838,
      "num_input_tokens_seen": 753380,
      "step": 31,
      "time_per_iteration": 2.6225290298461914
    },
    {
      "auxiliary_loss_clip": 0.02697098,
      "auxiliary_loss_mlp": 0.01375693,
      "balance_loss_clip": 1.56319976,
      "balance_loss_mlp": 1.10332382,
      "epoch": 0.003847772500450911,
      "flos": 22674608282880.0,
      "grad_norm": 2.45100144899735,
      "language_loss": 0.98840976,
      "learning_rate": 2.510738338534912e-06,
      "loss": 1.02913761,
      "num_input_tokens_seen": 772105,
      "step": 32,
      "time_per_iteration": 2.649449348449707
    },
    {
      "auxiliary_loss_clip": 0.02559674,
      "auxiliary_loss_mlp": 0.01359604,
      "balance_loss_clip": 1.52039504,
      "balance_loss_mlp": 1.09200263,
      "epoch": 0.003968015391090002,
      "flos": 17967796882560.0,
      "grad_norm": 2.721918852117027,
      "language_loss": 1.02625132,
      "learning_rate": 2.5330307420306648e-06,
      "loss": 1.06544399,
      "num_input_tokens_seen": 788955,
      "step": 33,
      "time_per_iteration": 2.6268293857574463
    },
    {
      "auxiliary_loss_clip": 0.02520656,
      "auxiliary_loss_mlp": 0.01344013,
      "balance_loss_clip": 1.51562619,
      "balance_loss_mlp": 1.10006297,
      "epoch": 0.004088258281729093,
      "flos": 27304103658240.0,
      "grad_norm": 2.643964959570546,
      "language_loss": 0.88083756,
      "learning_rate": 2.554657600279796e-06,
      "loss": 0.91948426,
      "num_input_tokens_seen": 810230,
      "step": 34,
      "time_per_iteration": 2.7650465965270996
    },
    {
      "auxiliary_loss_clip": 0.02501756,
      "auxiliary_loss_mlp": 0.01324716,
      "balance_loss_clip": 1.50747418,
      "balance_loss_mlp": 1.07313645,
      "epoch": 0.004208501172368184,
      "flos": 23258587599360.0,
      "grad_norm": 2.1435482384685067,
      "language_loss": 1.03435397,
      "learning_rate": 2.5756575039679493e-06,
      "loss": 1.07261872,
      "num_input_tokens_seen": 829780,
      "step": 35,
      "time_per_iteration": 2.741372585296631
    },
    {
      "auxiliary_loss_clip": 0.02464117,
      "auxiliary_loss_mlp": 0.01351179,
      "balance_loss_clip": 1.49650097,
      "balance_loss_mlp": 1.09998131,
      "epoch": 0.0043287440630072746,
      "flos": 17312104062720.0,
      "grad_norm": 1.9700471122117738,
      "language_loss": 0.94982445,
      "learning_rate": 2.5960657816942747e-06,
      "loss": 0.98797739,
      "num_input_tokens_seen": 848695,
      "step": 36,
      "time_per_iteration": 2.711890935897827
    },
    {
      "auxiliary_loss_clip": 0.02303775,
      "auxiliary_loss_mlp": 0.01393126,
      "balance_loss_clip": 1.57803583,
      "balance_loss_mlp": 1.25198257,
      "epoch": 0.004448986953646365,
      "flos": 53092491160320.0,
      "grad_norm": 1.398194431895382,
      "language_loss": 0.60974598,
      "learning_rate": 2.6159148575788668e-06,
      "loss": 0.64671493,
      "num_input_tokens_seen": 906730,
      "step": 37,
      "time_per_iteration": 3.112426996231079
    },
    {
      "auxiliary_loss_clip": 0.02412854,
      "auxiliary_loss_mlp": 0.013603,
      "balance_loss_clip": 1.48573279,
      "balance_loss_mlp": 1.1108191,
      "epoch": 0.004569229844285457,
      "flos": 13444165866240.0,
      "grad_norm": 2.5257416776216592,
      "language_loss": 0.98745996,
      "learning_rate": 2.635234561171e-06,
      "loss": 1.02519155,
      "num_input_tokens_seen": 925125,
      "step": 38,
      "time_per_iteration": 2.670102596282959
    },
    {
      "auxiliary_loss_clip": 0.02388864,
      "auxiliary_loss_mlp": 0.01327486,
      "balance_loss_clip": 1.47793186,
      "balance_loss_mlp": 1.09345436,
      "epoch": 0.0046894727349245475,
      "flos": 16209609966720.0,
      "grad_norm": 2.2874117518999193,
      "language_loss": 0.94215786,
      "learning_rate": 2.6540523970949877e-06,
      "loss": 0.97932136,
      "num_input_tokens_seen": 939970,
      "step": 39,
      "time_per_iteration": 2.7218737602233887
    },
    {
      "auxiliary_loss_clip": 0.02357519,
      "auxiliary_loss_mlp": 0.01333414,
      "balance_loss_clip": 1.47663307,
      "balance_loss_mlp": 1.09728432,
      "epoch": 0.004809715625563638,
      "flos": 23914244505600.0,
      "grad_norm": 2.6644391722813188,
      "language_loss": 0.92516059,
      "learning_rate": 2.6723937805519533e-06,
      "loss": 0.96206993,
      "num_input_tokens_seen": 957470,
      "step": 40,
      "time_per_iteration": 2.7147934436798096
    },
    {
      "auxiliary_loss_clip": 0.02345942,
      "auxiliary_loss_mlp": 0.01304425,
      "balance_loss_clip": 1.46480012,
      "balance_loss_mlp": 1.07916677,
      "epoch": 0.00492995851620273,
      "flos": 20773030273920.0,
      "grad_norm": 2.171423286546705,
      "language_loss": 0.92990512,
      "learning_rate": 2.690282243737839e-06,
      "loss": 0.96640879,
      "num_input_tokens_seen": 976405,
      "step": 41,
      "time_per_iteration": 2.6997878551483154
    },
    {
      "auxiliary_loss_clip": 0.02310364,
      "auxiliary_loss_mlp": 0.01327203,
      "balance_loss_clip": 1.45331836,
      "balance_loss_mlp": 1.09736788,
      "epoch": 0.0050502014068418205,
      "flos": 20338655103360.0,
      "grad_norm": 2.662969012086038,
      "language_loss": 0.99230605,
      "learning_rate": 2.7077396173840807e-06,
      "loss": 1.02868176,
      "num_input_tokens_seen": 994690,
      "step": 42,
      "time_per_iteration": 3.478790521621704
    },
    {
      "auxiliary_loss_clip": 0.02285866,
      "auxiliary_loss_mlp": 0.01314746,
      "balance_loss_clip": 1.44597411,
      "balance_loss_mlp": 1.09521067,
      "epoch": 0.005170444297480911,
      "flos": 25994872834560.0,
      "grad_norm": 2.6198556118564804,
      "language_loss": 0.92591667,
      "learning_rate": 2.7247861909342594e-06,
      "loss": 0.96192282,
      "num_input_tokens_seen": 1015615,
      "step": 43,
      "time_per_iteration": 4.0540406703948975
    },
    {
      "auxiliary_loss_clip": 0.02282108,
      "auxiliary_loss_mlp": 0.01309658,
      "balance_loss_clip": 1.44477665,
      "balance_loss_mlp": 1.09345984,
      "epoch": 0.005290687188120003,
      "flos": 20954055841920.0,
      "grad_norm": 2.286658410201937,
      "language_loss": 0.8293969,
      "learning_rate": 2.7414408543044743e-06,
      "loss": 0.86531454,
      "num_input_tokens_seen": 1031255,
      "step": 44,
      "time_per_iteration": 2.6883347034454346
    },
    {
      "auxiliary_loss_clip": 0.02231397,
      "auxiliary_loss_mlp": 0.01332251,
      "balance_loss_clip": 1.43015862,
      "balance_loss_mlp": 1.11261952,
      "epoch": 0.005410930078759093,
      "flos": 15851401585920.0,
      "grad_norm": 10.068546149385137,
      "language_loss": 0.79058111,
      "learning_rate": 2.7577212237113157e-06,
      "loss": 0.82621753,
      "num_input_tokens_seen": 1048295,
      "step": 45,
      "time_per_iteration": 2.682742118835449
    },
    {
      "auxiliary_loss_clip": 0.02217989,
      "auxiliary_loss_mlp": 0.01308043,
      "balance_loss_clip": 1.42400014,
      "balance_loss_mlp": 1.09499252,
      "epoch": 0.005531172969398184,
      "flos": 21104988791040.0,
      "grad_norm": 2.1281741110699595,
      "language_loss": 1.04237497,
      "learning_rate": 2.7736437536690466e-06,
      "loss": 1.07763529,
      "num_input_tokens_seen": 1067925,
      "step": 46,
      "time_per_iteration": 2.7103476524353027
    },
    {
      "auxiliary_loss_clip": 0.02207815,
      "auxiliary_loss_mlp": 0.01276385,
      "balance_loss_clip": 1.42449737,
      "balance_loss_mlp": 1.06781626,
      "epoch": 0.005651415860037276,
      "flos": 20844887431680.0,
      "grad_norm": 2.3266364602082144,
      "language_loss": 1.07720137,
      "learning_rate": 2.789223836941131e-06,
      "loss": 1.1120435,
      "num_input_tokens_seen": 1088060,
      "step": 47,
      "time_per_iteration": 2.6737117767333984
    },
    {
      "auxiliary_loss_clip": 0.02172153,
      "auxiliary_loss_mlp": 0.01285319,
      "balance_loss_clip": 1.41179073,
      "balance_loss_mlp": 1.08085096,
      "epoch": 0.005771658750676366,
      "flos": 13260195383040.0,
      "grad_norm": 2.388761675367142,
      "language_loss": 1.08692598,
      "learning_rate": 2.8044758939680847e-06,
      "loss": 1.12150085,
      "num_input_tokens_seen": 1104130,
      "step": 48,
      "time_per_iteration": 2.612104892730713
    },
    {
      "auxiliary_loss_clip": 0.02147161,
      "auxiliary_loss_mlp": 0.01282264,
      "balance_loss_clip": 1.41145968,
      "balance_loss_mlp": 1.07655692,
      "epoch": 0.005891901641315457,
      "flos": 24425396997120.0,
      "grad_norm": 7.205833187648408,
      "language_loss": 1.02130783,
      "learning_rate": 2.8194134530738863e-06,
      "loss": 1.05560207,
      "num_input_tokens_seen": 1122900,
      "step": 49,
      "time_per_iteration": 2.681382894515991
    },
    {
      "auxiliary_loss_clip": 0.02139021,
      "auxiliary_loss_mlp": 0.01290139,
      "balance_loss_clip": 1.40579212,
      "balance_loss_mlp": 1.09387314,
      "epoch": 0.006012144531954548,
      "flos": 23076197314560.0,
      "grad_norm": 2.558727094572549,
      "language_loss": 0.90087247,
      "learning_rate": 2.834049222568994e-06,
      "loss": 0.93516409,
      "num_input_tokens_seen": 1140250,
      "step": 50,
      "time_per_iteration": 2.67350697517395
    },
    {
      "auxiliary_loss_clip": 0.02135085,
      "auxiliary_loss_mlp": 0.01255022,
      "balance_loss_clip": 1.40258718,
      "balance_loss_mlp": 1.06400096,
      "epoch": 0.006132387422593639,
      "flos": 22528775064960.0,
      "grad_norm": 1.8717953273445211,
      "language_loss": 0.92423445,
      "learning_rate": 2.848395155712969e-06,
      "loss": 0.95813555,
      "num_input_tokens_seen": 1160470,
      "step": 51,
      "time_per_iteration": 2.6298584938049316
    },
    {
      "auxiliary_loss_clip": 0.02118313,
      "auxiliary_loss_mlp": 0.01293401,
      "balance_loss_clip": 1.40357351,
      "balance_loss_mlp": 1.09875619,
      "epoch": 0.00625263031323273,
      "flos": 27628340751360.0,
      "grad_norm": 2.312437719329922,
      "language_loss": 0.97569245,
      "learning_rate": 2.8624625093687977e-06,
      "loss": 1.00980961,
      "num_input_tokens_seen": 1177605,
      "step": 52,
      "time_per_iteration": 2.705409288406372
    },
    {
      "auxiliary_loss_clip": 0.02100515,
      "auxiliary_loss_mlp": 0.0125987,
      "balance_loss_clip": 1.39357424,
      "balance_loss_mlp": 1.07705033,
      "epoch": 0.006372873203871821,
      "flos": 23110671392640.0,
      "grad_norm": 2.413575028050906,
      "language_loss": 0.88998222,
      "learning_rate": 2.876261897070029e-06,
      "loss": 0.92358613,
      "num_input_tokens_seen": 1197735,
      "step": 53,
      "time_per_iteration": 2.630678415298462
    },
    {
      "auxiliary_loss_clip": 0.02099512,
      "auxiliary_loss_mlp": 0.01277437,
      "balance_loss_clip": 1.39772594,
      "balance_loss_mlp": 1.09194767,
      "epoch": 0.006493116094510912,
      "flos": 22856028900480.0,
      "grad_norm": 2.470427108327774,
      "language_loss": 0.92309314,
      "learning_rate": 2.889803337127447e-06,
      "loss": 0.95686257,
      "num_input_tokens_seen": 1216335,
      "step": 54,
      "time_per_iteration": 2.6569180488586426
    },
    {
      "auxiliary_loss_clip": 0.02068233,
      "auxiliary_loss_mlp": 0.01296732,
      "balance_loss_clip": 1.38537741,
      "balance_loss_mlp": 1.10141969,
      "epoch": 0.006613358985150003,
      "flos": 23071708114560.0,
      "grad_norm": 4.798285227810707,
      "language_loss": 0.84695476,
      "learning_rate": 2.903096296321516e-06,
      "loss": 0.88060445,
      "num_input_tokens_seen": 1234480,
      "step": 55,
      "time_per_iteration": 2.6526994705200195
    },
    {
      "auxiliary_loss_clip": 0.02069055,
      "auxiliary_loss_mlp": 0.01250531,
      "balance_loss_clip": 1.38686192,
      "balance_loss_mlp": 1.07524538,
      "epoch": 0.006733601875789094,
      "flos": 26537662229760.0,
      "grad_norm": 1.9678578099806474,
      "language_loss": 0.91687977,
      "learning_rate": 2.9161497296578907e-06,
      "loss": 0.95007569,
      "num_input_tokens_seen": 1253870,
      "step": 56,
      "time_per_iteration": 2.70420241355896
    },
    {
      "auxiliary_loss_clip": 0.02049681,
      "auxiliary_loss_mlp": 0.01253205,
      "balance_loss_clip": 1.38082504,
      "balance_loss_mlp": 1.07563078,
      "epoch": 0.006853844766428185,
      "flos": 15523178083200.0,
      "grad_norm": 2.169600513230549,
      "language_loss": 0.85848916,
      "learning_rate": 2.928972116604173e-06,
      "loss": 0.891518,
      "num_input_tokens_seen": 1270145,
      "step": 57,
      "time_per_iteration": 2.5955557823181152
    },
    {
      "auxiliary_loss_clip": 0.02021701,
      "auxiliary_loss_mlp": 0.01234647,
      "balance_loss_clip": 1.37206626,
      "balance_loss_mlp": 1.0678494,
      "epoch": 0.006974087657067276,
      "flos": 24243760897920.0,
      "grad_norm": 5.973243230344974,
      "language_loss": 1.02024364,
      "learning_rate": 2.9415714941751377e-06,
      "loss": 1.05280709,
      "num_input_tokens_seen": 1291365,
      "step": 58,
      "time_per_iteration": 2.651109218597412
    },
    {
      "auxiliary_loss_clip": 0.02038122,
      "auxiliary_loss_mlp": 0.01256678,
      "balance_loss_clip": 1.37371969,
      "balance_loss_mlp": 1.08892632,
      "epoch": 0.007094330547706367,
      "flos": 25772513690880.0,
      "grad_norm": 1.8730097132514536,
      "language_loss": 0.93521208,
      "learning_rate": 2.9539554871897396e-06,
      "loss": 0.96816009,
      "num_input_tokens_seen": 1311535,
      "step": 59,
      "time_per_iteration": 2.6681530475616455
    },
    {
      "auxiliary_loss_clip": 0.02006071,
      "auxiliary_loss_mlp": 0.01242539,
      "balance_loss_clip": 1.36593437,
      "balance_loss_mlp": 1.07907915,
      "epoch": 0.007214573438345458,
      "flos": 21319015979520.0,
      "grad_norm": 2.527464112918325,
      "language_loss": 0.97403991,
      "learning_rate": 2.9661313359851253e-06,
      "loss": 1.00652599,
      "num_input_tokens_seen": 1329420,
      "step": 60,
      "time_per_iteration": 2.6241652965545654
    },
    {
      "auxiliary_loss_clip": 0.01984811,
      "auxiliary_loss_mlp": 0.01236678,
      "balance_loss_clip": 1.36365652,
      "balance_loss_mlp": 1.07741404,
      "epoch": 0.007334816328984549,
      "flos": 24937088192640.0,
      "grad_norm": 6.292004761511977,
      "language_loss": 0.94155359,
      "learning_rate": 2.978105921839922e-06,
      "loss": 0.97376847,
      "num_input_tokens_seen": 1349965,
      "step": 61,
      "time_per_iteration": 2.7245492935180664
    },
    {
      "auxiliary_loss_clip": 0.01971468,
      "auxiliary_loss_mlp": 0.01249403,
      "balance_loss_clip": 1.35985589,
      "balance_loss_mlp": 1.08889973,
      "epoch": 0.00745505921962364,
      "flos": 18510586277760.0,
      "grad_norm": 2.265863591350538,
      "language_loss": 0.72096038,
      "learning_rate": 2.9898857903302893e-06,
      "loss": 0.75316906,
      "num_input_tokens_seen": 1368915,
      "step": 62,
      "time_per_iteration": 2.6413767337799072
    },
    {
      "auxiliary_loss_clip": 0.01974162,
      "auxiliary_loss_mlp": 0.01254663,
      "balance_loss_clip": 1.35824609,
      "balance_loss_mlp": 1.08986795,
      "epoch": 0.007575302110262731,
      "flos": 18477656484480.0,
      "grad_norm": 3.620721746446778,
      "language_loss": 0.8797034,
      "learning_rate": 3.001477172817253e-06,
      "loss": 0.9119916,
      "num_input_tokens_seen": 1386805,
      "step": 63,
      "time_per_iteration": 2.5936529636383057
    },
    {
      "auxiliary_loss_clip": 0.01950344,
      "auxiliary_loss_mlp": 0.01226837,
      "balance_loss_clip": 1.35019684,
      "balance_loss_mlp": 1.07834983,
      "epoch": 0.007695545000901822,
      "flos": 24973178382720.0,
      "grad_norm": 2.721068187518123,
      "language_loss": 0.9643023,
      "learning_rate": 3.012886006241894e-06,
      "loss": 0.99607414,
      "num_input_tokens_seen": 1406190,
      "step": 64,
      "time_per_iteration": 2.636819839477539
    },
    {
      "auxiliary_loss_clip": 0.01955315,
      "auxiliary_loss_mlp": 0.01230422,
      "balance_loss_clip": 1.35175109,
      "balance_loss_mlp": 1.07440138,
      "epoch": 0.007815787891540913,
      "flos": 21324223451520.0,
      "grad_norm": 2.0721214351494353,
      "language_loss": 0.88308674,
      "learning_rate": 3.0241179513858383e-06,
      "loss": 0.91494405,
      "num_input_tokens_seen": 1425500,
      "step": 65,
      "time_per_iteration": 2.624528646469116
    },
    {
      "auxiliary_loss_clip": 0.01937021,
      "auxiliary_loss_mlp": 0.01255913,
      "balance_loss_clip": 1.34111071,
      "balance_loss_mlp": 1.09264445,
      "epoch": 0.007936030782180003,
      "flos": 21575777374080.0,
      "grad_norm": 2.369998740161197,
      "language_loss": 0.87739629,
      "learning_rate": 3.035178409737647e-06,
      "loss": 0.9093256,
      "num_input_tokens_seen": 1442950,
      "step": 66,
      "time_per_iteration": 2.6210875511169434
    },
    {
      "auxiliary_loss_clip": 0.01916913,
      "auxiliary_loss_mlp": 0.01219577,
      "balance_loss_clip": 1.33460462,
      "balance_loss_mlp": 1.08339238,
      "epoch": 0.008056273672819095,
      "flos": 20120785159680.0,
      "grad_norm": 2.271438418590118,
      "language_loss": 0.88988554,
      "learning_rate": 3.046072539090907e-06,
      "loss": 0.92125046,
      "num_input_tokens_seen": 1460915,
      "step": 67,
      "time_per_iteration": 2.658783435821533
    },
    {
      "auxiliary_loss_clip": 0.01911898,
      "auxiliary_loss_mlp": 0.01215404,
      "balance_loss_clip": 1.33377588,
      "balance_loss_mlp": 1.07292461,
      "epoch": 0.008176516563458186,
      "flos": 18333116156160.0,
      "grad_norm": 2.5127764173045657,
      "language_loss": 1.04831719,
      "learning_rate": 3.056805267986779e-06,
      "loss": 1.07959032,
      "num_input_tokens_seen": 1478385,
      "step": 68,
      "time_per_iteration": 2.6134274005889893
    },
    {
      "auxiliary_loss_clip": 0.01894127,
      "auxiliary_loss_mlp": 0.01219922,
      "balance_loss_clip": 1.32747531,
      "balance_loss_mlp": 1.08106661,
      "epoch": 0.008296759454097276,
      "flos": 21872076664320.0,
      "grad_norm": 2.294801356368828,
      "language_loss": 0.95265126,
      "learning_rate": 3.0673813091022194e-06,
      "loss": 0.98379177,
      "num_input_tokens_seen": 1497605,
      "step": 69,
      "time_per_iteration": 3.514554500579834
    },
    {
      "auxiliary_loss_clip": 0.01747915,
      "auxiliary_loss_mlp": 0.0119753,
      "balance_loss_clip": 1.3496182,
      "balance_loss_mlp": 1.13153541,
      "epoch": 0.008417002344736368,
      "flos": 63408228036480.0,
      "grad_norm": 1.2700818634017939,
      "language_loss": 0.62012094,
      "learning_rate": 3.0778051716749317e-06,
      "loss": 0.64957529,
      "num_input_tokens_seen": 1561150,
      "step": 70,
      "time_per_iteration": 4.728844165802002
    },
    {
      "auxiliary_loss_clip": 0.01866523,
      "auxiliary_loss_mlp": 0.01209449,
      "balance_loss_clip": 1.30933714,
      "balance_loss_mlp": 1.07231104,
      "epoch": 0.008537245235375458,
      "flos": 22966454286720.0,
      "grad_norm": 2.358332988880353,
      "language_loss": 0.90341187,
      "learning_rate": 3.0880811730470094e-06,
      "loss": 0.93417162,
      "num_input_tokens_seen": 1580605,
      "step": 71,
      "time_per_iteration": 2.62558913230896
    },
    {
      "auxiliary_loss_clip": 0.01714101,
      "auxiliary_loss_mlp": 0.01158795,
      "balance_loss_clip": 1.32948065,
      "balance_loss_mlp": 1.09737861,
      "epoch": 0.008657488126014549,
      "flos": 61984046712960.0,
      "grad_norm": 1.1377262086437336,
      "language_loss": 0.58551824,
      "learning_rate": 3.098213449401257e-06,
      "loss": 0.6142472,
      "num_input_tokens_seen": 1647535,
      "step": 72,
      "time_per_iteration": 3.1147620677948
    },
    {
      "auxiliary_loss_clip": 0.01856985,
      "auxiliary_loss_mlp": 0.01213376,
      "balance_loss_clip": 1.30769348,
      "balance_loss_mlp": 1.08215058,
      "epoch": 0.00877773101665364,
      "flos": 30296791152000.0,
      "grad_norm": 2.159606984850392,
      "language_loss": 0.98988783,
      "learning_rate": 3.1082059657570015e-06,
      "loss": 1.0205915,
      "num_input_tokens_seen": 1666770,
      "step": 73,
      "time_per_iteration": 2.6980302333831787
    },
    {
      "auxiliary_loss_clip": 0.01828996,
      "auxiliary_loss_mlp": 0.01201452,
      "balance_loss_clip": 1.29911363,
      "balance_loss_mlp": 1.06755626,
      "epoch": 0.00889797390729273,
      "flos": 23514056104320.0,
      "grad_norm": 2.902644079794967,
      "language_loss": 0.96767902,
      "learning_rate": 3.1180625252858496e-06,
      "loss": 0.99798346,
      "num_input_tokens_seen": 1685200,
      "step": 74,
      "time_per_iteration": 2.63293194770813
    },
    {
      "auxiliary_loss_clip": 0.01812889,
      "auxiliary_loss_mlp": 0.01209509,
      "balance_loss_clip": 1.28972054,
      "balance_loss_mlp": 1.08362412,
      "epoch": 0.009018216797931822,
      "flos": 23075838178560.0,
      "grad_norm": 2.637849473753041,
      "language_loss": 0.80078471,
      "learning_rate": 3.1277867780021663e-06,
      "loss": 0.83100867,
      "num_input_tokens_seen": 1701835,
      "step": 75,
      "time_per_iteration": 2.6503360271453857
    },
    {
      "auxiliary_loss_clip": 0.01791304,
      "auxiliary_loss_mlp": 0.01180957,
      "balance_loss_clip": 1.28232956,
      "balance_loss_mlp": 1.06537127,
      "epoch": 0.009138459688570914,
      "flos": 15918877284480.0,
      "grad_norm": 2.788882816295574,
      "language_loss": 0.95595336,
      "learning_rate": 3.1373822288779824e-06,
      "loss": 0.98567593,
      "num_input_tokens_seen": 1718415,
      "step": 76,
      "time_per_iteration": 2.6364893913269043
    },
    {
      "auxiliary_loss_clip": 0.01790247,
      "auxiliary_loss_mlp": 0.01211052,
      "balance_loss_clip": 1.28385448,
      "balance_loss_mlp": 1.08821845,
      "epoch": 0.009258702579210003,
      "flos": 27016531372800.0,
      "grad_norm": 3.180093307409489,
      "language_loss": 0.79535246,
      "learning_rate": 3.1468522454274533e-06,
      "loss": 0.82536548,
      "num_input_tokens_seen": 1738770,
      "step": 77,
      "time_per_iteration": 2.7492098808288574
    },
    {
      "auxiliary_loss_clip": 0.01780251,
      "auxiliary_loss_mlp": 0.0119471,
      "balance_loss_clip": 1.27814174,
      "balance_loss_mlp": 1.07387948,
      "epoch": 0.009378945469849095,
      "flos": 26903196984960.0,
      "grad_norm": 1.9469720948869846,
      "language_loss": 0.91859007,
      "learning_rate": 3.15620006480197e-06,
      "loss": 0.9483397,
      "num_input_tokens_seen": 1758040,
      "step": 78,
      "time_per_iteration": 2.665693998336792
    },
    {
      "auxiliary_loss_clip": 0.01776904,
      "auxiliary_loss_mlp": 0.01187715,
      "balance_loss_clip": 1.27530289,
      "balance_loss_mlp": 1.06774259,
      "epoch": 0.009499188360488187,
      "flos": 35694236327040.0,
      "grad_norm": 4.061064857383708,
      "language_loss": 0.74933708,
      "learning_rate": 3.1654288004333087e-06,
      "loss": 0.77898324,
      "num_input_tokens_seen": 1776705,
      "step": 79,
      "time_per_iteration": 2.714592933654785
    },
    {
      "auxiliary_loss_clip": 0.01755734,
      "auxiliary_loss_mlp": 0.01180228,
      "balance_loss_clip": 1.2690351,
      "balance_loss_mlp": 1.07036471,
      "epoch": 0.009619431251127276,
      "flos": 21503201944320.0,
      "grad_norm": 2.4294124591910675,
      "language_loss": 0.76052099,
      "learning_rate": 3.1745414482589353e-06,
      "loss": 0.78988063,
      "num_input_tokens_seen": 1795915,
      "step": 80,
      "time_per_iteration": 2.632139205932617
    },
    {
      "auxiliary_loss_clip": 0.01745899,
      "auxiliary_loss_mlp": 0.01173838,
      "balance_loss_clip": 1.26470017,
      "balance_loss_mlp": 1.06268764,
      "epoch": 0.009739674141766368,
      "flos": 17421056991360.0,
      "grad_norm": 3.186390088460552,
      "language_loss": 0.87151778,
      "learning_rate": 3.1835408925606204e-06,
      "loss": 0.90071511,
      "num_input_tokens_seen": 1814055,
      "step": 81,
      "time_per_iteration": 2.6621835231781006
    },
    {
      "auxiliary_loss_clip": 0.01725373,
      "auxiliary_loss_mlp": 0.01186511,
      "balance_loss_clip": 1.25754762,
      "balance_loss_mlp": 1.07650506,
      "epoch": 0.00985991703240546,
      "flos": 27527109246720.0,
      "grad_norm": 5.285462304250921,
      "language_loss": 0.89346701,
      "learning_rate": 3.1924299114448214e-06,
      "loss": 0.92258584,
      "num_input_tokens_seen": 1834535,
      "step": 82,
      "time_per_iteration": 2.671375274658203
    },
    {
      "auxiliary_loss_clip": 0.01735232,
      "auxiliary_loss_mlp": 0.01187334,
      "balance_loss_clip": 1.2621932,
      "balance_loss_mlp": 1.07756591,
      "epoch": 0.00998015992304455,
      "flos": 13808084509440.0,
      "grad_norm": 2.534205593001497,
      "language_loss": 0.83327866,
      "learning_rate": 3.2012111819909055e-06,
      "loss": 0.86250436,
      "num_input_tokens_seen": 1851865,
      "step": 83,
      "time_per_iteration": 2.600539445877075
    },
    {
      "auxiliary_loss_clip": 0.01723025,
      "auxiliary_loss_mlp": 0.01179222,
      "balance_loss_clip": 1.25518084,
      "balance_loss_mlp": 1.07226741,
      "epoch": 0.010100402813683641,
      "flos": 20191385341440.0,
      "grad_norm": 2.138824840512966,
      "language_loss": 0.95058811,
      "learning_rate": 3.2098872850910627e-06,
      "loss": 0.97961056,
      "num_input_tokens_seen": 1868540,
      "step": 84,
      "time_per_iteration": 2.599152088165283
    },
    {
      "auxiliary_loss_clip": 0.01720412,
      "auxiliary_loss_mlp": 0.01179937,
      "balance_loss_clip": 1.25694168,
      "balance_loss_mlp": 1.07703519,
      "epoch": 0.010220645704322733,
      "flos": 17201642762880.0,
      "grad_norm": 2.1010529484399623,
      "language_loss": 0.89264512,
      "learning_rate": 3.2184607100038194e-06,
      "loss": 0.92164862,
      "num_input_tokens_seen": 1887180,
      "step": 85,
      "time_per_iteration": 2.614490270614624
    },
    {
      "auxiliary_loss_clip": 0.01718053,
      "auxiliary_loss_mlp": 0.01182105,
      "balance_loss_clip": 1.25722146,
      "balance_loss_mlp": 1.08063459,
      "epoch": 0.010340888594961822,
      "flos": 21470415805440.0,
      "grad_norm": 5.072650567508388,
      "language_loss": 0.93178141,
      "learning_rate": 3.2269338586412414e-06,
      "loss": 0.960783,
      "num_input_tokens_seen": 1904765,
      "step": 86,
      "time_per_iteration": 2.6865406036376953
    },
    {
      "auxiliary_loss_clip": 0.01704298,
      "auxiliary_loss_mlp": 0.01173039,
      "balance_loss_clip": 1.24989474,
      "balance_loss_mlp": 1.0777669,
      "epoch": 0.010461131485600914,
      "flos": 23002831785600.0,
      "grad_norm": 2.5050926268469746,
      "language_loss": 0.96704745,
      "learning_rate": 3.2353090496083106e-06,
      "loss": 0.99582082,
      "num_input_tokens_seen": 1922600,
      "step": 87,
      "time_per_iteration": 2.676307439804077
    },
    {
      "auxiliary_loss_clip": 0.01681535,
      "auxiliary_loss_mlp": 0.01172089,
      "balance_loss_clip": 1.24031413,
      "balance_loss_mlp": 1.082968,
      "epoch": 0.010581374376240005,
      "flos": 33546850571520.0,
      "grad_norm": 1.9442446928374768,
      "language_loss": 0.81370449,
      "learning_rate": 3.2435885220114572e-06,
      "loss": 0.84224069,
      "num_input_tokens_seen": 1943950,
      "step": 88,
      "time_per_iteration": 2.7299387454986572
    },
    {
      "auxiliary_loss_clip": 0.0169003,
      "auxiliary_loss_mlp": 0.01156504,
      "balance_loss_clip": 1.24699545,
      "balance_loss_mlp": 1.06290126,
      "epoch": 0.010701617266879095,
      "flos": 21763087822080.0,
      "grad_norm": 3.6004270120009805,
      "language_loss": 0.93844104,
      "learning_rate": 3.2517744390519113e-06,
      "loss": 0.96690637,
      "num_input_tokens_seen": 1962815,
      "step": 89,
      "time_per_iteration": 2.6142044067382812
    },
    {
      "auxiliary_loss_clip": 0.01676169,
      "auxiliary_loss_mlp": 0.01156133,
      "balance_loss_clip": 1.23301625,
      "balance_loss_mlp": 1.06782269,
      "epoch": 0.010821860157518187,
      "flos": 19060199256960.0,
      "grad_norm": 2.431057334734209,
      "language_loss": 0.75148028,
      "learning_rate": 3.259868891418298e-06,
      "loss": 0.77980328,
      "num_input_tokens_seen": 1980580,
      "step": 90,
      "time_per_iteration": 2.592013359069824
    },
    {
      "auxiliary_loss_clip": 0.01684397,
      "auxiliary_loss_mlp": 0.01192342,
      "balance_loss_clip": 1.24279797,
      "balance_loss_mlp": 1.10045588,
      "epoch": 0.010942103048157278,
      "flos": 25447378757760.0,
      "grad_norm": 2.02081501395768,
      "language_loss": 0.8498618,
      "learning_rate": 3.2678739004917757e-06,
      "loss": 0.87862921,
      "num_input_tokens_seen": 2000315,
      "step": 91,
      "time_per_iteration": 2.637687921524048
    },
    {
      "auxiliary_loss_clip": 0.0166731,
      "auxiliary_loss_mlp": 0.01170394,
      "balance_loss_clip": 1.23643279,
      "balance_loss_mlp": 1.08461094,
      "epoch": 0.011062345938796368,
      "flos": 27493928058240.0,
      "grad_norm": 3.0767886788724335,
      "language_loss": 0.92124963,
      "learning_rate": 3.275791421376029e-06,
      "loss": 0.94962668,
      "num_input_tokens_seen": 2023760,
      "step": 92,
      "time_per_iteration": 2.6677346229553223
    },
    {
      "auxiliary_loss_clip": 0.01656367,
      "auxiliary_loss_mlp": 0.01147862,
      "balance_loss_clip": 1.2288785,
      "balance_loss_mlp": 1.0688982,
      "epoch": 0.01118258882943546,
      "flos": 16071210864000.0,
      "grad_norm": 2.4559507604207944,
      "language_loss": 0.96172017,
      "learning_rate": 3.2836233457634622e-06,
      "loss": 0.98976243,
      "num_input_tokens_seen": 2041895,
      "step": 93,
      "time_per_iteration": 2.5751163959503174
    },
    {
      "auxiliary_loss_clip": 0.01653702,
      "auxiliary_loss_mlp": 0.01183245,
      "balance_loss_clip": 1.22804332,
      "balance_loss_mlp": 1.08911729,
      "epoch": 0.011302831720074551,
      "flos": 20668602458880.0,
      "grad_norm": 3.2201338679645133,
      "language_loss": 0.85535169,
      "learning_rate": 3.2913715046481135e-06,
      "loss": 0.88372111,
      "num_input_tokens_seen": 2061640,
      "step": 94,
      "time_per_iteration": 2.6139183044433594
    },
    {
      "auxiliary_loss_clip": 0.01650393,
      "auxiliary_loss_mlp": 0.0116086,
      "balance_loss_clip": 1.22666478,
      "balance_loss_mlp": 1.08041704,
      "epoch": 0.011423074610713641,
      "flos": 13072238490240.0,
      "grad_norm": 6.133333790693309,
      "language_loss": 0.88872576,
      "learning_rate": 3.299037670895023e-06,
      "loss": 0.91683829,
      "num_input_tokens_seen": 2078255,
      "step": 95,
      "time_per_iteration": 2.619093418121338
    },
    {
      "auxiliary_loss_clip": 0.01652527,
      "auxiliary_loss_mlp": 0.01147409,
      "balance_loss_clip": 1.23226905,
      "balance_loss_mlp": 1.06486893,
      "epoch": 0.011543317501352733,
      "flos": 30335646689280.0,
      "grad_norm": 3.7049022681075114,
      "language_loss": 0.80362546,
      "learning_rate": 3.3066235616750667e-06,
      "loss": 0.83162487,
      "num_input_tokens_seen": 2099490,
      "step": 96,
      "time_per_iteration": 2.6571922302246094
    },
    {
      "auxiliary_loss_clip": 0.01631608,
      "auxiliary_loss_mlp": 0.01143059,
      "balance_loss_clip": 1.21968937,
      "balance_loss_mlp": 1.06500113,
      "epoch": 0.011663560391991824,
      "flos": 15522962601600.0,
      "grad_norm": 4.102507524041367,
      "language_loss": 0.9242425,
      "learning_rate": 3.3141308407736276e-06,
      "loss": 0.95198917,
      "num_input_tokens_seen": 2116125,
      "step": 97,
      "time_per_iteration": 4.277625560760498
    },
    {
      "auxiliary_loss_clip": 0.01637193,
      "auxiliary_loss_mlp": 0.01148706,
      "balance_loss_clip": 1.21698785,
      "balance_loss_mlp": 1.07117188,
      "epoch": 0.011783803282630914,
      "flos": 19902125116800.0,
      "grad_norm": 2.409503682699795,
      "language_loss": 0.86788988,
      "learning_rate": 3.321561120780869e-06,
      "loss": 0.89574891,
      "num_input_tokens_seen": 2134835,
      "step": 98,
      "time_per_iteration": 3.4901506900787354
    },
    {
      "auxiliary_loss_clip": 0.01626835,
      "auxiliary_loss_mlp": 0.01143323,
      "balance_loss_clip": 1.21967435,
      "balance_loss_mlp": 1.07394278,
      "epoch": 0.011904046173270006,
      "flos": 22340674517760.0,
      "grad_norm": 12.859579686898757,
      "language_loss": 1.0142858,
      "learning_rate": 3.3289159651708192e-06,
      "loss": 1.04198742,
      "num_input_tokens_seen": 2152410,
      "step": 99,
      "time_per_iteration": 2.60176420211792
    },
    {
      "auxiliary_loss_clip": 0.01625277,
      "auxiliary_loss_mlp": 0.01142614,
      "balance_loss_clip": 1.21657133,
      "balance_loss_mlp": 1.06670177,
      "epoch": 0.012024289063909096,
      "flos": 19100060375040.0,
      "grad_norm": 1.9609932207630814,
      "language_loss": 0.97639257,
      "learning_rate": 3.3361968902759768e-06,
      "loss": 1.00407147,
      "num_input_tokens_seen": 2172090,
      "step": 100,
      "time_per_iteration": 2.5887444019317627
    },
    {
      "auxiliary_loss_clip": 0.01619465,
      "auxiliary_loss_mlp": 0.01132177,
      "balance_loss_clip": 1.21476483,
      "balance_loss_mlp": 1.06503844,
      "epoch": 0.012144531954548187,
      "flos": 15012205159680.0,
      "grad_norm": 2.274671944173216,
      "language_loss": 0.93972164,
      "learning_rate": 3.343405367163663e-06,
      "loss": 0.96723807,
      "num_input_tokens_seen": 2189020,
      "step": 101,
      "time_per_iteration": 2.5869266986846924
    },
    {
      "auxiliary_loss_clip": 0.01623648,
      "auxiliary_loss_mlp": 0.01137326,
      "balance_loss_clip": 1.21493495,
      "balance_loss_mlp": 1.06751716,
      "epoch": 0.012264774845187279,
      "flos": 15122020014720.0,
      "grad_norm": 13.200211338057656,
      "language_loss": 0.81224948,
      "learning_rate": 3.350542823419951e-06,
      "loss": 0.83985919,
      "num_input_tokens_seen": 2205620,
      "step": 102,
      "time_per_iteration": 2.601289987564087
    },
    {
      "auxiliary_loss_clip": 0.01617352,
      "auxiliary_loss_mlp": 0.01152063,
      "balance_loss_clip": 1.20911241,
      "balance_loss_mlp": 1.08277845,
      "epoch": 0.012385017735826368,
      "flos": 13949248959360.0,
      "grad_norm": 3.5181414275455443,
      "language_loss": 0.87406003,
      "learning_rate": 3.3576106448465615e-06,
      "loss": 0.90175414,
      "num_input_tokens_seen": 2219000,
      "step": 103,
      "time_per_iteration": 2.575157880783081
    },
    {
      "auxiliary_loss_clip": 0.01606251,
      "auxiliary_loss_mlp": 0.01139622,
      "balance_loss_clip": 1.20703244,
      "balance_loss_mlp": 1.06952691,
      "epoch": 0.01250526062646546,
      "flos": 23623260428160.0,
      "grad_norm": 2.010437276769021,
      "language_loss": 0.88113737,
      "learning_rate": 3.3646101770757797e-06,
      "loss": 0.9085961,
      "num_input_tokens_seen": 2237790,
      "step": 104,
      "time_per_iteration": 2.609377861022949
    },
    {
      "auxiliary_loss_clip": 0.01599122,
      "auxiliary_loss_mlp": 0.0114088,
      "balance_loss_clip": 1.20334709,
      "balance_loss_mlp": 1.06735206,
      "epoch": 0.012625503517104552,
      "flos": 34640078958720.0,
      "grad_norm": 1.7085932268712503,
      "language_loss": 0.85761094,
      "learning_rate": 3.371542727108104e-06,
      "loss": 0.88501102,
      "num_input_tokens_seen": 2259965,
      "step": 105,
      "time_per_iteration": 2.728022813796997
    },
    {
      "auxiliary_loss_clip": 0.01602075,
      "auxiliary_loss_mlp": 0.01180103,
      "balance_loss_clip": 1.20619822,
      "balance_loss_mlp": 1.1089586,
      "epoch": 0.012745746407743641,
      "flos": 17821891837440.0,
      "grad_norm": 2.6058663153912716,
      "language_loss": 0.90225899,
      "learning_rate": 3.3784095647770114e-06,
      "loss": 0.93008077,
      "num_input_tokens_seen": 2278610,
      "step": 106,
      "time_per_iteration": 2.6113648414611816
    },
    {
      "auxiliary_loss_clip": 0.01592173,
      "auxiliary_loss_mlp": 0.01142748,
      "balance_loss_clip": 1.19625306,
      "balance_loss_mlp": 1.07260489,
      "epoch": 0.012865989298382733,
      "flos": 20595057361920.0,
      "grad_norm": 2.5375172503219687,
      "language_loss": 0.88876402,
      "learning_rate": 3.3852119241449547e-06,
      "loss": 0.91611332,
      "num_input_tokens_seen": 2297730,
      "step": 107,
      "time_per_iteration": 2.6060562133789062
    },
    {
      "auxiliary_loss_clip": 0.01587286,
      "auxiliary_loss_mlp": 0.01131537,
      "balance_loss_clip": 1.19514942,
      "balance_loss_mlp": 1.06435037,
      "epoch": 0.012986232189021825,
      "flos": 23948969978880.0,
      "grad_norm": 2.1263203971495717,
      "language_loss": 0.96458864,
      "learning_rate": 3.3919510048344295e-06,
      "loss": 0.99177688,
      "num_input_tokens_seen": 2315740,
      "step": 108,
      "time_per_iteration": 2.6117992401123047
    },
    {
      "auxiliary_loss_clip": 0.01576402,
      "auxiliary_loss_mlp": 0.01131786,
      "balance_loss_clip": 1.19047308,
      "balance_loss_mlp": 1.06994045,
      "epoch": 0.013106475079660914,
      "flos": 23725425686400.0,
      "grad_norm": 4.195907912856149,
      "language_loss": 0.86752558,
      "learning_rate": 3.3986279732976907e-06,
      "loss": 0.89460742,
      "num_input_tokens_seen": 2334215,
      "step": 109,
      "time_per_iteration": 2.633835792541504
    },
    {
      "auxiliary_loss_clip": 0.01570618,
      "auxiliary_loss_mlp": 0.01112179,
      "balance_loss_clip": 1.18739426,
      "balance_loss_mlp": 1.051144,
      "epoch": 0.013226717970300006,
      "flos": 21102438925440.0,
      "grad_norm": 2.031351529118376,
      "language_loss": 0.9572528,
      "learning_rate": 3.4052439640284983e-06,
      "loss": 0.98408073,
      "num_input_tokens_seen": 2353130,
      "step": 110,
      "time_per_iteration": 2.610891580581665
    },
    {
      "auxiliary_loss_clip": 0.01571483,
      "auxiliary_loss_mlp": 0.01129761,
      "balance_loss_clip": 1.1909548,
      "balance_loss_mlp": 1.06691408,
      "epoch": 0.013346960860939098,
      "flos": 24863902231680.0,
      "grad_norm": 1.7554086602588028,
      "language_loss": 0.8135798,
      "learning_rate": 3.4118000807190217e-06,
      "loss": 0.84059227,
      "num_input_tokens_seen": 2374010,
      "step": 111,
      "time_per_iteration": 2.696084499359131
    },
    {
      "auxiliary_loss_clip": 0.01574202,
      "auxiliary_loss_mlp": 0.01128873,
      "balance_loss_clip": 1.18895674,
      "balance_loss_mlp": 1.06802821,
      "epoch": 0.013467203751578187,
      "flos": 28181940140160.0,
      "grad_norm": 1.8017197257528965,
      "language_loss": 0.76225603,
      "learning_rate": 3.4182973973648723e-06,
      "loss": 0.78928673,
      "num_input_tokens_seen": 2395220,
      "step": 112,
      "time_per_iteration": 2.6860828399658203
    },
    {
      "auxiliary_loss_clip": 0.01561408,
      "auxiliary_loss_mlp": 0.01148678,
      "balance_loss_clip": 1.18500376,
      "balance_loss_mlp": 1.08821487,
      "epoch": 0.013587446642217279,
      "flos": 18916233546240.0,
      "grad_norm": 2.3543122290576055,
      "language_loss": 0.95194882,
      "learning_rate": 3.424736959321014e-06,
      "loss": 0.97904968,
      "num_input_tokens_seen": 2413025,
      "step": 113,
      "time_per_iteration": 2.575143575668335
    },
    {
      "auxiliary_loss_clip": 0.01564182,
      "auxiliary_loss_mlp": 0.01142649,
      "balance_loss_clip": 1.18463874,
      "balance_loss_mlp": 1.08058834,
      "epoch": 0.01370768953285637,
      "flos": 23988615615360.0,
      "grad_norm": 2.4744194211300097,
      "language_loss": 0.889431,
      "learning_rate": 3.431119784311155e-06,
      "loss": 0.91649926,
      "num_input_tokens_seen": 2432700,
      "step": 114,
      "time_per_iteration": 2.608177661895752
    },
    {
      "auxiliary_loss_clip": 0.01549295,
      "auxiliary_loss_mlp": 0.01130157,
      "balance_loss_clip": 1.17954278,
      "balance_loss_mlp": 1.07250714,
      "epoch": 0.01382793242349546,
      "flos": 39202565512320.0,
      "grad_norm": 1.638271705124018,
      "language_loss": 0.77672195,
      "learning_rate": 3.43744686339307e-06,
      "loss": 0.80351645,
      "num_input_tokens_seen": 2455020,
      "step": 115,
      "time_per_iteration": 2.7269628047943115
    },
    {
      "auxiliary_loss_clip": 0.01544967,
      "auxiliary_loss_mlp": 0.010953,
      "balance_loss_clip": 1.17391562,
      "balance_loss_mlp": 1.0419656,
      "epoch": 0.013948175314134552,
      "flos": 41353506714240.0,
      "grad_norm": 2.368189704300822,
      "language_loss": 0.9097631,
      "learning_rate": 3.44371916188212e-06,
      "loss": 0.93616581,
      "num_input_tokens_seen": 2475775,
      "step": 116,
      "time_per_iteration": 2.746272563934326
    },
    {
      "auxiliary_loss_clip": 0.01539977,
      "auxiliary_loss_mlp": 0.01110158,
      "balance_loss_clip": 1.17387915,
      "balance_loss_mlp": 1.05901718,
      "epoch": 0.014068418204773643,
      "flos": 22453542028800.0,
      "grad_norm": 2.4300197062005022,
      "language_loss": 0.86365074,
      "learning_rate": 3.449937620235143e-06,
      "loss": 0.8901521,
      "num_input_tokens_seen": 2496370,
      "step": 117,
      "time_per_iteration": 2.637294292449951
    },
    {
      "auxiliary_loss_clip": 0.01541673,
      "auxiliary_loss_mlp": 0.01114464,
      "balance_loss_clip": 1.1748333,
      "balance_loss_mlp": 1.060987,
      "epoch": 0.014188661095412733,
      "flos": 23805147922560.0,
      "grad_norm": 2.1574447369944614,
      "language_loss": 0.89566547,
      "learning_rate": 3.456103154896722e-06,
      "loss": 0.92222685,
      "num_input_tokens_seen": 2517645,
      "step": 118,
      "time_per_iteration": 2.635037660598755
    },
    {
      "auxiliary_loss_clip": 0.01528601,
      "auxiliary_loss_mlp": 0.0112393,
      "balance_loss_clip": 1.16721666,
      "balance_loss_mlp": 1.0726459,
      "epoch": 0.014308903986051825,
      "flos": 23660248458240.0,
      "grad_norm": 1.87527939650359,
      "language_loss": 0.92715997,
      "learning_rate": 3.462216659109757e-06,
      "loss": 0.95368528,
      "num_input_tokens_seen": 2537825,
      "step": 119,
      "time_per_iteration": 2.5826480388641357
    },
    {
      "auxiliary_loss_clip": 0.01548501,
      "auxiliary_loss_mlp": 0.011336,
      "balance_loss_clip": 1.17659259,
      "balance_loss_mlp": 1.08133817,
      "epoch": 0.014429146876690916,
      "flos": 20667991927680.0,
      "grad_norm": 2.574267891817223,
      "language_loss": 0.85374397,
      "learning_rate": 3.4682790036921077e-06,
      "loss": 0.88056493,
      "num_input_tokens_seen": 2556485,
      "step": 120,
      "time_per_iteration": 2.568709135055542
    },
    {
      "auxiliary_loss_clip": 0.01522954,
      "auxiliary_loss_mlp": 0.01109574,
      "balance_loss_clip": 1.1682452,
      "balance_loss_mlp": 1.06606269,
      "epoch": 0.014549389767330006,
      "flos": 20229199384320.0,
      "grad_norm": 1.861321878702916,
      "language_loss": 0.83202308,
      "learning_rate": 3.4742910377810193e-06,
      "loss": 0.85834837,
      "num_input_tokens_seen": 2573945,
      "step": 121,
      "time_per_iteration": 2.618236541748047
    },
    {
      "auxiliary_loss_clip": 0.01520957,
      "auxiliary_loss_mlp": 0.01119775,
      "balance_loss_clip": 1.16583323,
      "balance_loss_mlp": 1.07340288,
      "epoch": 0.014669632657969098,
      "flos": 18004174381440.0,
      "grad_norm": 2.4903333954254143,
      "language_loss": 0.88693422,
      "learning_rate": 3.4802535895469042e-06,
      "loss": 0.91334158,
      "num_input_tokens_seen": 2592695,
      "step": 122,
      "time_per_iteration": 2.7141149044036865
    },
    {
      "auxiliary_loss_clip": 0.01523539,
      "auxiliary_loss_mlp": 0.0111212,
      "balance_loss_clip": 1.1650703,
      "balance_loss_mlp": 1.06543732,
      "epoch": 0.01478987554860819,
      "flos": 22741796672640.0,
      "grad_norm": 2.0415616141647654,
      "language_loss": 0.89808935,
      "learning_rate": 3.4861674668779934e-06,
      "loss": 0.92444593,
      "num_input_tokens_seen": 2610925,
      "step": 123,
      "time_per_iteration": 2.693521022796631
    },
    {
      "auxiliary_loss_clip": 0.01514595,
      "auxiliary_loss_mlp": 0.01106452,
      "balance_loss_clip": 1.16026092,
      "balance_loss_mlp": 1.05831575,
      "epoch": 0.01491011843924728,
      "flos": 17198590106880.0,
      "grad_norm": 5.187727928313824,
      "language_loss": 0.84371674,
      "learning_rate": 3.492033458037272e-06,
      "loss": 0.86992723,
      "num_input_tokens_seen": 2629495,
      "step": 124,
      "time_per_iteration": 3.479440450668335
    },
    {
      "auxiliary_loss_clip": 0.01510728,
      "auxiliary_loss_mlp": 0.01110825,
      "balance_loss_clip": 1.15768313,
      "balance_loss_mlp": 1.06760001,
      "epoch": 0.01503036132988637,
      "flos": 17673867889920.0,
      "grad_norm": 2.664610213426129,
      "language_loss": 0.87365061,
      "learning_rate": 3.497852332293018e-06,
      "loss": 0.8998661,
      "num_input_tokens_seen": 2645070,
      "step": 125,
      "time_per_iteration": 3.4887735843658447
    },
    {
      "auxiliary_loss_clip": 0.01510354,
      "auxiliary_loss_mlp": 0.0111473,
      "balance_loss_clip": 1.15976214,
      "balance_loss_mlp": 1.07255375,
      "epoch": 0.015150604220525462,
      "flos": 18878239935360.0,
      "grad_norm": 2.541710048970901,
      "language_loss": 0.96627098,
      "learning_rate": 3.5036248405242356e-06,
      "loss": 0.99252188,
      "num_input_tokens_seen": 2663825,
      "step": 126,
      "time_per_iteration": 2.642979621887207
    },
    {
      "auxiliary_loss_clip": 0.0151066,
      "auxiliary_loss_mlp": 0.01113067,
      "balance_loss_clip": 1.15819657,
      "balance_loss_mlp": 1.0669564,
      "epoch": 0.015270847111164552,
      "flos": 39420184060800.0,
      "grad_norm": 1.864338377495763,
      "language_loss": 0.82909191,
      "learning_rate": 3.509351715802146e-06,
      "loss": 0.85532922,
      "num_input_tokens_seen": 2684710,
      "step": 127,
      "time_per_iteration": 2.7415287494659424
    },
    {
      "auxiliary_loss_clip": 0.01508807,
      "auxiliary_loss_mlp": 0.01124026,
      "balance_loss_clip": 1.1564486,
      "balance_loss_mlp": 1.07691443,
      "epoch": 0.015391090001803644,
      "flos": 43762466286720.0,
      "grad_norm": 6.730558788297959,
      "language_loss": 0.78354734,
      "learning_rate": 3.5150336739488763e-06,
      "loss": 0.80987561,
      "num_input_tokens_seen": 2706995,
      "step": 128,
      "time_per_iteration": 2.8090872764587402
    },
    {
      "auxiliary_loss_clip": 0.01503317,
      "auxiliary_loss_mlp": 0.01090642,
      "balance_loss_clip": 1.15659976,
      "balance_loss_mlp": 1.05115986,
      "epoch": 0.015511332892442733,
      "flos": 18916341287040.0,
      "grad_norm": 1.9427516344510842,
      "language_loss": 0.84346092,
      "learning_rate": 3.5206714140744143e-06,
      "loss": 0.8694005,
      "num_input_tokens_seen": 2727050,
      "step": 129,
      "time_per_iteration": 2.6808488368988037
    },
    {
      "auxiliary_loss_clip": 0.01505657,
      "auxiliary_loss_mlp": 0.0111845,
      "balance_loss_clip": 1.15936875,
      "balance_loss_mlp": 1.07577288,
      "epoch": 0.015631575783081827,
      "flos": 24535283679360.0,
      "grad_norm": 4.73754184153692,
      "language_loss": 0.87491655,
      "learning_rate": 3.5262656190928208e-06,
      "loss": 0.90115762,
      "num_input_tokens_seen": 2745350,
      "step": 130,
      "time_per_iteration": 2.5958337783813477
    },
    {
      "auxiliary_loss_clip": 0.01452486,
      "auxiliary_loss_mlp": 0.01031608,
      "balance_loss_clip": 1.17128634,
      "balance_loss_mlp": 1.01797032,
      "epoch": 0.015751818673720917,
      "flos": 62328536098560.0,
      "grad_norm": 1.0704955359163306,
      "language_loss": 0.7151376,
      "learning_rate": 3.5318169562186737e-06,
      "loss": 0.73997855,
      "num_input_tokens_seen": 2814195,
      "step": 131,
      "time_per_iteration": 3.183711528778076
    },
    {
      "auxiliary_loss_clip": 0.01492113,
      "auxiliary_loss_mlp": 0.01122596,
      "balance_loss_clip": 1.15217769,
      "balance_loss_mlp": 1.08251762,
      "epoch": 0.015872061564360006,
      "flos": 23878549365120.0,
      "grad_norm": 1.7558376584742281,
      "language_loss": 0.82305408,
      "learning_rate": 3.5373260774446292e-06,
      "loss": 0.8492012,
      "num_input_tokens_seen": 2834645,
      "step": 132,
      "time_per_iteration": 2.642026901245117
    },
    {
      "auxiliary_loss_clip": 0.01489732,
      "auxiliary_loss_mlp": 0.01116723,
      "balance_loss_clip": 1.15059876,
      "balance_loss_mlp": 1.07655001,
      "epoch": 0.0159923044549991,
      "flos": 23367899664000.0,
      "grad_norm": 1.9892344404559494,
      "language_loss": 0.90380722,
      "learning_rate": 3.542793620000961e-06,
      "loss": 0.9298718,
      "num_input_tokens_seen": 2854120,
      "step": 133,
      "time_per_iteration": 2.566098928451538
    },
    {
      "auxiliary_loss_clip": 0.01487185,
      "auxiliary_loss_mlp": 0.01106252,
      "balance_loss_clip": 1.14991784,
      "balance_loss_mlp": 1.06557775,
      "epoch": 0.01611254734563819,
      "flos": 17858305249920.0,
      "grad_norm": 2.376586049494334,
      "language_loss": 0.86960846,
      "learning_rate": 3.5482202067978894e-06,
      "loss": 0.8955428,
      "num_input_tokens_seen": 2871330,
      "step": 134,
      "time_per_iteration": 2.5655272006988525
    },
    {
      "auxiliary_loss_clip": 0.01485696,
      "auxiliary_loss_mlp": 0.01100606,
      "balance_loss_clip": 1.1504935,
      "balance_loss_mlp": 1.06079006,
      "epoch": 0.01623279023627728,
      "flos": 20954774113920.0,
      "grad_norm": 2.9914096949009026,
      "language_loss": 0.76311105,
      "learning_rate": 3.553606446851471e-06,
      "loss": 0.78897405,
      "num_input_tokens_seen": 2888070,
      "step": 135,
      "time_per_iteration": 2.590481996536255
    },
    {
      "auxiliary_loss_clip": 0.01472415,
      "auxiliary_loss_mlp": 0.0109796,
      "balance_loss_clip": 1.14223146,
      "balance_loss_mlp": 1.05871665,
      "epoch": 0.016353033126916373,
      "flos": 15742412743680.0,
      "grad_norm": 1.9208589061965191,
      "language_loss": 0.83461642,
      "learning_rate": 3.5589529356937613e-06,
      "loss": 0.86032015,
      "num_input_tokens_seen": 2906465,
      "step": 136,
      "time_per_iteration": 2.5364911556243896
    },
    {
      "auxiliary_loss_clip": 0.0148191,
      "auxiliary_loss_mlp": 0.01101312,
      "balance_loss_clip": 1.14590597,
      "balance_loss_mlp": 1.0625453,
      "epoch": 0.016473276017555463,
      "flos": 18807280617600.0,
      "grad_norm": 2.457584216005113,
      "language_loss": 0.77103424,
      "learning_rate": 3.5642602557679627e-06,
      "loss": 0.79686648,
      "num_input_tokens_seen": 2924915,
      "step": 137,
      "time_per_iteration": 2.5617358684539795
    },
    {
      "auxiliary_loss_clip": 0.01475814,
      "auxiliary_loss_mlp": 0.01091411,
      "balance_loss_clip": 1.15158582,
      "balance_loss_mlp": 1.05798447,
      "epoch": 0.016593518908194552,
      "flos": 24352641999360.0,
      "grad_norm": 3.3367477928744576,
      "language_loss": 0.84288585,
      "learning_rate": 3.569528976809202e-06,
      "loss": 0.86855817,
      "num_input_tokens_seen": 2942130,
      "step": 138,
      "time_per_iteration": 2.5901236534118652
    },
    {
      "auxiliary_loss_clip": 0.0147646,
      "auxiliary_loss_mlp": 0.01108912,
      "balance_loss_clip": 1.14605927,
      "balance_loss_mlp": 1.06919122,
      "epoch": 0.016713761798833646,
      "flos": 22346133384960.0,
      "grad_norm": 1.7246101274504957,
      "language_loss": 0.89972842,
      "learning_rate": 3.5747596562115522e-06,
      "loss": 0.92558217,
      "num_input_tokens_seen": 2962745,
      "step": 139,
      "time_per_iteration": 2.6615800857543945
    },
    {
      "auxiliary_loss_clip": 0.01480523,
      "auxiliary_loss_mlp": 0.01106728,
      "balance_loss_clip": 1.14758432,
      "balance_loss_mlp": 1.06929624,
      "epoch": 0.016834004689472735,
      "flos": 17821820010240.0,
      "grad_norm": 2.316607481235765,
      "language_loss": 0.90950894,
      "learning_rate": 3.5799528393819138e-06,
      "loss": 0.93538153,
      "num_input_tokens_seen": 2981825,
      "step": 140,
      "time_per_iteration": 2.668705701828003
    },
    {
      "auxiliary_loss_clip": 0.01462849,
      "auxiliary_loss_mlp": 0.01100261,
      "balance_loss_clip": 1.13782465,
      "balance_loss_mlp": 1.06509459,
      "epoch": 0.016954247580111825,
      "flos": 20519501103360.0,
      "grad_norm": 2.218289125596974,
      "language_loss": 0.8820675,
      "learning_rate": 3.585109060081286e-06,
      "loss": 0.90769857,
      "num_input_tokens_seen": 3001625,
      "step": 141,
      "time_per_iteration": 2.556135416030884
    },
    {
      "auxiliary_loss_clip": 0.01469179,
      "auxiliary_loss_mlp": 0.011012,
      "balance_loss_clip": 1.14193761,
      "balance_loss_mlp": 1.06574762,
      "epoch": 0.017074490470750915,
      "flos": 22088869200000.0,
      "grad_norm": 1.7956846231971892,
      "language_loss": 0.78660738,
      "learning_rate": 3.590228840753992e-06,
      "loss": 0.81231105,
      "num_input_tokens_seen": 3022055,
      "step": 142,
      "time_per_iteration": 2.580550193786621
    },
    {
      "auxiliary_loss_clip": 0.01459938,
      "auxiliary_loss_mlp": 0.01103534,
      "balance_loss_clip": 1.13899851,
      "balance_loss_mlp": 1.07010746,
      "epoch": 0.01719473336139001,
      "flos": 15997270717440.0,
      "grad_norm": 2.028740299574435,
      "language_loss": 0.87280518,
      "learning_rate": 3.5953126928453423e-06,
      "loss": 0.89843988,
      "num_input_tokens_seen": 3039605,
      "step": 143,
      "time_per_iteration": 2.516853094100952
    },
    {
      "auxiliary_loss_clip": 0.01457036,
      "auxiliary_loss_mlp": 0.01084669,
      "balance_loss_clip": 1.13552809,
      "balance_loss_mlp": 1.05260146,
      "epoch": 0.017314976252029098,
      "flos": 22492038430080.0,
      "grad_norm": 1.9655212222122012,
      "language_loss": 0.80619258,
      "learning_rate": 3.600361117108239e-06,
      "loss": 0.83160961,
      "num_input_tokens_seen": 3059405,
      "step": 144,
      "time_per_iteration": 2.574777364730835
    },
    {
      "auxiliary_loss_clip": 0.01461691,
      "auxiliary_loss_mlp": 0.01090405,
      "balance_loss_clip": 1.13679838,
      "balance_loss_mlp": 1.05633473,
      "epoch": 0.017435219142668188,
      "flos": 22018053536640.0,
      "grad_norm": 2.0625245833236034,
      "language_loss": 0.97196126,
      "learning_rate": 3.6053746038991616e-06,
      "loss": 0.99748224,
      "num_input_tokens_seen": 3078490,
      "step": 145,
      "time_per_iteration": 2.5558927059173584
    },
    {
      "auxiliary_loss_clip": 0.01407729,
      "auxiliary_loss_mlp": 0.01010883,
      "balance_loss_clip": 1.15419316,
      "balance_loss_mlp": 0.99982017,
      "epoch": 0.01755546203330728,
      "flos": 72240526149120.0,
      "grad_norm": 1.0618487226235793,
      "language_loss": 0.58423042,
      "learning_rate": 3.6103536334639843e-06,
      "loss": 0.60841656,
      "num_input_tokens_seen": 3131755,
      "step": 146,
      "time_per_iteration": 3.109741687774658
    },
    {
      "auxiliary_loss_clip": 0.01451087,
      "auxiliary_loss_mlp": 0.01087139,
      "balance_loss_clip": 1.13362956,
      "balance_loss_mlp": 1.05540562,
      "epoch": 0.01767570492394637,
      "flos": 25337061112320.0,
      "grad_norm": 1.9974747287052388,
      "language_loss": 0.85786569,
      "learning_rate": 3.615298676214041e-06,
      "loss": 0.88324791,
      "num_input_tokens_seen": 3152035,
      "step": 147,
      "time_per_iteration": 2.617709159851074
    },
    {
      "auxiliary_loss_clip": 0.01449112,
      "auxiliary_loss_mlp": 0.010997,
      "balance_loss_clip": 1.13172388,
      "balance_loss_mlp": 1.06900418,
      "epoch": 0.01779594781458546,
      "flos": 20449188230400.0,
      "grad_norm": 2.1250273792771837,
      "language_loss": 0.88978142,
      "learning_rate": 3.6202101929928317e-06,
      "loss": 0.91526949,
      "num_input_tokens_seen": 3170625,
      "step": 148,
      "time_per_iteration": 2.5402638912200928
    },
    {
      "auxiliary_loss_clip": 0.01443985,
      "auxiliary_loss_mlp": 0.01093728,
      "balance_loss_clip": 1.12985492,
      "balance_loss_mlp": 1.06410456,
      "epoch": 0.017916190705224554,
      "flos": 16253601148800.0,
      "grad_norm": 2.0052308562285535,
      "language_loss": 0.88471079,
      "learning_rate": 3.6250886353337413e-06,
      "loss": 0.91008788,
      "num_input_tokens_seen": 3188155,
      "step": 149,
      "time_per_iteration": 2.5184099674224854
    },
    {
      "auxiliary_loss_clip": 0.01456577,
      "auxiliary_loss_mlp": 0.01095739,
      "balance_loss_clip": 1.13688076,
      "balance_loss_mlp": 1.06607938,
      "epoch": 0.018036433595863644,
      "flos": 23330588411520.0,
      "grad_norm": 2.5097661483820266,
      "language_loss": 0.86353678,
      "learning_rate": 3.6299344457091488e-06,
      "loss": 0.8890599,
      "num_input_tokens_seen": 3209015,
      "step": 150,
      "time_per_iteration": 2.6522512435913086
    },
    {
      "auxiliary_loss_clip": 0.01448062,
      "auxiliary_loss_mlp": 0.01085606,
      "balance_loss_clip": 1.1330452,
      "balance_loss_mlp": 1.05735326,
      "epoch": 0.018156676486502734,
      "flos": 18588010043520.0,
      "grad_norm": 2.1232483638502346,
      "language_loss": 0.93856239,
      "learning_rate": 3.634748057771256e-06,
      "loss": 0.96389908,
      "num_input_tokens_seen": 3224955,
      "step": 151,
      "time_per_iteration": 4.405697584152222
    },
    {
      "auxiliary_loss_clip": 0.01441102,
      "auxiliary_loss_mlp": 0.01090787,
      "balance_loss_clip": 1.13099337,
      "balance_loss_mlp": 1.06257057,
      "epoch": 0.018276919377141827,
      "flos": 25448707560960.0,
      "grad_norm": 1.8128308807158346,
      "language_loss": 0.85941714,
      "learning_rate": 3.639529896584965e-06,
      "loss": 0.88473606,
      "num_input_tokens_seen": 3246330,
      "step": 152,
      "time_per_iteration": 3.4095232486724854
    },
    {
      "auxiliary_loss_clip": 0.01442831,
      "auxiliary_loss_mlp": 0.01080838,
      "balance_loss_clip": 1.13019538,
      "balance_loss_mlp": 1.05102372,
      "epoch": 0.018397162267780917,
      "flos": 20047311889920.0,
      "grad_norm": 2.9186730443635307,
      "language_loss": 0.89245641,
      "learning_rate": 3.6442803788531233e-06,
      "loss": 0.91769302,
      "num_input_tokens_seen": 3264290,
      "step": 153,
      "time_per_iteration": 3.3136463165283203
    },
    {
      "auxiliary_loss_clip": 0.0144434,
      "auxiliary_loss_mlp": 0.01092801,
      "balance_loss_clip": 1.12938178,
      "balance_loss_mlp": 1.06197381,
      "epoch": 0.018517405158420007,
      "flos": 27565282425600.0,
      "grad_norm": 2.2904171111991043,
      "language_loss": 0.96016854,
      "learning_rate": 3.6489999131344357e-06,
      "loss": 0.98553991,
      "num_input_tokens_seen": 3287065,
      "step": 154,
      "time_per_iteration": 2.584468126296997
    },
    {
      "auxiliary_loss_clip": 0.01432768,
      "auxiliary_loss_mlp": 0.0109073,
      "balance_loss_clip": 1.12618279,
      "balance_loss_mlp": 1.06401515,
      "epoch": 0.0186376480490591,
      "flos": 19354056422400.0,
      "grad_norm": 1.7776320407094601,
      "language_loss": 0.9060998,
      "learning_rate": 3.653688900054313e-06,
      "loss": 0.93133479,
      "num_input_tokens_seen": 3305595,
      "step": 155,
      "time_per_iteration": 2.5302257537841797
    },
    {
      "auxiliary_loss_clip": 0.01435272,
      "auxiliary_loss_mlp": 0.01069098,
      "balance_loss_clip": 1.12378955,
      "balance_loss_mlp": 1.04096508,
      "epoch": 0.01875789093969819,
      "flos": 26687840993280.0,
      "grad_norm": 2.417757028032494,
      "language_loss": 0.76047444,
      "learning_rate": 3.6583477325089526e-06,
      "loss": 0.78551811,
      "num_input_tokens_seen": 3326135,
      "step": 156,
      "time_per_iteration": 2.585200548171997
    },
    {
      "auxiliary_loss_clip": 0.01429677,
      "auxiliary_loss_mlp": 0.01079277,
      "balance_loss_clip": 1.12295747,
      "balance_loss_mlp": 1.0518589,
      "epoch": 0.01887813383033728,
      "flos": 24353001135360.0,
      "grad_norm": 2.207466324006412,
      "language_loss": 1.04357588,
      "learning_rate": 3.6629767958628916e-06,
      "loss": 1.0686655,
      "num_input_tokens_seen": 3343510,
      "step": 157,
      "time_per_iteration": 2.5631606578826904
    },
    {
      "auxiliary_loss_clip": 0.01426265,
      "auxiliary_loss_mlp": 0.01080537,
      "balance_loss_clip": 1.12468541,
      "balance_loss_mlp": 1.05251074,
      "epoch": 0.018998376720976373,
      "flos": 14647532330880.0,
      "grad_norm": 2.4247202971029784,
      "language_loss": 0.85458946,
      "learning_rate": 3.667576468140291e-06,
      "loss": 0.87965751,
      "num_input_tokens_seen": 3361325,
      "step": 158,
      "time_per_iteration": 2.5459694862365723
    },
    {
      "auxiliary_loss_clip": 0.01419749,
      "auxiliary_loss_mlp": 0.01064302,
      "balance_loss_clip": 1.11828184,
      "balance_loss_mlp": 1.03892207,
      "epoch": 0.019118619611615463,
      "flos": 29305261146240.0,
      "grad_norm": 2.211349109085119,
      "language_loss": 0.88938743,
      "learning_rate": 3.672147120210184e-06,
      "loss": 0.91422796,
      "num_input_tokens_seen": 3377925,
      "step": 159,
      "time_per_iteration": 2.599454164505005
    },
    {
      "auxiliary_loss_clip": 0.01426579,
      "auxiliary_loss_mlp": 0.01076261,
      "balance_loss_clip": 1.12569165,
      "balance_loss_mlp": 1.0512867,
      "epoch": 0.019238862502254553,
      "flos": 20886723797760.0,
      "grad_norm": 1.9756263559848335,
      "language_loss": 0.86620474,
      "learning_rate": 3.6766891159659177e-06,
      "loss": 0.89123309,
      "num_input_tokens_seen": 3396335,
      "step": 160,
      "time_per_iteration": 2.537170886993408
    },
    {
      "auxiliary_loss_clip": 0.01426305,
      "auxiliary_loss_mlp": 0.01078336,
      "balance_loss_clip": 1.12688708,
      "balance_loss_mlp": 1.05296862,
      "epoch": 0.019359105392893646,
      "flos": 21360672777600.0,
      "grad_norm": 3.2657088674070787,
      "language_loss": 0.8801288,
      "learning_rate": 3.6812028124990075e-06,
      "loss": 0.90517521,
      "num_input_tokens_seen": 3413605,
      "step": 161,
      "time_per_iteration": 2.6275370121002197
    },
    {
      "auxiliary_loss_clip": 0.01421357,
      "auxiliary_loss_mlp": 0.01082785,
      "balance_loss_clip": 1.1231519,
      "balance_loss_mlp": 1.05837083,
      "epoch": 0.019479348283532736,
      "flos": 16283729681280.0,
      "grad_norm": 4.047007746984585,
      "language_loss": 0.81460655,
      "learning_rate": 3.6856885602676016e-06,
      "loss": 0.83964801,
      "num_input_tokens_seen": 3429640,
      "step": 162,
      "time_per_iteration": 2.507634162902832
    },
    {
      "auxiliary_loss_clip": 0.01420076,
      "auxiliary_loss_mlp": 0.01084318,
      "balance_loss_clip": 1.12288117,
      "balance_loss_mlp": 1.06010652,
      "epoch": 0.019599591174171826,
      "flos": 22091239497600.0,
      "grad_norm": 2.044509185201751,
      "language_loss": 0.94333065,
      "learning_rate": 3.6901467032597733e-06,
      "loss": 0.96837461,
      "num_input_tokens_seen": 3448125,
      "step": 163,
      "time_per_iteration": 2.6433823108673096
    },
    {
      "auxiliary_loss_clip": 0.01422195,
      "auxiliary_loss_mlp": 0.01070052,
      "balance_loss_clip": 1.12144852,
      "balance_loss_mlp": 1.04351616,
      "epoch": 0.01971983406481092,
      "flos": 19609668581760.0,
      "grad_norm": 2.2499308276572494,
      "language_loss": 0.87472332,
      "learning_rate": 3.694577579151804e-06,
      "loss": 0.89964581,
      "num_input_tokens_seen": 3466535,
      "step": 164,
      "time_per_iteration": 2.6180763244628906
    },
    {
      "auxiliary_loss_clip": 0.01421243,
      "auxiliary_loss_mlp": 0.01079805,
      "balance_loss_clip": 1.12326443,
      "balance_loss_mlp": 1.05434155,
      "epoch": 0.01984007695545001,
      "flos": 19099342103040.0,
      "grad_norm": 2.2160376331278835,
      "language_loss": 0.73723412,
      "learning_rate": 3.6989815194616703e-06,
      "loss": 0.76224458,
      "num_input_tokens_seen": 3483730,
      "step": 165,
      "time_per_iteration": 2.5633740425109863
    },
    {
      "auxiliary_loss_clip": 0.01420118,
      "auxiliary_loss_mlp": 0.01080578,
      "balance_loss_clip": 1.11949909,
      "balance_loss_mlp": 1.05358934,
      "epoch": 0.0199603198460891,
      "flos": 20848406964480.0,
      "grad_norm": 2.8469479377242544,
      "language_loss": 0.79968739,
      "learning_rate": 3.703358849697888e-06,
      "loss": 0.8246944,
      "num_input_tokens_seen": 3503640,
      "step": 166,
      "time_per_iteration": 2.601308584213257
    },
    {
      "auxiliary_loss_clip": 0.01417344,
      "auxiliary_loss_mlp": 0.01089618,
      "balance_loss_clip": 1.12307215,
      "balance_loss_mlp": 1.06593156,
      "epoch": 0.020080562736728192,
      "flos": 21870747861120.0,
      "grad_norm": 1.7661579960379223,
      "language_loss": 0.82708067,
      "learning_rate": 3.7077098895038803e-06,
      "loss": 0.85215032,
      "num_input_tokens_seen": 3523010,
      "step": 167,
      "time_per_iteration": 2.554431200027466
    },
    {
      "auxiliary_loss_clip": 0.01416286,
      "auxiliary_loss_mlp": 0.01076642,
      "balance_loss_clip": 1.12099493,
      "balance_loss_mlp": 1.05264544,
      "epoch": 0.020200805627367282,
      "flos": 21688788539520.0,
      "grad_norm": 2.228780715329253,
      "language_loss": 0.97167206,
      "learning_rate": 3.712034952798045e-06,
      "loss": 0.99660122,
      "num_input_tokens_seen": 3541125,
      "step": 168,
      "time_per_iteration": 2.521315097808838
    },
    {
      "auxiliary_loss_clip": 0.0141256,
      "auxiliary_loss_mlp": 0.01084325,
      "balance_loss_clip": 1.1159389,
      "balance_loss_mlp": 1.05980313,
      "epoch": 0.02032104851800637,
      "flos": 33543043729920.0,
      "grad_norm": 5.603565505184597,
      "language_loss": 0.84590304,
      "learning_rate": 3.7163343479096656e-06,
      "loss": 0.87087184,
      "num_input_tokens_seen": 3562700,
      "step": 169,
      "time_per_iteration": 2.6373016834259033
    },
    {
      "auxiliary_loss_clip": 0.01410202,
      "auxiliary_loss_mlp": 0.01076741,
      "balance_loss_clip": 1.11913192,
      "balance_loss_mlp": 1.05486584,
      "epoch": 0.020441291408645465,
      "flos": 31686965274240.0,
      "grad_norm": 2.948590484015234,
      "language_loss": 0.83056402,
      "learning_rate": 3.720608377710802e-06,
      "loss": 0.8554334,
      "num_input_tokens_seen": 3582790,
      "step": 170,
      "time_per_iteration": 2.628444194793701
    },
    {
      "auxiliary_loss_clip": 0.01403473,
      "auxiliary_loss_mlp": 0.01087102,
      "balance_loss_clip": 1.1130209,
      "balance_loss_mlp": 1.0624969,
      "epoch": 0.020561534299284555,
      "flos": 20886687884160.0,
      "grad_norm": 2.0495006095686255,
      "language_loss": 0.86342776,
      "learning_rate": 3.7248573397443277e-06,
      "loss": 0.88833356,
      "num_input_tokens_seen": 3601715,
      "step": 171,
      "time_per_iteration": 2.5745770931243896
    },
    {
      "auxiliary_loss_clip": 0.01409233,
      "auxiliary_loss_mlp": 0.0109003,
      "balance_loss_clip": 1.11958981,
      "balance_loss_mlp": 1.06468582,
      "epoch": 0.020681777189923645,
      "flos": 20996610480000.0,
      "grad_norm": 2.0797316398658534,
      "language_loss": 0.97813094,
      "learning_rate": 3.729081526348224e-06,
      "loss": 1.00312352,
      "num_input_tokens_seen": 3620245,
      "step": 172,
      "time_per_iteration": 2.6195945739746094
    },
    {
      "auxiliary_loss_clip": 0.01410013,
      "auxiliary_loss_mlp": 0.01068789,
      "balance_loss_clip": 1.11768854,
      "balance_loss_mlp": 1.04669952,
      "epoch": 0.020802020080562738,
      "flos": 28257532312320.0,
      "grad_norm": 1.936306971283965,
      "language_loss": 0.84899962,
      "learning_rate": 3.7332812247762777e-06,
      "loss": 0.87378764,
      "num_input_tokens_seen": 3641545,
      "step": 173,
      "time_per_iteration": 2.6253838539123535
    },
    {
      "auxiliary_loss_clip": 0.01410098,
      "auxiliary_loss_mlp": 0.01066071,
      "balance_loss_clip": 1.12051344,
      "balance_loss_mlp": 1.04311168,
      "epoch": 0.020922262971201828,
      "flos": 19681274344320.0,
      "grad_norm": 2.4853459240993656,
      "language_loss": 0.95718849,
      "learning_rate": 3.737456717315293e-06,
      "loss": 0.98195016,
      "num_input_tokens_seen": 3660510,
      "step": 174,
      "time_per_iteration": 2.6722049713134766
    },
    {
      "auxiliary_loss_clip": 0.01398191,
      "auxiliary_loss_mlp": 0.01087759,
      "balance_loss_clip": 1.11585402,
      "balance_loss_mlp": 1.06469226,
      "epoch": 0.021042505861840918,
      "flos": 15666353694720.0,
      "grad_norm": 1.7243215051599294,
      "language_loss": 0.9063713,
      "learning_rate": 3.7416082813989552e-06,
      "loss": 0.93123078,
      "num_input_tokens_seen": 3677505,
      "step": 175,
      "time_per_iteration": 2.5443568229675293
    },
    {
      "auxiliary_loss_clip": 0.01406896,
      "auxiliary_loss_mlp": 0.01079216,
      "balance_loss_clip": 1.11785543,
      "balance_loss_mlp": 1.05564809,
      "epoch": 0.02116274875248001,
      "flos": 21142012734720.0,
      "grad_norm": 1.9655892053854447,
      "language_loss": 0.89469743,
      "learning_rate": 3.745736189718439e-06,
      "loss": 0.91955858,
      "num_input_tokens_seen": 3696760,
      "step": 176,
      "time_per_iteration": 2.535235643386841
    },
    {
      "auxiliary_loss_clip": 0.01396913,
      "auxiliary_loss_mlp": 0.01065821,
      "balance_loss_clip": 1.11271834,
      "balance_loss_mlp": 1.04332662,
      "epoch": 0.0212829916431191,
      "flos": 24715770543360.0,
      "grad_norm": 2.7040320607491797,
      "language_loss": 0.72632468,
      "learning_rate": 3.749840710329894e-06,
      "loss": 0.75095206,
      "num_input_tokens_seen": 3717465,
      "step": 177,
      "time_per_iteration": 2.575251817703247
    },
    {
      "auxiliary_loss_clip": 0.01408161,
      "auxiliary_loss_mlp": 0.01085532,
      "balance_loss_clip": 1.11644828,
      "balance_loss_mlp": 1.06078386,
      "epoch": 0.02140323453375819,
      "flos": 16645493508480.0,
      "grad_norm": 3.6053867680093465,
      "language_loss": 0.98011255,
      "learning_rate": 3.7539221067588938e-06,
      "loss": 1.00504947,
      "num_input_tokens_seen": 3731440,
      "step": 178,
      "time_per_iteration": 4.1849565505981445
    },
    {
      "auxiliary_loss_clip": 0.01402694,
      "auxiliary_loss_mlp": 0.01084266,
      "balance_loss_clip": 1.11476171,
      "balance_loss_mlp": 1.0603168,
      "epoch": 0.021523477424397284,
      "flos": 20299332689280.0,
      "grad_norm": 4.752666604168373,
      "language_loss": 0.93415415,
      "learning_rate": 3.757980638101964e-06,
      "loss": 0.95902377,
      "num_input_tokens_seen": 3744935,
      "step": 179,
      "time_per_iteration": 3.2877705097198486
    },
    {
      "auxiliary_loss_clip": 0.01405218,
      "auxiliary_loss_mlp": 0.01076861,
      "balance_loss_clip": 1.11657786,
      "balance_loss_mlp": 1.05145788,
      "epoch": 0.021643720315036374,
      "flos": 26104005331200.0,
      "grad_norm": 2.4511298870031917,
      "language_loss": 0.8913734,
      "learning_rate": 3.7620165591252806e-06,
      "loss": 0.9161942,
      "num_input_tokens_seen": 3763035,
      "step": 180,
      "time_per_iteration": 3.358435869216919
    },
    {
      "auxiliary_loss_clip": 0.01394805,
      "auxiliary_loss_mlp": 0.01072653,
      "balance_loss_clip": 1.11491966,
      "balance_loss_mlp": 1.05084956,
      "epoch": 0.021763963205675464,
      "flos": 24787663614720.0,
      "grad_norm": 1.7665753306860168,
      "language_loss": 0.9454397,
      "learning_rate": 3.766030120360636e-06,
      "loss": 0.97011423,
      "num_input_tokens_seen": 3782665,
      "step": 181,
      "time_per_iteration": 2.548003911972046
    },
    {
      "auxiliary_loss_clip": 0.01401304,
      "auxiliary_loss_mlp": 0.01074823,
      "balance_loss_clip": 1.11475885,
      "balance_loss_mlp": 1.05270934,
      "epoch": 0.021884206096314557,
      "flos": 25813559957760.0,
      "grad_norm": 2.1208633989575887,
      "language_loss": 0.90331209,
      "learning_rate": 3.7700215681987578e-06,
      "loss": 0.92807329,
      "num_input_tokens_seen": 3802435,
      "step": 182,
      "time_per_iteration": 2.590991258621216
    },
    {
      "auxiliary_loss_clip": 0.01393931,
      "auxiliary_loss_mlp": 0.01085684,
      "balance_loss_clip": 1.11251068,
      "balance_loss_mlp": 1.06237805,
      "epoch": 0.022004448986953647,
      "flos": 20082719721600.0,
      "grad_norm": 1.9346909068289775,
      "language_loss": 0.82292002,
      "learning_rate": 3.7739911449800767e-06,
      "loss": 0.84771615,
      "num_input_tokens_seen": 3822490,
      "step": 183,
      "time_per_iteration": 2.6303699016571045
    },
    {
      "auxiliary_loss_clip": 0.01394469,
      "auxiliary_loss_mlp": 0.01085175,
      "balance_loss_clip": 1.11118197,
      "balance_loss_mlp": 1.06401491,
      "epoch": 0.022124691877592736,
      "flos": 20480609652480.0,
      "grad_norm": 2.4506076648523143,
      "language_loss": 0.80815208,
      "learning_rate": 3.7779390890830114e-06,
      "loss": 0.83294851,
      "num_input_tokens_seen": 3841140,
      "step": 184,
      "time_per_iteration": 2.582005500793457
    },
    {
      "auxiliary_loss_clip": 0.01394968,
      "auxiliary_loss_mlp": 0.01084229,
      "balance_loss_clip": 1.11145091,
      "balance_loss_mlp": 1.06118631,
      "epoch": 0.02224493476823183,
      "flos": 23586847015680.0,
      "grad_norm": 2.5213755960115054,
      "language_loss": 0.85919791,
      "learning_rate": 3.7818656350098723e-06,
      "loss": 0.88398993,
      "num_input_tokens_seen": 3862090,
      "step": 185,
      "time_per_iteration": 2.622337818145752
    },
    {
      "auxiliary_loss_clip": 0.0138934,
      "auxiliary_loss_mlp": 0.01075058,
      "balance_loss_clip": 1.10778046,
      "balance_loss_mlp": 1.05164576,
      "epoch": 0.02236517765887092,
      "flos": 16909940413440.0,
      "grad_norm": 2.5423084023343017,
      "language_loss": 0.7713865,
      "learning_rate": 3.7857710134704447e-06,
      "loss": 0.79603046,
      "num_input_tokens_seen": 3881025,
      "step": 186,
      "time_per_iteration": 2.530702829360962
    },
    {
      "auxiliary_loss_clip": 0.01388985,
      "auxiliary_loss_mlp": 0.01057624,
      "balance_loss_clip": 1.1122731,
      "balance_loss_mlp": 1.0367744,
      "epoch": 0.02248542054951001,
      "flos": 43508182930560.0,
      "grad_norm": 3.055214000823869,
      "language_loss": 0.79375297,
      "learning_rate": 3.7896554514633234e-06,
      "loss": 0.81821907,
      "num_input_tokens_seen": 3905310,
      "step": 187,
      "time_per_iteration": 2.703484296798706
    },
    {
      "auxiliary_loss_clip": 0.01387329,
      "auxiliary_loss_mlp": 0.01068577,
      "balance_loss_clip": 1.11025524,
      "balance_loss_mlp": 1.04735792,
      "epoch": 0.022605663440149103,
      "flos": 23367648268800.0,
      "grad_norm": 3.9384464579489027,
      "language_loss": 0.84562314,
      "learning_rate": 3.7935191723550955e-06,
      "loss": 0.87018216,
      "num_input_tokens_seen": 3924265,
      "step": 188,
      "time_per_iteration": 2.602079391479492
    },
    {
      "auxiliary_loss_clip": 0.01386209,
      "auxiliary_loss_mlp": 0.01071827,
      "balance_loss_clip": 1.10891056,
      "balance_loss_mlp": 1.05145431,
      "epoch": 0.022725906330788193,
      "flos": 29019915504000.0,
      "grad_norm": 2.1404268308171406,
      "language_loss": 0.88815933,
      "learning_rate": 3.797362395957408e-06,
      "loss": 0.91273969,
      "num_input_tokens_seen": 3944830,
      "step": 189,
      "time_per_iteration": 2.6051111221313477
    },
    {
      "auxiliary_loss_clip": 0.01396753,
      "auxiliary_loss_mlp": 0.01070363,
      "balance_loss_clip": 1.11555684,
      "balance_loss_mlp": 1.04886961,
      "epoch": 0.022846149221427282,
      "flos": 24496176746880.0,
      "grad_norm": 2.0484068975039684,
      "language_loss": 0.7848438,
      "learning_rate": 3.8011853386020055e-06,
      "loss": 0.809515,
      "num_input_tokens_seen": 3965735,
      "step": 190,
      "time_per_iteration": 2.59596586227417
    },
    {
      "auxiliary_loss_clip": 0.01392041,
      "auxiliary_loss_mlp": 0.01084012,
      "balance_loss_clip": 1.11264849,
      "balance_loss_mlp": 1.06164813,
      "epoch": 0.022966392112066376,
      "flos": 15523537219200.0,
      "grad_norm": 3.5710873371832825,
      "language_loss": 0.90078062,
      "learning_rate": 3.804988213213804e-06,
      "loss": 0.92554116,
      "num_input_tokens_seen": 3983975,
      "step": 191,
      "time_per_iteration": 2.5427801609039307
    },
    {
      "auxiliary_loss_clip": 0.01378642,
      "auxiliary_loss_mlp": 0.01020948,
      "balance_loss_clip": 1.15688062,
      "balance_loss_mlp": 1.00916982,
      "epoch": 0.023086635002705466,
      "flos": 55650408433920.0,
      "grad_norm": 1.0193135268128133,
      "language_loss": 0.6316992,
      "learning_rate": 3.808771229382049e-06,
      "loss": 0.65569508,
      "num_input_tokens_seen": 4043440,
      "step": 192,
      "time_per_iteration": 3.0310378074645996
    },
    {
      "auxiliary_loss_clip": 0.01383525,
      "auxiliary_loss_mlp": 0.01079605,
      "balance_loss_clip": 1.11008716,
      "balance_loss_mlp": 1.0592916,
      "epoch": 0.023206877893344555,
      "flos": 19313441118720.0,
      "grad_norm": 2.3585366127604876,
      "language_loss": 0.84396625,
      "learning_rate": 3.8125345934296324e-06,
      "loss": 0.86859757,
      "num_input_tokens_seen": 4061750,
      "step": 193,
      "time_per_iteration": 2.5186355113983154
    },
    {
      "auxiliary_loss_clip": 0.01384921,
      "auxiliary_loss_mlp": 0.01075543,
      "balance_loss_clip": 1.10954964,
      "balance_loss_mlp": 1.05296457,
      "epoch": 0.02332712078398365,
      "flos": 23072965090560.0,
      "grad_norm": 2.753611041070088,
      "language_loss": 0.88016462,
      "learning_rate": 3.81627850848061e-06,
      "loss": 0.9047693,
      "num_input_tokens_seen": 4082345,
      "step": 194,
      "time_per_iteration": 2.5787055492401123
    },
    {
      "auxiliary_loss_clip": 0.01379018,
      "auxiliary_loss_mlp": 0.01066263,
      "balance_loss_clip": 1.10537577,
      "balance_loss_mlp": 1.04605734,
      "epoch": 0.02344736367462274,
      "flos": 24425971614720.0,
      "grad_norm": 2.2039583775921376,
      "language_loss": 0.86284089,
      "learning_rate": 3.820003174525994e-06,
      "loss": 0.88729376,
      "num_input_tokens_seen": 4101770,
      "step": 195,
      "time_per_iteration": 2.538210868835449
    },
    {
      "auxiliary_loss_clip": 0.01383532,
      "auxiliary_loss_mlp": 0.01074209,
      "balance_loss_clip": 1.11042321,
      "balance_loss_mlp": 1.05376482,
      "epoch": 0.02356760656526183,
      "flos": 21579799697280.0,
      "grad_norm": 2.3914382843859765,
      "language_loss": 0.82955569,
      "learning_rate": 3.823708788487851e-06,
      "loss": 0.85413301,
      "num_input_tokens_seen": 4118770,
      "step": 196,
      "time_per_iteration": 2.5229756832122803
    },
    {
      "auxiliary_loss_clip": 0.01379316,
      "auxiliary_loss_mlp": 0.01082421,
      "balance_loss_clip": 1.10739207,
      "balance_loss_mlp": 1.06291866,
      "epoch": 0.02368784945590092,
      "flos": 25193598192000.0,
      "grad_norm": 2.6305134989326135,
      "language_loss": 0.84352314,
      "learning_rate": 3.827395544281781e-06,
      "loss": 0.86814046,
      "num_input_tokens_seen": 4141110,
      "step": 197,
      "time_per_iteration": 2.6391611099243164
    },
    {
      "auxiliary_loss_clip": 0.01386246,
      "auxiliary_loss_mlp": 0.01080686,
      "balance_loss_clip": 1.1109786,
      "balance_loss_mlp": 1.06018174,
      "epoch": 0.02380809234654001,
      "flos": 27562481164800.0,
      "grad_norm": 1.999182722210569,
      "language_loss": 0.78948164,
      "learning_rate": 3.831063632877802e-06,
      "loss": 0.81415093,
      "num_input_tokens_seen": 4161430,
      "step": 198,
      "time_per_iteration": 2.576688289642334
    },
    {
      "auxiliary_loss_clip": 0.01381736,
      "auxiliary_loss_mlp": 0.01072961,
      "balance_loss_clip": 1.11401629,
      "balance_loss_mlp": 1.05416203,
      "epoch": 0.0239283352371791,
      "flos": 18259786540800.0,
      "grad_norm": 2.379044868441493,
      "language_loss": 0.76057124,
      "learning_rate": 3.834713242359712e-06,
      "loss": 0.78511822,
      "num_input_tokens_seen": 4179260,
      "step": 199,
      "time_per_iteration": 2.490786552429199
    },
    {
      "auxiliary_loss_clip": 0.01384182,
      "auxiliary_loss_mlp": 0.01073777,
      "balance_loss_clip": 1.10767508,
      "balance_loss_mlp": 1.05190182,
      "epoch": 0.02404857812781819,
      "flos": 21395110942080.0,
      "grad_norm": 3.457610276393604,
      "language_loss": 0.87166214,
      "learning_rate": 3.838344557982959e-06,
      "loss": 0.89624172,
      "num_input_tokens_seen": 4200640,
      "step": 200,
      "time_per_iteration": 2.572953224182129
    },
    {
      "auxiliary_loss_clip": 0.01377653,
      "auxiliary_loss_mlp": 0.01075539,
      "balance_loss_clip": 1.10674405,
      "balance_loss_mlp": 1.05415249,
      "epoch": 0.024168821018457284,
      "flos": 16654256426880.0,
      "grad_norm": 3.3374394347500886,
      "language_loss": 0.84899002,
      "learning_rate": 3.841957762231063e-06,
      "loss": 0.87352198,
      "num_input_tokens_seen": 4218170,
      "step": 201,
      "time_per_iteration": 2.5260581970214844
    },
    {
      "auxiliary_loss_clip": 0.01374378,
      "auxiliary_loss_mlp": 0.01063737,
      "balance_loss_clip": 1.1042887,
      "balance_loss_mlp": 1.0435431,
      "epoch": 0.024289063909096374,
      "flos": 22820872464000.0,
      "grad_norm": 2.4385184604658967,
      "language_loss": 0.87626147,
      "learning_rate": 3.8455530348706454e-06,
      "loss": 0.90064269,
      "num_input_tokens_seen": 4237770,
      "step": 202,
      "time_per_iteration": 2.571443557739258
    },
    {
      "auxiliary_loss_clip": 0.01375074,
      "auxiliary_loss_mlp": 0.01073974,
      "balance_loss_clip": 1.10634232,
      "balance_loss_mlp": 1.05511534,
      "epoch": 0.024409306799735464,
      "flos": 17748598135680.0,
      "grad_norm": 2.153162568101511,
      "language_loss": 0.77302998,
      "learning_rate": 3.849130553005099e-06,
      "loss": 0.79752046,
      "num_input_tokens_seen": 4255985,
      "step": 203,
      "time_per_iteration": 2.48757266998291
    },
    {
      "auxiliary_loss_clip": 0.01375607,
      "auxiliary_loss_mlp": 0.01067165,
      "balance_loss_clip": 1.10425639,
      "balance_loss_mlp": 1.04836547,
      "epoch": 0.024529549690374557,
      "flos": 21616213109760.0,
      "grad_norm": 3.480862266978662,
      "language_loss": 0.8341502,
      "learning_rate": 3.852690491126933e-06,
      "loss": 0.85857791,
      "num_input_tokens_seen": 4276035,
      "step": 204,
      "time_per_iteration": 3.2992236614227295
    },
    {
      "auxiliary_loss_clip": 0.01370282,
      "auxiliary_loss_mlp": 0.01062416,
      "balance_loss_clip": 1.10067284,
      "balance_loss_mlp": 1.04217386,
      "epoch": 0.024649792581013647,
      "flos": 25551662918400.0,
      "grad_norm": 3.0605471564654763,
      "language_loss": 0.91415983,
      "learning_rate": 3.856233021168845e-06,
      "loss": 0.93848681,
      "num_input_tokens_seen": 4295730,
      "step": 205,
      "time_per_iteration": 3.370774984359741
    },
    {
      "auxiliary_loss_clip": 0.0136434,
      "auxiliary_loss_mlp": 0.01054383,
      "balance_loss_clip": 1.10094357,
      "balance_loss_mlp": 1.03669214,
      "epoch": 0.024770035471652737,
      "flos": 34495574544000.0,
      "grad_norm": 2.242911687815196,
      "language_loss": 0.91352785,
      "learning_rate": 3.859758312553544e-06,
      "loss": 0.93771505,
      "num_input_tokens_seen": 4317950,
      "step": 206,
      "time_per_iteration": 4.205277681350708
    },
    {
      "auxiliary_loss_clip": 0.01373758,
      "auxiliary_loss_mlp": 0.01069276,
      "balance_loss_clip": 1.1070987,
      "balance_loss_mlp": 1.05090547,
      "epoch": 0.02489027836229183,
      "flos": 21505428587520.0,
      "grad_norm": 2.119683256324153,
      "language_loss": 0.91625947,
      "learning_rate": 3.8632665322423735e-06,
      "loss": 0.9406898,
      "num_input_tokens_seen": 4337605,
      "step": 207,
      "time_per_iteration": 2.5087318420410156
    },
    {
      "auxiliary_loss_clip": 0.01371615,
      "auxiliary_loss_mlp": 0.01065658,
      "balance_loss_clip": 1.10397506,
      "balance_loss_mlp": 1.0465014,
      "epoch": 0.02501052125293092,
      "flos": 23219013790080.0,
      "grad_norm": 1.841662127677893,
      "language_loss": 0.86065757,
      "learning_rate": 3.866757844782762e-06,
      "loss": 0.88503033,
      "num_input_tokens_seen": 4358110,
      "step": 208,
      "time_per_iteration": 2.6086056232452393
    },
    {
      "auxiliary_loss_clip": 0.01371228,
      "auxiliary_loss_mlp": 0.01063733,
      "balance_loss_clip": 1.10509634,
      "balance_loss_mlp": 1.04489827,
      "epoch": 0.02513076414357001,
      "flos": 26388920010240.0,
      "grad_norm": 2.8526319740064374,
      "language_loss": 0.91058338,
      "learning_rate": 3.870232412354527e-06,
      "loss": 0.93493301,
      "num_input_tokens_seen": 4374955,
      "step": 209,
      "time_per_iteration": 2.523590326309204
    },
    {
      "auxiliary_loss_clip": 0.01367578,
      "auxiliary_loss_mlp": 0.01064445,
      "balance_loss_clip": 1.10237503,
      "balance_loss_mlp": 1.04549098,
      "epoch": 0.025251007034209103,
      "flos": 13590430047360.0,
      "grad_norm": 2.0385818608497526,
      "language_loss": 0.92587775,
      "learning_rate": 3.873690394815086e-06,
      "loss": 0.95019805,
      "num_input_tokens_seen": 4391535,
      "step": 210,
      "time_per_iteration": 2.495131015777588
    },
    {
      "auxiliary_loss_clip": 0.01365097,
      "auxiliary_loss_mlp": 0.01058858,
      "balance_loss_clip": 1.09899962,
      "balance_loss_mlp": 1.04014182,
      "epoch": 0.025371249924848193,
      "flos": 15049229103360.0,
      "grad_norm": 2.5820015441999047,
      "language_loss": 0.91305387,
      "learning_rate": 3.877131949743587e-06,
      "loss": 0.93729341,
      "num_input_tokens_seen": 4408400,
      "step": 211,
      "time_per_iteration": 2.478599786758423
    },
    {
      "auxiliary_loss_clip": 0.01367094,
      "auxiliary_loss_mlp": 0.01078591,
      "balance_loss_clip": 1.10253382,
      "balance_loss_mlp": 1.05948234,
      "epoch": 0.025491492815487283,
      "flos": 25553853648000.0,
      "grad_norm": 2.2464687423561025,
      "language_loss": 0.77946156,
      "learning_rate": 3.880557232483993e-06,
      "loss": 0.80391848,
      "num_input_tokens_seen": 4427840,
      "step": 212,
      "time_per_iteration": 2.5540060997009277
    },
    {
      "auxiliary_loss_clip": 0.01366147,
      "auxiliary_loss_mlp": 0.01061694,
      "balance_loss_clip": 1.09939647,
      "balance_loss_mlp": 1.04216766,
      "epoch": 0.025611735706126376,
      "flos": 20630752502400.0,
      "grad_norm": 2.013256768384452,
      "language_loss": 0.86888659,
      "learning_rate": 3.883966396187164e-06,
      "loss": 0.89316499,
      "num_input_tokens_seen": 4447110,
      "step": 213,
      "time_per_iteration": 2.5085840225219727
    },
    {
      "auxiliary_loss_clip": 0.0136812,
      "auxiliary_loss_mlp": 0.01060864,
      "balance_loss_clip": 1.10369563,
      "balance_loss_mlp": 1.04280365,
      "epoch": 0.025731978596765466,
      "flos": 19062282245760.0,
      "grad_norm": 2.048296030302431,
      "language_loss": 0.90014124,
      "learning_rate": 3.887359591851937e-06,
      "loss": 0.92443109,
      "num_input_tokens_seen": 4464715,
      "step": 214,
      "time_per_iteration": 2.543308734893799
    },
    {
      "auxiliary_loss_clip": 0.01363345,
      "auxiliary_loss_mlp": 0.01058674,
      "balance_loss_clip": 1.10138774,
      "balance_loss_mlp": 1.03925502,
      "epoch": 0.025852221487404556,
      "flos": 22163814927360.0,
      "grad_norm": 2.2926812510931076,
      "language_loss": 0.92373013,
      "learning_rate": 3.890736968365265e-06,
      "loss": 0.94795024,
      "num_input_tokens_seen": 4485030,
      "step": 215,
      "time_per_iteration": 2.5321481227874756
    },
    {
      "auxiliary_loss_clip": 0.01364322,
      "auxiliary_loss_mlp": 0.0106298,
      "balance_loss_clip": 1.10002589,
      "balance_loss_mlp": 1.04281032,
      "epoch": 0.02597246437804365,
      "flos": 26541971861760.0,
      "grad_norm": 1.9901051216940264,
      "language_loss": 0.85252273,
      "learning_rate": 3.894098672541412e-06,
      "loss": 0.87679577,
      "num_input_tokens_seen": 4505935,
      "step": 216,
      "time_per_iteration": 2.6220383644104004
    },
    {
      "auxiliary_loss_clip": 0.01364433,
      "auxiliary_loss_mlp": 0.01066895,
      "balance_loss_clip": 1.10046506,
      "balance_loss_mlp": 1.04668856,
      "epoch": 0.02609270726868274,
      "flos": 32671671696000.0,
      "grad_norm": 1.7493344947840754,
      "language_loss": 0.75335574,
      "learning_rate": 3.89744484916025e-06,
      "loss": 0.77766901,
      "num_input_tokens_seen": 4527045,
      "step": 217,
      "time_per_iteration": 2.65970516204834
    },
    {
      "auxiliary_loss_clip": 0.01366411,
      "auxiliary_loss_mlp": 0.01068807,
      "balance_loss_clip": 1.10224271,
      "balance_loss_mlp": 1.04861248,
      "epoch": 0.02621295015932183,
      "flos": 26243553669120.0,
      "grad_norm": 2.3008581604823672,
      "language_loss": 0.87159967,
      "learning_rate": 3.900775641004673e-06,
      "loss": 0.89595187,
      "num_input_tokens_seen": 4546360,
      "step": 218,
      "time_per_iteration": 2.546806573867798
    },
    {
      "auxiliary_loss_clip": 0.01371726,
      "auxiliary_loss_mlp": 0.01074983,
      "balance_loss_clip": 1.10462165,
      "balance_loss_mlp": 1.05239296,
      "epoch": 0.026333193049960922,
      "flos": 42921402353280.0,
      "grad_norm": 3.320789338224337,
      "language_loss": 0.73600757,
      "learning_rate": 3.904091188897156e-06,
      "loss": 0.76047468,
      "num_input_tokens_seen": 4565495,
      "step": 219,
      "time_per_iteration": 2.669745922088623
    },
    {
      "auxiliary_loss_clip": 0.01362539,
      "auxiliary_loss_mlp": 0.01073098,
      "balance_loss_clip": 1.10027313,
      "balance_loss_mlp": 1.05253482,
      "epoch": 0.026453435940600012,
      "flos": 17963846386560.0,
      "grad_norm": 2.020720471020208,
      "language_loss": 0.81937492,
      "learning_rate": 3.90739163173548e-06,
      "loss": 0.84373134,
      "num_input_tokens_seen": 4583330,
      "step": 220,
      "time_per_iteration": 2.5275635719299316
    },
    {
      "auxiliary_loss_clip": 0.01360372,
      "auxiliary_loss_mlp": 0.01069607,
      "balance_loss_clip": 1.09959698,
      "balance_loss_mlp": 1.05024719,
      "epoch": 0.026573678831239102,
      "flos": 18984319776000.0,
      "grad_norm": 2.9456030849177925,
      "language_loss": 0.88444632,
      "learning_rate": 3.910677106527646e-06,
      "loss": 0.90874612,
      "num_input_tokens_seen": 4600520,
      "step": 221,
      "time_per_iteration": 2.4854769706726074
    },
    {
      "auxiliary_loss_clip": 0.01358389,
      "auxiliary_loss_mlp": 0.01069227,
      "balance_loss_clip": 1.09946918,
      "balance_loss_mlp": 1.05108368,
      "epoch": 0.026693921721878195,
      "flos": 29241448634880.0,
      "grad_norm": 2.5311777765044643,
      "language_loss": 0.84321344,
      "learning_rate": 3.913947748426004e-06,
      "loss": 0.86748958,
      "num_input_tokens_seen": 4617340,
      "step": 222,
      "time_per_iteration": 2.6192891597747803
    },
    {
      "auxiliary_loss_clip": 0.01364779,
      "auxiliary_loss_mlp": 0.01070756,
      "balance_loss_clip": 1.10314584,
      "balance_loss_mlp": 1.05221844,
      "epoch": 0.026814164612517285,
      "flos": 14128083797760.0,
      "grad_norm": 2.854249974962943,
      "language_loss": 0.76695639,
      "learning_rate": 3.9172036907606136e-06,
      "loss": 0.79131174,
      "num_input_tokens_seen": 4630820,
      "step": 223,
      "time_per_iteration": 2.4835925102233887
    },
    {
      "auxiliary_loss_clip": 0.01362118,
      "auxiliary_loss_mlp": 0.01064354,
      "balance_loss_clip": 1.09909868,
      "balance_loss_mlp": 1.04522085,
      "epoch": 0.026934407503156375,
      "flos": 23511973115520.0,
      "grad_norm": 1.8045871885028086,
      "language_loss": 0.94935501,
      "learning_rate": 3.920445065071855e-06,
      "loss": 0.97361982,
      "num_input_tokens_seen": 4651985,
      "step": 224,
      "time_per_iteration": 2.5434398651123047
    },
    {
      "auxiliary_loss_clip": 0.01358694,
      "auxiliary_loss_mlp": 0.0107522,
      "balance_loss_clip": 1.09892631,
      "balance_loss_mlp": 1.05577648,
      "epoch": 0.027054650393795468,
      "flos": 28950356816640.0,
      "grad_norm": 2.3382265797310966,
      "language_loss": 0.7993409,
      "learning_rate": 3.923672001142322e-06,
      "loss": 0.82368004,
      "num_input_tokens_seen": 4672295,
      "step": 225,
      "time_per_iteration": 2.5866034030914307
    },
    {
      "auxiliary_loss_clip": 0.01355271,
      "auxiliary_loss_mlp": 0.01076068,
      "balance_loss_clip": 1.09733748,
      "balance_loss_mlp": 1.05650616,
      "epoch": 0.027174893284434558,
      "flos": 31431568596480.0,
      "grad_norm": 4.077587836933165,
      "language_loss": 0.84481573,
      "learning_rate": 3.926884627027996e-06,
      "loss": 0.86912912,
      "num_input_tokens_seen": 4696065,
      "step": 226,
      "time_per_iteration": 2.692692518234253
    },
    {
      "auxiliary_loss_clip": 0.01357076,
      "auxiliary_loss_mlp": 0.01070759,
      "balance_loss_clip": 1.09662664,
      "balance_loss_mlp": 1.05248427,
      "epoch": 0.027295136175073648,
      "flos": 22054466949120.0,
      "grad_norm": 2.107326939323988,
      "language_loss": 0.77363896,
      "learning_rate": 3.930083069088744e-06,
      "loss": 0.79791731,
      "num_input_tokens_seen": 4716065,
      "step": 227,
      "time_per_iteration": 2.503667116165161
    },
    {
      "auxiliary_loss_clip": 0.01322995,
      "auxiliary_loss_mlp": 0.01016696,
      "balance_loss_clip": 1.1227802,
      "balance_loss_mlp": 1.00658703,
      "epoch": 0.02741537906571274,
      "flos": 60800752972800.0,
      "grad_norm": 0.9749441330164644,
      "language_loss": 0.59297657,
      "learning_rate": 3.933267452018137e-06,
      "loss": 0.61637342,
      "num_input_tokens_seen": 4775860,
      "step": 228,
      "time_per_iteration": 3.094857692718506
    },
    {
      "auxiliary_loss_clip": 0.01355364,
      "auxiliary_loss_mlp": 0.01062031,
      "balance_loss_clip": 1.099015,
      "balance_loss_mlp": 1.04311216,
      "epoch": 0.02753562195635183,
      "flos": 24606278910720.0,
      "grad_norm": 2.208420289896797,
      "language_loss": 0.8430475,
      "learning_rate": 3.936437898872622e-06,
      "loss": 0.86722136,
      "num_input_tokens_seen": 4795835,
      "step": 229,
      "time_per_iteration": 2.54166841506958
    },
    {
      "auxiliary_loss_clip": 0.01357199,
      "auxiliary_loss_mlp": 0.0105513,
      "balance_loss_clip": 1.09940708,
      "balance_loss_mlp": 1.03717709,
      "epoch": 0.02765586484699092,
      "flos": 34094236907520.0,
      "grad_norm": 3.2385977808700495,
      "language_loss": 0.79844141,
      "learning_rate": 3.9395945311000525e-06,
      "loss": 0.8225646,
      "num_input_tokens_seen": 4817460,
      "step": 230,
      "time_per_iteration": 2.6339735984802246
    },
    {
      "auxiliary_loss_clip": 0.0135729,
      "auxiliary_loss_mlp": 0.01069433,
      "balance_loss_clip": 1.09914613,
      "balance_loss_mlp": 1.05052686,
      "epoch": 0.027776107737630014,
      "flos": 14829922615680.0,
      "grad_norm": 3.8376158589856786,
      "language_loss": 0.90692508,
      "learning_rate": 3.942737468567608e-06,
      "loss": 0.93119228,
      "num_input_tokens_seen": 4835475,
      "step": 231,
      "time_per_iteration": 2.497300863265991
    },
    {
      "auxiliary_loss_clip": 0.01355252,
      "auxiliary_loss_mlp": 0.01069152,
      "balance_loss_clip": 1.09883857,
      "balance_loss_mlp": 1.05079365,
      "epoch": 0.027896350628269104,
      "flos": 47920347066240.0,
      "grad_norm": 2.0804334918992344,
      "language_loss": 0.8606438,
      "learning_rate": 3.9458668295891026e-06,
      "loss": 0.88488781,
      "num_input_tokens_seen": 4857760,
      "step": 232,
      "time_per_iteration": 3.556023120880127
    },
    {
      "auxiliary_loss_clip": 0.01351136,
      "auxiliary_loss_mlp": 0.0106241,
      "balance_loss_clip": 1.09420633,
      "balance_loss_mlp": 1.04269302,
      "epoch": 0.028016593518908194,
      "flos": 21684550734720.0,
      "grad_norm": 2.3319487340446874,
      "language_loss": 0.86540693,
      "learning_rate": 3.948982730951712e-06,
      "loss": 0.88954234,
      "num_input_tokens_seen": 4875855,
      "step": 233,
      "time_per_iteration": 3.401474952697754
    },
    {
      "auxiliary_loss_clip": 0.01354923,
      "auxiliary_loss_mlp": 0.0106241,
      "balance_loss_clip": 1.09745383,
      "balance_loss_mlp": 1.04310989,
      "epoch": 0.028136836409547287,
      "flos": 18439483305600.0,
      "grad_norm": 2.467211104804349,
      "language_loss": 0.82109368,
      "learning_rate": 3.9520852879421254e-06,
      "loss": 0.84526706,
      "num_input_tokens_seen": 4893200,
      "step": 234,
      "time_per_iteration": 3.2872390747070312
    },
    {
      "auxiliary_loss_clip": 0.01349852,
      "auxiliary_loss_mlp": 0.01065635,
      "balance_loss_clip": 1.09696901,
      "balance_loss_mlp": 1.04802775,
      "epoch": 0.028257079300186377,
      "flos": 31576934937600.0,
      "grad_norm": 2.247225688848913,
      "language_loss": 0.81712079,
      "learning_rate": 3.955174614372137e-06,
      "loss": 0.84127569,
      "num_input_tokens_seen": 4912965,
      "step": 235,
      "time_per_iteration": 2.5721051692962646
    },
    {
      "auxiliary_loss_clip": 0.01352424,
      "auxiliary_loss_mlp": 0.01068071,
      "balance_loss_clip": 1.09749591,
      "balance_loss_mlp": 1.04923606,
      "epoch": 0.028377322190825467,
      "flos": 23513337832320.0,
      "grad_norm": 2.3448512452873995,
      "language_loss": 0.84337473,
      "learning_rate": 3.9582508226037045e-06,
      "loss": 0.8675797,
      "num_input_tokens_seen": 4933105,
      "step": 236,
      "time_per_iteration": 2.538154363632202
    },
    {
      "auxiliary_loss_clip": 0.01359055,
      "auxiliary_loss_mlp": 0.01070912,
      "balance_loss_clip": 1.09861803,
      "balance_loss_mlp": 1.05113506,
      "epoch": 0.02849756508146456,
      "flos": 20479604071680.0,
      "grad_norm": 2.440995852846626,
      "language_loss": 0.94367313,
      "learning_rate": 3.9613140235734636e-06,
      "loss": 0.96797276,
      "num_input_tokens_seen": 4950085,
      "step": 237,
      "time_per_iteration": 2.5157129764556885
    },
    {
      "auxiliary_loss_clip": 0.01350154,
      "auxiliary_loss_mlp": 0.01063875,
      "balance_loss_clip": 1.09532285,
      "balance_loss_mlp": 1.04457521,
      "epoch": 0.02861780797210365,
      "flos": 14283362292480.0,
      "grad_norm": 1.992663850482671,
      "language_loss": 0.80907106,
      "learning_rate": 3.96436432681674e-06,
      "loss": 0.83321142,
      "num_input_tokens_seen": 4968075,
      "step": 238,
      "time_per_iteration": 2.4828989505767822
    },
    {
      "auxiliary_loss_clip": 0.01349834,
      "auxiliary_loss_mlp": 0.01069475,
      "balance_loss_clip": 1.09542072,
      "balance_loss_mlp": 1.05052102,
      "epoch": 0.02873805086274274,
      "flos": 25808532053760.0,
      "grad_norm": 1.9908416034147012,
      "language_loss": 0.88966727,
      "learning_rate": 3.967401840491044e-06,
      "loss": 0.91386038,
      "num_input_tokens_seen": 4987355,
      "step": 239,
      "time_per_iteration": 2.5395402908325195
    },
    {
      "auxiliary_loss_clip": 0.01347916,
      "auxiliary_loss_mlp": 0.01063631,
      "balance_loss_clip": 1.09742951,
      "balance_loss_mlp": 1.04706097,
      "epoch": 0.028858293753381833,
      "flos": 17304238984320.0,
      "grad_norm": 2.672957067561469,
      "language_loss": 0.87890118,
      "learning_rate": 3.97042667139909e-06,
      "loss": 0.90301669,
      "num_input_tokens_seen": 5004680,
      "step": 240,
      "time_per_iteration": 2.4862513542175293
    },
    {
      "auxiliary_loss_clip": 0.01349318,
      "auxiliary_loss_mlp": 0.010588,
      "balance_loss_clip": 1.09697223,
      "balance_loss_mlp": 1.0407877,
      "epoch": 0.028978536644020923,
      "flos": 23038347358080.0,
      "grad_norm": 2.174193489347373,
      "language_loss": 0.87252986,
      "learning_rate": 3.973438925011327e-06,
      "loss": 0.89661103,
      "num_input_tokens_seen": 5022965,
      "step": 241,
      "time_per_iteration": 2.523435115814209
    },
    {
      "auxiliary_loss_clip": 0.01348963,
      "auxiliary_loss_mlp": 0.01051689,
      "balance_loss_clip": 1.09396458,
      "balance_loss_mlp": 1.03308034,
      "epoch": 0.029098779534660012,
      "flos": 28329712692480.0,
      "grad_norm": 9.429468356383522,
      "language_loss": 0.91451728,
      "learning_rate": 3.976438705488002e-06,
      "loss": 0.93852377,
      "num_input_tokens_seen": 5042625,
      "step": 242,
      "time_per_iteration": 2.5604324340820312
    },
    {
      "auxiliary_loss_clip": 0.01346862,
      "auxiliary_loss_mlp": 0.01061335,
      "balance_loss_clip": 1.09626973,
      "balance_loss_mlp": 1.04419231,
      "epoch": 0.029219022425299106,
      "flos": 13881665520000.0,
      "grad_norm": 8.943295230343054,
      "language_loss": 0.93283665,
      "learning_rate": 3.9794261157007744e-06,
      "loss": 0.9569186,
      "num_input_tokens_seen": 5060380,
      "step": 243,
      "time_per_iteration": 2.5408549308776855
    },
    {
      "auxiliary_loss_clip": 0.01352182,
      "auxiliary_loss_mlp": 0.01058749,
      "balance_loss_clip": 1.0975039,
      "balance_loss_mlp": 1.03918707,
      "epoch": 0.029339265315938196,
      "flos": 19422501788160.0,
      "grad_norm": 2.3217988013801363,
      "language_loss": 0.85028291,
      "learning_rate": 3.982401257253887e-06,
      "loss": 0.87439227,
      "num_input_tokens_seen": 5078720,
      "step": 244,
      "time_per_iteration": 2.4937031269073486
    },
    {
      "auxiliary_loss_clip": 0.01348208,
      "auxiliary_loss_mlp": 0.0105766,
      "balance_loss_clip": 1.0945996,
      "balance_loss_mlp": 1.04029155,
      "epoch": 0.029459508206577285,
      "flos": 15669550005120.0,
      "grad_norm": 2.3041590113838204,
      "language_loss": 0.89937472,
      "learning_rate": 3.985364230504893e-06,
      "loss": 0.92343342,
      "num_input_tokens_seen": 5096605,
      "step": 245,
      "time_per_iteration": 2.4814274311065674
    },
    {
      "auxiliary_loss_clip": 0.01354748,
      "auxiliary_loss_mlp": 0.01061326,
      "balance_loss_clip": 1.10030985,
      "balance_loss_mlp": 1.04437447,
      "epoch": 0.02957975109721638,
      "flos": 28220975245440.0,
      "grad_norm": 2.013585364659552,
      "language_loss": 0.84621513,
      "learning_rate": 3.988315134584976e-06,
      "loss": 0.87037593,
      "num_input_tokens_seen": 5116285,
      "step": 246,
      "time_per_iteration": 2.5597591400146484
    },
    {
      "auxiliary_loss_clip": 0.01352394,
      "auxiliary_loss_mlp": 0.01068742,
      "balance_loss_clip": 1.0980742,
      "balance_loss_mlp": 1.05041969,
      "epoch": 0.02969999398785547,
      "flos": 24315869450880.0,
      "grad_norm": 1.909379917416462,
      "language_loss": 0.80429095,
      "learning_rate": 3.991254067418851e-06,
      "loss": 0.8285023,
      "num_input_tokens_seen": 5136825,
      "step": 247,
      "time_per_iteration": 2.5331099033355713
    },
    {
      "auxiliary_loss_clip": 0.01342871,
      "auxiliary_loss_mlp": 0.01067055,
      "balance_loss_clip": 1.09528852,
      "balance_loss_mlp": 1.04994798,
      "epoch": 0.02982023687849456,
      "flos": 35078584193280.0,
      "grad_norm": 2.4163010074261524,
      "language_loss": 0.83139467,
      "learning_rate": 3.994181125744254e-06,
      "loss": 0.8554939,
      "num_input_tokens_seen": 5158630,
      "step": 248,
      "time_per_iteration": 2.6819405555725098
    },
    {
      "auxiliary_loss_clip": 0.01346289,
      "auxiliary_loss_mlp": 0.01057225,
      "balance_loss_clip": 1.09550762,
      "balance_loss_mlp": 1.04008269,
      "epoch": 0.02994047976913365,
      "flos": 26177155378560.0,
      "grad_norm": 2.3555214822289794,
      "language_loss": 0.74047112,
      "learning_rate": 3.99709640513106e-06,
      "loss": 0.76450622,
      "num_input_tokens_seen": 5179510,
      "step": 249,
      "time_per_iteration": 2.6884233951568604
    },
    {
      "auxiliary_loss_clip": 0.01347535,
      "auxiliary_loss_mlp": 0.01071898,
      "balance_loss_clip": 1.09291124,
      "balance_loss_mlp": 1.05293226,
      "epoch": 0.03006072265977274,
      "flos": 25625028447360.0,
      "grad_norm": 2.3579285643860426,
      "language_loss": 0.85599041,
      "learning_rate": 4e-06,
      "loss": 0.88018471,
      "num_input_tokens_seen": 5199345,
      "step": 250,
      "time_per_iteration": 2.6508238315582275
    },
    {
      "auxiliary_loss_clip": 0.01348462,
      "auxiliary_loss_mlp": 0.01056314,
      "balance_loss_clip": 1.09756136,
      "balance_loss_mlp": 1.03913593,
      "epoch": 0.03018096555041183,
      "flos": 22127078292480.0,
      "grad_norm": 6.856721845759342,
      "language_loss": 0.88815314,
      "learning_rate": 3.999999848300794e-06,
      "loss": 0.91220093,
      "num_input_tokens_seen": 5218330,
      "step": 251,
      "time_per_iteration": 2.561594009399414
    },
    {
      "auxiliary_loss_clip": 0.01340508,
      "auxiliary_loss_mlp": 0.0105576,
      "balance_loss_clip": 1.09116793,
      "balance_loss_mlp": 1.03841567,
      "epoch": 0.030301208441050925,
      "flos": 30188197359360.0,
      "grad_norm": 1.6813670069668534,
      "language_loss": 0.89249742,
      "learning_rate": 3.999999393203203e-06,
      "loss": 0.91646004,
      "num_input_tokens_seen": 5240740,
      "step": 252,
      "time_per_iteration": 2.600980520248413
    },
    {
      "auxiliary_loss_clip": 0.01340194,
      "auxiliary_loss_mlp": 0.01058007,
      "balance_loss_clip": 1.08968043,
      "balance_loss_mlp": 1.04090023,
      "epoch": 0.030421451331690014,
      "flos": 23621392920960.0,
      "grad_norm": 2.0489429878677483,
      "language_loss": 0.85192108,
      "learning_rate": 3.999998634707293e-06,
      "loss": 0.87590307,
      "num_input_tokens_seen": 5260290,
      "step": 253,
      "time_per_iteration": 2.5189058780670166
    },
    {
      "auxiliary_loss_clip": 0.01349752,
      "auxiliary_loss_mlp": 0.01063256,
      "balance_loss_clip": 1.09779787,
      "balance_loss_mlp": 1.04561341,
      "epoch": 0.030541694222329104,
      "flos": 27928446883200.0,
      "grad_norm": 3.2957676838366092,
      "language_loss": 0.96555686,
      "learning_rate": 3.999997572813182e-06,
      "loss": 0.98968697,
      "num_input_tokens_seen": 5278100,
      "step": 254,
      "time_per_iteration": 2.553295373916626
    },
    {
      "auxiliary_loss_clip": 0.01344141,
      "auxiliary_loss_mlp": 0.01069777,
      "balance_loss_clip": 1.09273148,
      "balance_loss_mlp": 1.05239654,
      "epoch": 0.030661937112968194,
      "flos": 18588441006720.0,
      "grad_norm": 1.8447062214383305,
      "language_loss": 0.87863976,
      "learning_rate": 3.999996207521028e-06,
      "loss": 0.90277898,
      "num_input_tokens_seen": 5296810,
      "step": 255,
      "time_per_iteration": 2.519864559173584
    },
    {
      "auxiliary_loss_clip": 0.01346642,
      "auxiliary_loss_mlp": 0.01058714,
      "balance_loss_clip": 1.09154677,
      "balance_loss_mlp": 1.03992653,
      "epoch": 0.030782180003607287,
      "flos": 12969139478400.0,
      "grad_norm": 2.1470557493906837,
      "language_loss": 0.82066244,
      "learning_rate": 3.999994538831039e-06,
      "loss": 0.84471607,
      "num_input_tokens_seen": 5313395,
      "step": 256,
      "time_per_iteration": 2.5047779083251953
    },
    {
      "auxiliary_loss_clip": 0.013442,
      "auxiliary_loss_mlp": 0.01059467,
      "balance_loss_clip": 1.09288216,
      "balance_loss_mlp": 1.04066777,
      "epoch": 0.030902422894246377,
      "flos": 23335364920320.0,
      "grad_norm": 2.4750030227833117,
      "language_loss": 0.85999751,
      "learning_rate": 3.99999256674347e-06,
      "loss": 0.88403416,
      "num_input_tokens_seen": 5333545,
      "step": 257,
      "time_per_iteration": 2.577547788619995
    },
    {
      "auxiliary_loss_clip": 0.01287211,
      "auxiliary_loss_mlp": 0.01008034,
      "balance_loss_clip": 1.09815097,
      "balance_loss_mlp": 0.99759179,
      "epoch": 0.031022665784885467,
      "flos": 55094151438720.0,
      "grad_norm": 1.004923587208518,
      "language_loss": 0.5349049,
      "learning_rate": 3.999990291258618e-06,
      "loss": 0.55785733,
      "num_input_tokens_seen": 5392235,
      "step": 258,
      "time_per_iteration": 3.0803494453430176
    },
    {
      "auxiliary_loss_clip": 0.01343399,
      "auxiliary_loss_mlp": 0.01061069,
      "balance_loss_clip": 1.09277654,
      "balance_loss_mlp": 1.0433073,
      "epoch": 0.03114290867552456,
      "flos": 19317786664320.0,
      "grad_norm": 2.3273693251046295,
      "language_loss": 0.86488926,
      "learning_rate": 3.999987712376829e-06,
      "loss": 0.8889339,
      "num_input_tokens_seen": 5410555,
      "step": 259,
      "time_per_iteration": 2.5353946685791016
    },
    {
      "auxiliary_loss_clip": 0.01342276,
      "auxiliary_loss_mlp": 0.01060603,
      "balance_loss_clip": 1.09426641,
      "balance_loss_mlp": 1.04290009,
      "epoch": 0.031263151566163654,
      "flos": 20959442881920.0,
      "grad_norm": 2.1982399831404953,
      "language_loss": 0.82236338,
      "learning_rate": 3.999984830098494e-06,
      "loss": 0.84639215,
      "num_input_tokens_seen": 5430135,
      "step": 260,
      "time_per_iteration": 4.185324668884277
    },
    {
      "auxiliary_loss_clip": 0.01339203,
      "auxiliary_loss_mlp": 0.01063006,
      "balance_loss_clip": 1.09065187,
      "balance_loss_mlp": 1.04500532,
      "epoch": 0.03138339445680274,
      "flos": 14793006412800.0,
      "grad_norm": 2.9356722304355114,
      "language_loss": 0.98046136,
      "learning_rate": 3.999981644424051e-06,
      "loss": 1.00448346,
      "num_input_tokens_seen": 5444935,
      "step": 261,
      "time_per_iteration": 3.1868677139282227
    },
    {
      "auxiliary_loss_clip": 0.01341159,
      "auxiliary_loss_mlp": 0.01069933,
      "balance_loss_clip": 1.0940057,
      "balance_loss_mlp": 1.05084765,
      "epoch": 0.03150363734744183,
      "flos": 11655599022720.0,
      "grad_norm": 2.645400996478755,
      "language_loss": 0.86089593,
      "learning_rate": 3.999978155353982e-06,
      "loss": 0.88500684,
      "num_input_tokens_seen": 5462080,
      "step": 262,
      "time_per_iteration": 2.502997636795044
    },
    {
      "auxiliary_loss_clip": 0.01338949,
      "auxiliary_loss_mlp": 0.01063926,
      "balance_loss_clip": 1.09073329,
      "balance_loss_mlp": 1.04542446,
      "epoch": 0.03162388023808092,
      "flos": 33727732485120.0,
      "grad_norm": 2.267351285021861,
      "language_loss": 0.80341858,
      "learning_rate": 3.9999743628888186e-06,
      "loss": 0.8274473,
      "num_input_tokens_seen": 5483870,
      "step": 263,
      "time_per_iteration": 2.6104564666748047
    },
    {
      "auxiliary_loss_clip": 0.01332334,
      "auxiliary_loss_mlp": 0.01060338,
      "balance_loss_clip": 1.08722758,
      "balance_loss_mlp": 1.0426352,
      "epoch": 0.03174412312872001,
      "flos": 20810952057600.0,
      "grad_norm": 2.7662914793297864,
      "language_loss": 0.89219964,
      "learning_rate": 3.999970267029133e-06,
      "loss": 0.91612631,
      "num_input_tokens_seen": 5502830,
      "step": 264,
      "time_per_iteration": 2.5291385650634766
    },
    {
      "auxiliary_loss_clip": 0.01335072,
      "auxiliary_loss_mlp": 0.01055993,
      "balance_loss_clip": 1.0904026,
      "balance_loss_mlp": 1.03866029,
      "epoch": 0.0318643660193591,
      "flos": 23727939638400.0,
      "grad_norm": 2.4709643370415564,
      "language_loss": 0.80019975,
      "learning_rate": 3.999965867775548e-06,
      "loss": 0.82411039,
      "num_input_tokens_seen": 5523225,
      "step": 265,
      "time_per_iteration": 2.5426151752471924
    },
    {
      "auxiliary_loss_clip": 0.01337859,
      "auxiliary_loss_mlp": 0.01066422,
      "balance_loss_clip": 1.09038353,
      "balance_loss_mlp": 1.04913712,
      "epoch": 0.0319846089099982,
      "flos": 13917863450880.0,
      "grad_norm": 2.5103618653069684,
      "language_loss": 0.86971104,
      "learning_rate": 3.9999611651287315e-06,
      "loss": 0.89375389,
      "num_input_tokens_seen": 5541380,
      "step": 266,
      "time_per_iteration": 2.5159478187561035
    },
    {
      "auxiliary_loss_clip": 0.01340565,
      "auxiliary_loss_mlp": 0.01060924,
      "balance_loss_clip": 1.09252357,
      "balance_loss_mlp": 1.04366207,
      "epoch": 0.03210485180063729,
      "flos": 14753253035520.0,
      "grad_norm": 2.393723200460593,
      "language_loss": 0.78739327,
      "learning_rate": 3.999956159089396e-06,
      "loss": 0.81140816,
      "num_input_tokens_seen": 5558830,
      "step": 267,
      "time_per_iteration": 2.4616403579711914
    },
    {
      "auxiliary_loss_clip": 0.0133729,
      "auxiliary_loss_mlp": 0.01065539,
      "balance_loss_clip": 1.09130144,
      "balance_loss_mlp": 1.04777694,
      "epoch": 0.03222509469127638,
      "flos": 28913153304960.0,
      "grad_norm": 2.3866526384486466,
      "language_loss": 0.79274267,
      "learning_rate": 3.999950849658302e-06,
      "loss": 0.81677091,
      "num_input_tokens_seen": 5577750,
      "step": 268,
      "time_per_iteration": 2.5542900562286377
    },
    {
      "auxiliary_loss_clip": 0.0134391,
      "auxiliary_loss_mlp": 0.01067759,
      "balance_loss_clip": 1.09357536,
      "balance_loss_mlp": 1.05040216,
      "epoch": 0.03234533758191547,
      "flos": 16946389739520.0,
      "grad_norm": 2.16188505563306,
      "language_loss": 0.841959,
      "learning_rate": 3.999945236836254e-06,
      "loss": 0.86607569,
      "num_input_tokens_seen": 5596715,
      "step": 269,
      "time_per_iteration": 2.50282883644104
    },
    {
      "auxiliary_loss_clip": 0.0134402,
      "auxiliary_loss_mlp": 0.01066178,
      "balance_loss_clip": 1.09516478,
      "balance_loss_mlp": 1.04737902,
      "epoch": 0.03246558047255456,
      "flos": 18989096284800.0,
      "grad_norm": 2.4772982015118155,
      "language_loss": 0.94841027,
      "learning_rate": 3.999939320624103e-06,
      "loss": 0.97251225,
      "num_input_tokens_seen": 5611865,
      "step": 270,
      "time_per_iteration": 2.509799003601074
    },
    {
      "auxiliary_loss_clip": 0.01341445,
      "auxiliary_loss_mlp": 0.0106344,
      "balance_loss_clip": 1.09389699,
      "balance_loss_mlp": 1.04583311,
      "epoch": 0.03258582336319365,
      "flos": 23728334688000.0,
      "grad_norm": 1.9824435272516618,
      "language_loss": 0.90065706,
      "learning_rate": 3.999933101022749e-06,
      "loss": 0.92470586,
      "num_input_tokens_seen": 5632270,
      "step": 271,
      "time_per_iteration": 2.596348285675049
    },
    {
      "auxiliary_loss_clip": 0.01337975,
      "auxiliary_loss_mlp": 0.01065526,
      "balance_loss_clip": 1.09219527,
      "balance_loss_mlp": 1.04808581,
      "epoch": 0.032706066253832745,
      "flos": 27670823562240.0,
      "grad_norm": 2.134231135976669,
      "language_loss": 0.86807013,
      "learning_rate": 3.999926578033132e-06,
      "loss": 0.8921051,
      "num_input_tokens_seen": 5652085,
      "step": 272,
      "time_per_iteration": 2.589371919631958
    },
    {
      "auxiliary_loss_clip": 0.01337107,
      "auxiliary_loss_mlp": 0.01063806,
      "balance_loss_clip": 1.08852267,
      "balance_loss_mlp": 1.04598451,
      "epoch": 0.032826309144471835,
      "flos": 45624685968000.0,
      "grad_norm": 1.9736985798727305,
      "language_loss": 0.63150817,
      "learning_rate": 3.999919751656244e-06,
      "loss": 0.65551728,
      "num_input_tokens_seen": 5678985,
      "step": 273,
      "time_per_iteration": 2.71744441986084
    },
    {
      "auxiliary_loss_clip": 0.01333368,
      "auxiliary_loss_mlp": 0.01057577,
      "balance_loss_clip": 1.08777916,
      "balance_loss_mlp": 1.03900385,
      "epoch": 0.032946552035110925,
      "flos": 25812374808960.0,
      "grad_norm": 2.8679144539871495,
      "language_loss": 0.75774014,
      "learning_rate": 3.9999126218931195e-06,
      "loss": 0.78164953,
      "num_input_tokens_seen": 5697020,
      "step": 274,
      "time_per_iteration": 2.5234222412109375
    },
    {
      "auxiliary_loss_clip": 0.01340223,
      "auxiliary_loss_mlp": 0.01052168,
      "balance_loss_clip": 1.09333205,
      "balance_loss_mlp": 1.03462029,
      "epoch": 0.033066794925750015,
      "flos": 15121984101120.0,
      "grad_norm": 2.1604332970218394,
      "language_loss": 0.89828002,
      "learning_rate": 3.99990518874484e-06,
      "loss": 0.92220396,
      "num_input_tokens_seen": 5713460,
      "step": 275,
      "time_per_iteration": 2.514495849609375
    },
    {
      "auxiliary_loss_clip": 0.01337893,
      "auxiliary_loss_mlp": 0.01068565,
      "balance_loss_clip": 1.09233379,
      "balance_loss_mlp": 1.05123186,
      "epoch": 0.033187037816389105,
      "flos": 22776593973120.0,
      "grad_norm": 2.295405872491499,
      "language_loss": 0.92337883,
      "learning_rate": 3.999897452212534e-06,
      "loss": 0.94744337,
      "num_input_tokens_seen": 5730790,
      "step": 276,
      "time_per_iteration": 2.5258681774139404
    },
    {
      "auxiliary_loss_clip": 0.01333459,
      "auxiliary_loss_mlp": 0.01064446,
      "balance_loss_clip": 1.0898875,
      "balance_loss_mlp": 1.0465169,
      "epoch": 0.033307280707028195,
      "flos": 23331414424320.0,
      "grad_norm": 2.170279915796394,
      "language_loss": 1.00139761,
      "learning_rate": 3.999889412297374e-06,
      "loss": 1.02537668,
      "num_input_tokens_seen": 5750215,
      "step": 277,
      "time_per_iteration": 2.5562474727630615
    },
    {
      "auxiliary_loss_clip": 0.01333571,
      "auxiliary_loss_mlp": 0.01043413,
      "balance_loss_clip": 1.08807111,
      "balance_loss_mlp": 1.02710533,
      "epoch": 0.03342752359766729,
      "flos": 28840290566400.0,
      "grad_norm": 2.066377876317811,
      "language_loss": 0.79029882,
      "learning_rate": 3.999881069000581e-06,
      "loss": 0.81406868,
      "num_input_tokens_seen": 5769945,
      "step": 278,
      "time_per_iteration": 2.58473539352417
    },
    {
      "auxiliary_loss_clip": 0.01335182,
      "auxiliary_loss_mlp": 0.01054719,
      "balance_loss_clip": 1.08848381,
      "balance_loss_mlp": 1.03621721,
      "epoch": 0.03354776648830638,
      "flos": 19384544090880.0,
      "grad_norm": 3.367739707248116,
      "language_loss": 0.87200749,
      "learning_rate": 3.99987242232342e-06,
      "loss": 0.89590657,
      "num_input_tokens_seen": 5784950,
      "step": 279,
      "time_per_iteration": 2.507972240447998
    },
    {
      "auxiliary_loss_clip": 0.0133659,
      "auxiliary_loss_mlp": 0.010664,
      "balance_loss_clip": 1.09167504,
      "balance_loss_mlp": 1.04811358,
      "epoch": 0.03366800937894547,
      "flos": 17858628472320.0,
      "grad_norm": 1.9479442345584634,
      "language_loss": 0.79613489,
      "learning_rate": 3.9998634722672026e-06,
      "loss": 0.8201648,
      "num_input_tokens_seen": 5805005,
      "step": 280,
      "time_per_iteration": 2.526787757873535
    },
    {
      "auxiliary_loss_clip": 0.01336722,
      "auxiliary_loss_mlp": 0.010581,
      "balance_loss_clip": 1.09226966,
      "balance_loss_mlp": 1.04107666,
      "epoch": 0.03378825226958456,
      "flos": 35951033635200.0,
      "grad_norm": 2.120507500201445,
      "language_loss": 0.78565633,
      "learning_rate": 3.999854218833286e-06,
      "loss": 0.80960453,
      "num_input_tokens_seen": 5825825,
      "step": 281,
      "time_per_iteration": 2.6457369327545166
    },
    {
      "auxiliary_loss_clip": 0.01335434,
      "auxiliary_loss_mlp": 0.01062124,
      "balance_loss_clip": 1.09196496,
      "balance_loss_mlp": 1.04435015,
      "epoch": 0.03390849516022365,
      "flos": 25702488126720.0,
      "grad_norm": 2.24343246384953,
      "language_loss": 0.81778347,
      "learning_rate": 3.999844662023075e-06,
      "loss": 0.84175909,
      "num_input_tokens_seen": 5845700,
      "step": 282,
      "time_per_iteration": 2.545872688293457
    },
    {
      "auxiliary_loss_clip": 0.0132775,
      "auxiliary_loss_mlp": 0.01058647,
      "balance_loss_clip": 1.08758342,
      "balance_loss_mlp": 1.04140937,
      "epoch": 0.03402873805086274,
      "flos": 21284505987840.0,
      "grad_norm": 1.8115762606009012,
      "language_loss": 0.92019212,
      "learning_rate": 3.999834801838018e-06,
      "loss": 0.94405603,
      "num_input_tokens_seen": 5864680,
      "step": 283,
      "time_per_iteration": 2.5256025791168213
    },
    {
      "auxiliary_loss_clip": 0.01329167,
      "auxiliary_loss_mlp": 0.01056436,
      "balance_loss_clip": 1.08850718,
      "balance_loss_mlp": 1.03924584,
      "epoch": 0.03414898094150183,
      "flos": 22710913954560.0,
      "grad_norm": 1.8912749750533082,
      "language_loss": 0.74029875,
      "learning_rate": 3.9998246382796115e-06,
      "loss": 0.76415479,
      "num_input_tokens_seen": 5884260,
      "step": 284,
      "time_per_iteration": 2.5002176761627197
    },
    {
      "auxiliary_loss_clip": 0.01333349,
      "auxiliary_loss_mlp": 0.01052163,
      "balance_loss_clip": 1.08690763,
      "balance_loss_mlp": 1.03386402,
      "epoch": 0.03426922383214093,
      "flos": 18879927874560.0,
      "grad_norm": 2.256044220588054,
      "language_loss": 0.90905637,
      "learning_rate": 3.999814171349399e-06,
      "loss": 0.93291152,
      "num_input_tokens_seen": 5902120,
      "step": 285,
      "time_per_iteration": 2.501950979232788
    },
    {
      "auxiliary_loss_clip": 0.0132887,
      "auxiliary_loss_mlp": 0.01055901,
      "balance_loss_clip": 1.08813155,
      "balance_loss_mlp": 1.03940225,
      "epoch": 0.03438946672278002,
      "flos": 34752012716160.0,
      "grad_norm": 1.6021543378440304,
      "language_loss": 0.73486495,
      "learning_rate": 3.9998034010489655e-06,
      "loss": 0.75871265,
      "num_input_tokens_seen": 5925810,
      "step": 286,
      "time_per_iteration": 2.619114875793457
    },
    {
      "auxiliary_loss_clip": 0.01328761,
      "auxiliary_loss_mlp": 0.01061034,
      "balance_loss_clip": 1.08955002,
      "balance_loss_mlp": 1.04446411,
      "epoch": 0.03450970961341911,
      "flos": 22164102236160.0,
      "grad_norm": 2.637695448636838,
      "language_loss": 0.76006019,
      "learning_rate": 3.999792327379946e-06,
      "loss": 0.78395808,
      "num_input_tokens_seen": 5945185,
      "step": 287,
      "time_per_iteration": 3.3553969860076904
    },
    {
      "auxiliary_loss_clip": 0.01334865,
      "auxiliary_loss_mlp": 0.01063927,
      "balance_loss_clip": 1.09394956,
      "balance_loss_mlp": 1.04707134,
      "epoch": 0.034629952504058197,
      "flos": 21725740656000.0,
      "grad_norm": 2.1228582628350092,
      "language_loss": 0.96314919,
      "learning_rate": 3.999780950344021e-06,
      "loss": 0.98713708,
      "num_input_tokens_seen": 5963375,
      "step": 288,
      "time_per_iteration": 3.356580972671509
    },
    {
      "auxiliary_loss_clip": 0.01336544,
      "auxiliary_loss_mlp": 0.01065429,
      "balance_loss_clip": 1.09137404,
      "balance_loss_mlp": 1.04733264,
      "epoch": 0.034750195394697286,
      "flos": 20047994248320.0,
      "grad_norm": 1.923355404044138,
      "language_loss": 0.82856649,
      "learning_rate": 3.999769269942916e-06,
      "loss": 0.85258615,
      "num_input_tokens_seen": 5983415,
      "step": 289,
      "time_per_iteration": 3.273599863052368
    },
    {
      "auxiliary_loss_clip": 0.01329649,
      "auxiliary_loss_mlp": 0.01055401,
      "balance_loss_clip": 1.08848834,
      "balance_loss_mlp": 1.03786576,
      "epoch": 0.034870438285336376,
      "flos": 27965865876480.0,
      "grad_norm": 1.7660963300444175,
      "language_loss": 0.8114922,
      "learning_rate": 3.999757286178402e-06,
      "loss": 0.83534265,
      "num_input_tokens_seen": 6005850,
      "step": 290,
      "time_per_iteration": 2.571743965148926
    },
    {
      "auxiliary_loss_clip": 0.01332818,
      "auxiliary_loss_mlp": 0.01050877,
      "balance_loss_clip": 1.09096825,
      "balance_loss_mlp": 1.03334117,
      "epoch": 0.03499068117597547,
      "flos": 22017514832640.0,
      "grad_norm": 2.817466840867797,
      "language_loss": 0.90789264,
      "learning_rate": 3.999744999052299e-06,
      "loss": 0.93172961,
      "num_input_tokens_seen": 6027240,
      "step": 291,
      "time_per_iteration": 2.5321621894836426
    },
    {
      "auxiliary_loss_clip": 0.0127556,
      "auxiliary_loss_mlp": 0.01032921,
      "balance_loss_clip": 1.09456384,
      "balance_loss_mlp": 1.02276409,
      "epoch": 0.03511092406661456,
      "flos": 57242147725440.0,
      "grad_norm": 0.9579212115780174,
      "language_loss": 0.61122328,
      "learning_rate": 3.9997324085664675e-06,
      "loss": 0.6343081,
      "num_input_tokens_seen": 6087470,
      "step": 292,
      "time_per_iteration": 3.0541954040527344
    },
    {
      "auxiliary_loss_clip": 0.0132714,
      "auxiliary_loss_mlp": 0.01057525,
      "balance_loss_clip": 1.08639526,
      "balance_loss_mlp": 1.03988171,
      "epoch": 0.03523116695725365,
      "flos": 22928065626240.0,
      "grad_norm": 2.8206649418372063,
      "language_loss": 0.92031407,
      "learning_rate": 3.999719514722821e-06,
      "loss": 0.94416064,
      "num_input_tokens_seen": 6107600,
      "step": 293,
      "time_per_iteration": 2.5209949016571045
    },
    {
      "auxiliary_loss_clip": 0.01324131,
      "auxiliary_loss_mlp": 0.0105454,
      "balance_loss_clip": 1.08604956,
      "balance_loss_mlp": 1.0381012,
      "epoch": 0.03535140984789274,
      "flos": 36903241226880.0,
      "grad_norm": 2.9812749147255326,
      "language_loss": 0.74878967,
      "learning_rate": 3.999706317523314e-06,
      "loss": 0.77257627,
      "num_input_tokens_seen": 6126160,
      "step": 294,
      "time_per_iteration": 2.658825635910034
    },
    {
      "auxiliary_loss_clip": 0.01325559,
      "auxiliary_loss_mlp": 0.01051473,
      "balance_loss_clip": 1.08720243,
      "balance_loss_mlp": 1.03529596,
      "epoch": 0.03547165273853183,
      "flos": 20449152316800.0,
      "grad_norm": 2.0086204547311945,
      "language_loss": 0.85961372,
      "learning_rate": 3.999692816969948e-06,
      "loss": 0.88338405,
      "num_input_tokens_seen": 6145695,
      "step": 295,
      "time_per_iteration": 2.555886745452881
    },
    {
      "auxiliary_loss_clip": 0.01264517,
      "auxiliary_loss_mlp": 0.01014689,
      "balance_loss_clip": 1.08577347,
      "balance_loss_mlp": 1.00429428,
      "epoch": 0.03559189562917092,
      "flos": 69850564871040.0,
      "grad_norm": 0.9962205863226998,
      "language_loss": 0.69369656,
      "learning_rate": 3.999679013064772e-06,
      "loss": 0.71648866,
      "num_input_tokens_seen": 6212440,
      "step": 296,
      "time_per_iteration": 3.1484200954437256
    },
    {
      "auxiliary_loss_clip": 0.01328386,
      "auxiliary_loss_mlp": 0.01055972,
      "balance_loss_clip": 1.08866513,
      "balance_loss_mlp": 1.03918695,
      "epoch": 0.03571213851981002,
      "flos": 21651944163840.0,
      "grad_norm": 2.7196133241091807,
      "language_loss": 0.85931814,
      "learning_rate": 3.99966490580988e-06,
      "loss": 0.88316172,
      "num_input_tokens_seen": 6229800,
      "step": 297,
      "time_per_iteration": 2.5340359210968018
    },
    {
      "auxiliary_loss_clip": 0.01329534,
      "auxiliary_loss_mlp": 0.01061466,
      "balance_loss_clip": 1.0877459,
      "balance_loss_mlp": 1.04449058,
      "epoch": 0.03583238141044911,
      "flos": 43945610757120.0,
      "grad_norm": 2.2396501534292472,
      "language_loss": 0.65846479,
      "learning_rate": 3.999650495207411e-06,
      "loss": 0.68237484,
      "num_input_tokens_seen": 6255825,
      "step": 298,
      "time_per_iteration": 2.708862543106079
    },
    {
      "auxiliary_loss_clip": 0.01322169,
      "auxiliary_loss_mlp": 0.01058964,
      "balance_loss_clip": 1.08597827,
      "balance_loss_mlp": 1.0416193,
      "epoch": 0.0359526243010882,
      "flos": 18910810592640.0,
      "grad_norm": 3.2348747223661514,
      "language_loss": 0.90328115,
      "learning_rate": 3.999635781259553e-06,
      "loss": 0.92709249,
      "num_input_tokens_seen": 6271090,
      "step": 299,
      "time_per_iteration": 2.509911298751831
    },
    {
      "auxiliary_loss_clip": 0.0124728,
      "auxiliary_loss_mlp": 0.01009398,
      "balance_loss_clip": 1.07158875,
      "balance_loss_mlp": 0.99881184,
      "epoch": 0.03607286719172729,
      "flos": 61668892782720.0,
      "grad_norm": 0.9174435658753497,
      "language_loss": 0.5222618,
      "learning_rate": 3.999620763968535e-06,
      "loss": 0.54482859,
      "num_input_tokens_seen": 6329965,
      "step": 300,
      "time_per_iteration": 2.985281467437744
    },
    {
      "auxiliary_loss_clip": 0.01322019,
      "auxiliary_loss_mlp": 0.01054684,
      "balance_loss_clip": 1.08697486,
      "balance_loss_mlp": 1.0374819,
      "epoch": 0.03619311008236638,
      "flos": 27819062991360.0,
      "grad_norm": 1.6924198064025273,
      "language_loss": 0.86637765,
      "learning_rate": 3.999605443336638e-06,
      "loss": 0.89014471,
      "num_input_tokens_seen": 6352095,
      "step": 301,
      "time_per_iteration": 2.6531150341033936
    },
    {
      "auxiliary_loss_clip": 0.01328685,
      "auxiliary_loss_mlp": 0.01060367,
      "balance_loss_clip": 1.08937466,
      "balance_loss_mlp": 1.04327273,
      "epoch": 0.03631335297300547,
      "flos": 13621133197440.0,
      "grad_norm": 2.5111007791976414,
      "language_loss": 0.8970139,
      "learning_rate": 3.999589819366185e-06,
      "loss": 0.9209044,
      "num_input_tokens_seen": 6365885,
      "step": 302,
      "time_per_iteration": 2.500898599624634
    },
    {
      "auxiliary_loss_clip": 0.01328616,
      "auxiliary_loss_mlp": 0.01056838,
      "balance_loss_clip": 1.08870649,
      "balance_loss_mlp": 1.03904057,
      "epoch": 0.036433595863644565,
      "flos": 27631788456960.0,
      "grad_norm": 2.09520069352677,
      "language_loss": 0.84832335,
      "learning_rate": 3.999573892059547e-06,
      "loss": 0.87217796,
      "num_input_tokens_seen": 6385015,
      "step": 303,
      "time_per_iteration": 2.5832653045654297
    },
    {
      "auxiliary_loss_clip": 0.01331332,
      "auxiliary_loss_mlp": 0.0106247,
      "balance_loss_clip": 1.08946693,
      "balance_loss_mlp": 1.04401696,
      "epoch": 0.036553838754283655,
      "flos": 24572020314240.0,
      "grad_norm": 2.161686598713949,
      "language_loss": 0.81187534,
      "learning_rate": 3.999557661419138e-06,
      "loss": 0.8358134,
      "num_input_tokens_seen": 6405165,
      "step": 304,
      "time_per_iteration": 2.521466016769409
    },
    {
      "auxiliary_loss_clip": 0.01329276,
      "auxiliary_loss_mlp": 0.01057483,
      "balance_loss_clip": 1.09052372,
      "balance_loss_mlp": 1.04122305,
      "epoch": 0.036674081644922744,
      "flos": 23404313076480.0,
      "grad_norm": 1.9459934439852062,
      "language_loss": 0.81442571,
      "learning_rate": 3.9995411274474225e-06,
      "loss": 0.83829331,
      "num_input_tokens_seen": 6424445,
      "step": 305,
      "time_per_iteration": 2.5334625244140625
    },
    {
      "auxiliary_loss_clip": 0.01327331,
      "auxiliary_loss_mlp": 0.01065835,
      "balance_loss_clip": 1.08737397,
      "balance_loss_mlp": 1.047858,
      "epoch": 0.036794324535561834,
      "flos": 27489690253440.0,
      "grad_norm": 2.0954074933042506,
      "language_loss": 0.81599385,
      "learning_rate": 3.999524290146908e-06,
      "loss": 0.83992547,
      "num_input_tokens_seen": 6444650,
      "step": 306,
      "time_per_iteration": 2.6260616779327393
    },
    {
      "auxiliary_loss_clip": 0.01325214,
      "auxiliary_loss_mlp": 0.01066193,
      "balance_loss_clip": 1.08915496,
      "balance_loss_mlp": 1.04903829,
      "epoch": 0.036914567426200924,
      "flos": 19463476227840.0,
      "grad_norm": 2.635763558910293,
      "language_loss": 0.92800236,
      "learning_rate": 3.9995071495201485e-06,
      "loss": 0.95191634,
      "num_input_tokens_seen": 6461755,
      "step": 307,
      "time_per_iteration": 2.5212206840515137
    },
    {
      "auxiliary_loss_clip": 0.01325563,
      "auxiliary_loss_mlp": 0.01061206,
      "balance_loss_clip": 1.08881021,
      "balance_loss_mlp": 1.04326475,
      "epoch": 0.037034810316840014,
      "flos": 22309324922880.0,
      "grad_norm": 2.9360639537587425,
      "language_loss": 0.97812188,
      "learning_rate": 3.999489705569744e-06,
      "loss": 1.0019896,
      "num_input_tokens_seen": 6479455,
      "step": 308,
      "time_per_iteration": 2.5483994483947754
    },
    {
      "auxiliary_loss_clip": 0.0132289,
      "auxiliary_loss_mlp": 0.01058299,
      "balance_loss_clip": 1.08516765,
      "balance_loss_mlp": 1.04157436,
      "epoch": 0.03715505320747911,
      "flos": 18588333265920.0,
      "grad_norm": 2.6182183455842285,
      "language_loss": 0.86391532,
      "learning_rate": 3.999471958298341e-06,
      "loss": 0.88772726,
      "num_input_tokens_seen": 6498365,
      "step": 309,
      "time_per_iteration": 2.507850408554077
    },
    {
      "auxiliary_loss_clip": 0.01329665,
      "auxiliary_loss_mlp": 0.01067838,
      "balance_loss_clip": 1.09073079,
      "balance_loss_mlp": 1.04976583,
      "epoch": 0.0372752960981182,
      "flos": 35955343267200.0,
      "grad_norm": 2.0334745302381405,
      "language_loss": 0.76096392,
      "learning_rate": 3.999453907708631e-06,
      "loss": 0.78493893,
      "num_input_tokens_seen": 6520770,
      "step": 310,
      "time_per_iteration": 2.6196281909942627
    },
    {
      "auxiliary_loss_clip": 0.01326735,
      "auxiliary_loss_mlp": 0.01053964,
      "balance_loss_clip": 1.08909392,
      "balance_loss_mlp": 1.03751278,
      "epoch": 0.03739553898875729,
      "flos": 20814040627200.0,
      "grad_norm": 1.7142251968670252,
      "language_loss": 0.81299865,
      "learning_rate": 3.999435553803353e-06,
      "loss": 0.83680564,
      "num_input_tokens_seen": 6540170,
      "step": 311,
      "time_per_iteration": 2.539167642593384
    },
    {
      "auxiliary_loss_clip": 0.01323714,
      "auxiliary_loss_mlp": 0.01061315,
      "balance_loss_clip": 1.0877887,
      "balance_loss_mlp": 1.04450667,
      "epoch": 0.03751578187939638,
      "flos": 20264140339200.0,
      "grad_norm": 2.3178425497429402,
      "language_loss": 0.8306613,
      "learning_rate": 3.999416896585292e-06,
      "loss": 0.85451156,
      "num_input_tokens_seen": 6557200,
      "step": 312,
      "time_per_iteration": 2.4963958263397217
    },
    {
      "auxiliary_loss_clip": 0.01325557,
      "auxiliary_loss_mlp": 0.01055698,
      "balance_loss_clip": 1.08774781,
      "balance_loss_mlp": 1.0384481,
      "epoch": 0.03763602477003547,
      "flos": 20668063754880.0,
      "grad_norm": 7.972621094044825,
      "language_loss": 0.85312146,
      "learning_rate": 3.9993979360572775e-06,
      "loss": 0.87693405,
      "num_input_tokens_seen": 6577340,
      "step": 313,
      "time_per_iteration": 3.361898422241211
    },
    {
      "auxiliary_loss_clip": 0.01332534,
      "auxiliary_loss_mlp": 0.0106335,
      "balance_loss_clip": 1.09138894,
      "balance_loss_mlp": 1.04605293,
      "epoch": 0.03775626766067456,
      "flos": 16691352197760.0,
      "grad_norm": 2.9572263982059153,
      "language_loss": 0.8326993,
      "learning_rate": 3.999378672222185e-06,
      "loss": 0.8566581,
      "num_input_tokens_seen": 6595125,
      "step": 314,
      "time_per_iteration": 2.488062620162964
    },
    {
      "auxiliary_loss_clip": 0.01327943,
      "auxiliary_loss_mlp": 0.01056861,
      "balance_loss_clip": 1.09060633,
      "balance_loss_mlp": 1.03814507,
      "epoch": 0.03787651055131366,
      "flos": 21141797253120.0,
      "grad_norm": 2.348276936918884,
      "language_loss": 0.82739699,
      "learning_rate": 3.9993591050829385e-06,
      "loss": 0.85124505,
      "num_input_tokens_seen": 6612990,
      "step": 315,
      "time_per_iteration": 4.209430932998657
    },
    {
      "auxiliary_loss_clip": 0.01327291,
      "auxiliary_loss_mlp": 0.01065795,
      "balance_loss_clip": 1.09015453,
      "balance_loss_mlp": 1.04814053,
      "epoch": 0.037996753441952746,
      "flos": 22018089450240.0,
      "grad_norm": 2.3789878680614844,
      "language_loss": 0.79256058,
      "learning_rate": 3.999339234642506e-06,
      "loss": 0.81649148,
      "num_input_tokens_seen": 6632740,
      "step": 316,
      "time_per_iteration": 2.517179489135742
    },
    {
      "auxiliary_loss_clip": 0.01328657,
      "auxiliary_loss_mlp": 0.01050433,
      "balance_loss_clip": 1.09153104,
      "balance_loss_mlp": 1.03221774,
      "epoch": 0.038116996332591836,
      "flos": 27709391790720.0,
      "grad_norm": 3.3013038311176386,
      "language_loss": 0.83545929,
      "learning_rate": 3.9993190609038994e-06,
      "loss": 0.85925019,
      "num_input_tokens_seen": 6651505,
      "step": 317,
      "time_per_iteration": 3.337282419204712
    },
    {
      "auxiliary_loss_clip": 0.01319379,
      "auxiliary_loss_mlp": 0.01049781,
      "balance_loss_clip": 1.08588779,
      "balance_loss_mlp": 1.03243613,
      "epoch": 0.038237239223230926,
      "flos": 21178067011200.0,
      "grad_norm": 2.0830086482639034,
      "language_loss": 0.833009,
      "learning_rate": 3.999298583870182e-06,
      "loss": 0.85670054,
      "num_input_tokens_seen": 6671090,
      "step": 318,
      "time_per_iteration": 2.5150322914123535
    },
    {
      "auxiliary_loss_clip": 0.01322628,
      "auxiliary_loss_mlp": 0.01057801,
      "balance_loss_clip": 1.08708167,
      "balance_loss_mlp": 1.04056382,
      "epoch": 0.038357482113870016,
      "flos": 25556618995200.0,
      "grad_norm": 2.7204237437115717,
      "language_loss": 0.77655065,
      "learning_rate": 3.999277803544458e-06,
      "loss": 0.80035496,
      "num_input_tokens_seen": 6691245,
      "step": 319,
      "time_per_iteration": 2.5506980419158936
    },
    {
      "auxiliary_loss_clip": 0.01230239,
      "auxiliary_loss_mlp": 0.01019467,
      "balance_loss_clip": 1.0673337,
      "balance_loss_mlp": 1.01012063,
      "epoch": 0.038477725004509106,
      "flos": 59227578034560.0,
      "grad_norm": 0.9630927809641395,
      "language_loss": 0.6238597,
      "learning_rate": 3.999256719929882e-06,
      "loss": 0.64635676,
      "num_input_tokens_seen": 6752520,
      "step": 320,
      "time_per_iteration": 3.092270612716675
    },
    {
      "auxiliary_loss_clip": 0.01228995,
      "auxiliary_loss_mlp": 0.01011204,
      "balance_loss_clip": 1.06638336,
      "balance_loss_mlp": 1.00195336,
      "epoch": 0.0385979678951482,
      "flos": 67317676398720.0,
      "grad_norm": 1.2175475917935272,
      "language_loss": 0.67108715,
      "learning_rate": 3.999235333029651e-06,
      "loss": 0.69348913,
      "num_input_tokens_seen": 6806460,
      "step": 321,
      "time_per_iteration": 2.9851176738739014
    },
    {
      "auxiliary_loss_clip": 0.01322092,
      "auxiliary_loss_mlp": 0.01058274,
      "balance_loss_clip": 1.08937657,
      "balance_loss_mlp": 1.04152548,
      "epoch": 0.03871821078578729,
      "flos": 22746752749440.0,
      "grad_norm": 1.8569989273025194,
      "language_loss": 0.81926775,
      "learning_rate": 3.999213642847009e-06,
      "loss": 0.84307146,
      "num_input_tokens_seen": 6827045,
      "step": 322,
      "time_per_iteration": 2.5512428283691406
    },
    {
      "auxiliary_loss_clip": 0.0132256,
      "auxiliary_loss_mlp": 0.01057646,
      "balance_loss_clip": 1.08720958,
      "balance_loss_mlp": 1.04084945,
      "epoch": 0.03883845367642638,
      "flos": 26280613526400.0,
      "grad_norm": 1.7577695843165855,
      "language_loss": 0.91229081,
      "learning_rate": 3.999191649385247e-06,
      "loss": 0.93609285,
      "num_input_tokens_seen": 6848220,
      "step": 323,
      "time_per_iteration": 2.6603574752807617
    },
    {
      "auxiliary_loss_clip": 0.0122134,
      "auxiliary_loss_mlp": 0.01009497,
      "balance_loss_clip": 1.06197786,
      "balance_loss_mlp": 1.00077116,
      "epoch": 0.03895869656706547,
      "flos": 56962835568000.0,
      "grad_norm": 0.906988018940969,
      "language_loss": 0.59800816,
      "learning_rate": 3.999169352647702e-06,
      "loss": 0.62031662,
      "num_input_tokens_seen": 6909400,
      "step": 324,
      "time_per_iteration": 2.9916763305664062
    },
    {
      "auxiliary_loss_clip": 0.01323623,
      "auxiliary_loss_mlp": 0.01081837,
      "balance_loss_clip": 1.08758521,
      "balance_loss_mlp": 1.06322801,
      "epoch": 0.03907893945770456,
      "flos": 24863363527680.0,
      "grad_norm": 1.8507078007250286,
      "language_loss": 0.83109725,
      "learning_rate": 3.999146752637755e-06,
      "loss": 0.85515183,
      "num_input_tokens_seen": 6930445,
      "step": 325,
      "time_per_iteration": 2.549497604370117
    },
    {
      "auxiliary_loss_clip": 0.01322288,
      "auxiliary_loss_mlp": 0.01060797,
      "balance_loss_clip": 1.08705592,
      "balance_loss_mlp": 1.04311848,
      "epoch": 0.03919918234834365,
      "flos": 18368595815040.0,
      "grad_norm": 2.8756009460010863,
      "language_loss": 0.89516294,
      "learning_rate": 3.999123849358836e-06,
      "loss": 0.91899383,
      "num_input_tokens_seen": 6948110,
      "step": 326,
      "time_per_iteration": 2.487290382385254
    },
    {
      "auxiliary_loss_clip": 0.01322495,
      "auxiliary_loss_mlp": 0.01065955,
      "balance_loss_clip": 1.08741999,
      "balance_loss_mlp": 1.04758453,
      "epoch": 0.03931942523898275,
      "flos": 25225414663680.0,
      "grad_norm": 2.2656996093550297,
      "language_loss": 0.74592853,
      "learning_rate": 3.999100642814418e-06,
      "loss": 0.76981294,
      "num_input_tokens_seen": 6968550,
      "step": 327,
      "time_per_iteration": 2.5387938022613525
    },
    {
      "auxiliary_loss_clip": 0.01321406,
      "auxiliary_loss_mlp": 0.01061238,
      "balance_loss_clip": 1.08763158,
      "balance_loss_mlp": 1.04345202,
      "epoch": 0.03943966812962184,
      "flos": 23257905240960.0,
      "grad_norm": 2.3694148212369535,
      "language_loss": 0.88497484,
      "learning_rate": 3.999077133008022e-06,
      "loss": 0.90880126,
      "num_input_tokens_seen": 6987135,
      "step": 328,
      "time_per_iteration": 2.5143589973449707
    },
    {
      "auxiliary_loss_clip": 0.0132301,
      "auxiliary_loss_mlp": 0.01064153,
      "balance_loss_clip": 1.08795619,
      "balance_loss_mlp": 1.04462671,
      "epoch": 0.03955991102026093,
      "flos": 29168837291520.0,
      "grad_norm": 2.2706797785465076,
      "language_loss": 0.90551168,
      "learning_rate": 3.9990533199432145e-06,
      "loss": 0.92938328,
      "num_input_tokens_seen": 7008630,
      "step": 329,
      "time_per_iteration": 2.554593563079834
    },
    {
      "auxiliary_loss_clip": 0.01320879,
      "auxiliary_loss_mlp": 0.01057919,
      "balance_loss_clip": 1.08672369,
      "balance_loss_mlp": 1.04010868,
      "epoch": 0.03968015391090002,
      "flos": 17602441695360.0,
      "grad_norm": 2.5874890831835398,
      "language_loss": 0.7574122,
      "learning_rate": 3.999029203623608e-06,
      "loss": 0.78120023,
      "num_input_tokens_seen": 7026350,
      "step": 330,
      "time_per_iteration": 2.4613447189331055
    },
    {
      "auxiliary_loss_clip": 0.01317444,
      "auxiliary_loss_mlp": 0.01055333,
      "balance_loss_clip": 1.08607543,
      "balance_loss_mlp": 1.03749931,
      "epoch": 0.03980039680153911,
      "flos": 21799285752960.0,
      "grad_norm": 2.173477655795099,
      "language_loss": 0.86607158,
      "learning_rate": 3.99900478405286e-06,
      "loss": 0.88979924,
      "num_input_tokens_seen": 7045660,
      "step": 331,
      "time_per_iteration": 2.490781545639038
    },
    {
      "auxiliary_loss_clip": 0.01318404,
      "auxiliary_loss_mlp": 0.01061005,
      "balance_loss_clip": 1.08941865,
      "balance_loss_mlp": 1.04522133,
      "epoch": 0.0399206396921782,
      "flos": 15195134148480.0,
      "grad_norm": 2.5433866667122227,
      "language_loss": 0.82356119,
      "learning_rate": 3.998980061234676e-06,
      "loss": 0.84735525,
      "num_input_tokens_seen": 7063575,
      "step": 332,
      "time_per_iteration": 2.4944329261779785
    },
    {
      "auxiliary_loss_clip": 0.01325771,
      "auxiliary_loss_mlp": 0.01053483,
      "balance_loss_clip": 1.08843422,
      "balance_loss_mlp": 1.03524363,
      "epoch": 0.040040882582817294,
      "flos": 14422910630400.0,
      "grad_norm": 2.6891676369336652,
      "language_loss": 0.75439119,
      "learning_rate": 3.9989550351728055e-06,
      "loss": 0.77818376,
      "num_input_tokens_seen": 7080505,
      "step": 333,
      "time_per_iteration": 2.4763119220733643
    },
    {
      "auxiliary_loss_clip": 0.01319959,
      "auxiliary_loss_mlp": 0.01056229,
      "balance_loss_clip": 1.08849204,
      "balance_loss_mlp": 1.0393132,
      "epoch": 0.040161125473456384,
      "flos": 19280906375040.0,
      "grad_norm": 2.4471815199456306,
      "language_loss": 0.84606612,
      "learning_rate": 3.998929705871046e-06,
      "loss": 0.86982799,
      "num_input_tokens_seen": 7097860,
      "step": 334,
      "time_per_iteration": 2.4974420070648193
    },
    {
      "auxiliary_loss_clip": 0.01318843,
      "auxiliary_loss_mlp": 0.01057949,
      "balance_loss_clip": 1.08930421,
      "balance_loss_mlp": 1.04095018,
      "epoch": 0.040281368364095474,
      "flos": 17821101738240.0,
      "grad_norm": 2.692403906654916,
      "language_loss": 0.88927627,
      "learning_rate": 3.99890407333324e-06,
      "loss": 0.91304421,
      "num_input_tokens_seen": 7116390,
      "step": 335,
      "time_per_iteration": 2.4892184734344482
    },
    {
      "auxiliary_loss_clip": 0.01315432,
      "auxiliary_loss_mlp": 0.01056401,
      "balance_loss_clip": 1.08302426,
      "balance_loss_mlp": 1.03863895,
      "epoch": 0.040401611254734564,
      "flos": 19573757959680.0,
      "grad_norm": 1.9567657765697963,
      "language_loss": 0.87152272,
      "learning_rate": 3.998878137563275e-06,
      "loss": 0.89524102,
      "num_input_tokens_seen": 7135940,
      "step": 336,
      "time_per_iteration": 2.5106372833251953
    },
    {
      "auxiliary_loss_clip": 0.01317852,
      "auxiliary_loss_mlp": 0.0105336,
      "balance_loss_clip": 1.08562124,
      "balance_loss_mlp": 1.03564513,
      "epoch": 0.040521854145373654,
      "flos": 22054466949120.0,
      "grad_norm": 1.943193987780698,
      "language_loss": 0.85139,
      "learning_rate": 3.998851898565085e-06,
      "loss": 0.8751021,
      "num_input_tokens_seen": 7155745,
      "step": 337,
      "time_per_iteration": 2.5173726081848145
    },
    {
      "auxiliary_loss_clip": 0.01314726,
      "auxiliary_loss_mlp": 0.01048657,
      "balance_loss_clip": 1.08427358,
      "balance_loss_mlp": 1.03190768,
      "epoch": 0.04064209703601274,
      "flos": 22674644196480.0,
      "grad_norm": 2.1707250638736135,
      "language_loss": 0.82978356,
      "learning_rate": 3.998825356342653e-06,
      "loss": 0.85341734,
      "num_input_tokens_seen": 7175920,
      "step": 338,
      "time_per_iteration": 2.5226211547851562
    },
    {
      "auxiliary_loss_clip": 0.0131692,
      "auxiliary_loss_mlp": 0.01064318,
      "balance_loss_clip": 1.0839572,
      "balance_loss_mlp": 1.04700851,
      "epoch": 0.04076233992665183,
      "flos": 38582172783360.0,
      "grad_norm": 2.348032330359331,
      "language_loss": 0.73048615,
      "learning_rate": 3.998798510900003e-06,
      "loss": 0.75429857,
      "num_input_tokens_seen": 7198720,
      "step": 339,
      "time_per_iteration": 2.6424903869628906
    },
    {
      "auxiliary_loss_clip": 0.01317186,
      "auxiliary_loss_mlp": 0.01053957,
      "balance_loss_clip": 1.08507109,
      "balance_loss_mlp": 1.03671885,
      "epoch": 0.04088258281729093,
      "flos": 25885309374720.0,
      "grad_norm": 4.171316213342774,
      "language_loss": 0.83692461,
      "learning_rate": 3.998771362241207e-06,
      "loss": 0.86063612,
      "num_input_tokens_seen": 7219125,
      "step": 340,
      "time_per_iteration": 2.5377330780029297
    },
    {
      "auxiliary_loss_clip": 0.01311549,
      "auxiliary_loss_mlp": 0.01054975,
      "balance_loss_clip": 1.08336401,
      "balance_loss_mlp": 1.03810692,
      "epoch": 0.04100282570793002,
      "flos": 19789832223360.0,
      "grad_norm": 1.8624021442822696,
      "language_loss": 0.876616,
      "learning_rate": 3.998743910370385e-06,
      "loss": 0.90028119,
      "num_input_tokens_seen": 7237985,
      "step": 341,
      "time_per_iteration": 3.3244218826293945
    },
    {
      "auxiliary_loss_clip": 0.01322695,
      "auxiliary_loss_mlp": 0.01049569,
      "balance_loss_clip": 1.09355974,
      "balance_loss_mlp": 1.03086543,
      "epoch": 0.04112306859856911,
      "flos": 22565152563840.0,
      "grad_norm": 2.0866266673048233,
      "language_loss": 0.73385412,
      "learning_rate": 3.998716155291702e-06,
      "loss": 0.75757676,
      "num_input_tokens_seen": 7255825,
      "step": 342,
      "time_per_iteration": 3.295494556427002
    },
    {
      "auxiliary_loss_clip": 0.01316787,
      "auxiliary_loss_mlp": 0.0105937,
      "balance_loss_clip": 1.08823419,
      "balance_loss_mlp": 1.04179811,
      "epoch": 0.0412433114892082,
      "flos": 25040654081280.0,
      "grad_norm": 1.7813420202434957,
      "language_loss": 0.90508056,
      "learning_rate": 3.998688097009366e-06,
      "loss": 0.92884207,
      "num_input_tokens_seen": 7276590,
      "step": 343,
      "time_per_iteration": 3.392993688583374
    },
    {
      "auxiliary_loss_clip": 0.01317468,
      "auxiliary_loss_mlp": 0.01054606,
      "balance_loss_clip": 1.08683801,
      "balance_loss_mlp": 1.03829789,
      "epoch": 0.04136355437984729,
      "flos": 25191371548800.0,
      "grad_norm": 2.13551725510914,
      "language_loss": 0.80172688,
      "learning_rate": 3.998659735527636e-06,
      "loss": 0.82544762,
      "num_input_tokens_seen": 7295680,
      "step": 344,
      "time_per_iteration": 3.27579665184021
    },
    {
      "auxiliary_loss_clip": 0.01314617,
      "auxiliary_loss_mlp": 0.0105403,
      "balance_loss_clip": 1.08498883,
      "balance_loss_mlp": 1.03670919,
      "epoch": 0.04148379727048638,
      "flos": 22966777509120.0,
      "grad_norm": 1.8327557282777152,
      "language_loss": 0.77637661,
      "learning_rate": 3.998631070850813e-06,
      "loss": 0.80006313,
      "num_input_tokens_seen": 7316300,
      "step": 345,
      "time_per_iteration": 2.50992488861084
    },
    {
      "auxiliary_loss_clip": 0.01312904,
      "auxiliary_loss_mlp": 0.01064336,
      "balance_loss_clip": 1.08687079,
      "balance_loss_mlp": 1.04860044,
      "epoch": 0.041604040161125476,
      "flos": 14063481187200.0,
      "grad_norm": 2.960064737292621,
      "language_loss": 0.83638465,
      "learning_rate": 3.9986021029832455e-06,
      "loss": 0.86015701,
      "num_input_tokens_seen": 7333615,
      "step": 346,
      "time_per_iteration": 2.4711108207702637
    },
    {
      "auxiliary_loss_clip": 0.01312436,
      "auxiliary_loss_mlp": 0.01055771,
      "balance_loss_clip": 1.08320427,
      "balance_loss_mlp": 1.03686416,
      "epoch": 0.041724283051764566,
      "flos": 12091877614080.0,
      "grad_norm": 4.551994444278132,
      "language_loss": 0.91755617,
      "learning_rate": 3.9985728319293285e-06,
      "loss": 0.94123816,
      "num_input_tokens_seen": 7347590,
      "step": 347,
      "time_per_iteration": 2.4509387016296387
    },
    {
      "auxiliary_loss_clip": 0.01317267,
      "auxiliary_loss_mlp": 0.01054278,
      "balance_loss_clip": 1.08412385,
      "balance_loss_mlp": 1.03663468,
      "epoch": 0.041844525942403656,
      "flos": 12385303816320.0,
      "grad_norm": 2.3224384490317864,
      "language_loss": 0.8525244,
      "learning_rate": 3.998543257693501e-06,
      "loss": 0.8762399,
      "num_input_tokens_seen": 7364345,
      "step": 348,
      "time_per_iteration": 2.4601216316223145
    },
    {
      "auxiliary_loss_clip": 0.01314363,
      "auxiliary_loss_mlp": 0.01065657,
      "balance_loss_clip": 1.08635926,
      "balance_loss_mlp": 1.0492177,
      "epoch": 0.041964768833042745,
      "flos": 23769345041280.0,
      "grad_norm": 2.248021269233625,
      "language_loss": 0.87798762,
      "learning_rate": 3.998513380280251e-06,
      "loss": 0.90178782,
      "num_input_tokens_seen": 7384625,
      "step": 349,
      "time_per_iteration": 2.522496461868286
    },
    {
      "auxiliary_loss_clip": 0.01317368,
      "auxiliary_loss_mlp": 0.01069265,
      "balance_loss_clip": 1.08583426,
      "balance_loss_mlp": 1.05029845,
      "epoch": 0.042085011723681835,
      "flos": 11875336473600.0,
      "grad_norm": 2.8138389743529975,
      "language_loss": 0.95031047,
      "learning_rate": 3.99848319969411e-06,
      "loss": 0.97417688,
      "num_input_tokens_seen": 7402225,
      "step": 350,
      "time_per_iteration": 2.458815097808838
    },
    {
      "auxiliary_loss_clip": 0.01319429,
      "auxiliary_loss_mlp": 0.01063121,
      "balance_loss_clip": 1.08800197,
      "balance_loss_mlp": 1.04487002,
      "epoch": 0.042205254614320925,
      "flos": 16873957964160.0,
      "grad_norm": 2.282321927813428,
      "language_loss": 0.79577434,
      "learning_rate": 3.9984527159396564e-06,
      "loss": 0.81959981,
      "num_input_tokens_seen": 7420865,
      "step": 351,
      "time_per_iteration": 2.467747449874878
    },
    {
      "auxiliary_loss_clip": 0.01313021,
      "auxiliary_loss_mlp": 0.01055937,
      "balance_loss_clip": 1.08246279,
      "balance_loss_mlp": 1.03935456,
      "epoch": 0.04232549750496002,
      "flos": 25118508810240.0,
      "grad_norm": 2.2321062629986983,
      "language_loss": 0.84833091,
      "learning_rate": 3.9984219290215154e-06,
      "loss": 0.87202048,
      "num_input_tokens_seen": 7441040,
      "step": 352,
      "time_per_iteration": 2.5033609867095947
    },
    {
      "auxiliary_loss_clip": 0.01311988,
      "auxiliary_loss_mlp": 0.01048693,
      "balance_loss_clip": 1.08593845,
      "balance_loss_mlp": 1.03308797,
      "epoch": 0.04244574039559911,
      "flos": 26724541714560.0,
      "grad_norm": 3.782901108407086,
      "language_loss": 0.89161533,
      "learning_rate": 3.998390838944356e-06,
      "loss": 0.91522217,
      "num_input_tokens_seen": 7462545,
      "step": 353,
      "time_per_iteration": 2.5348618030548096
    },
    {
      "auxiliary_loss_clip": 0.01313427,
      "auxiliary_loss_mlp": 0.01060902,
      "balance_loss_clip": 1.08533001,
      "balance_loss_mlp": 1.04473722,
      "epoch": 0.0425659832862382,
      "flos": 20923244951040.0,
      "grad_norm": 2.1331455820739187,
      "language_loss": 0.90390348,
      "learning_rate": 3.998359445712895e-06,
      "loss": 0.92764676,
      "num_input_tokens_seen": 7481650,
      "step": 354,
      "time_per_iteration": 2.5024843215942383
    },
    {
      "auxiliary_loss_clip": 0.01310854,
      "auxiliary_loss_mlp": 0.01050327,
      "balance_loss_clip": 1.08206868,
      "balance_loss_mlp": 1.03448427,
      "epoch": 0.04268622617687729,
      "flos": 23331127115520.0,
      "grad_norm": 2.2933715559087076,
      "language_loss": 0.81207919,
      "learning_rate": 3.9983277493318955e-06,
      "loss": 0.83569098,
      "num_input_tokens_seen": 7500945,
      "step": 355,
      "time_per_iteration": 2.5418589115142822
    },
    {
      "auxiliary_loss_clip": 0.01314101,
      "auxiliary_loss_mlp": 0.01053424,
      "balance_loss_clip": 1.08197987,
      "balance_loss_mlp": 1.03690147,
      "epoch": 0.04280646906751638,
      "flos": 25994010908160.0,
      "grad_norm": 1.6714723407154795,
      "language_loss": 0.81337923,
      "learning_rate": 3.998295749806165e-06,
      "loss": 0.83705449,
      "num_input_tokens_seen": 7522170,
      "step": 356,
      "time_per_iteration": 2.523563861846924
    },
    {
      "auxiliary_loss_clip": 0.01314028,
      "auxiliary_loss_mlp": 0.01068793,
      "balance_loss_clip": 1.08721352,
      "balance_loss_mlp": 1.05179369,
      "epoch": 0.04292671195815547,
      "flos": 26906824258560.0,
      "grad_norm": 1.8169540640186919,
      "language_loss": 0.83275414,
      "learning_rate": 3.998263447140558e-06,
      "loss": 0.85658234,
      "num_input_tokens_seen": 7542370,
      "step": 357,
      "time_per_iteration": 2.538599729537964
    },
    {
      "auxiliary_loss_clip": 0.01309809,
      "auxiliary_loss_mlp": 0.0104728,
      "balance_loss_clip": 1.08093023,
      "balance_loss_mlp": 1.03097248,
      "epoch": 0.04304695484879457,
      "flos": 39457315745280.0,
      "grad_norm": 1.753371143794966,
      "language_loss": 0.81925076,
      "learning_rate": 3.998230841339976e-06,
      "loss": 0.84282172,
      "num_input_tokens_seen": 7564380,
      "step": 358,
      "time_per_iteration": 2.637113332748413
    },
    {
      "auxiliary_loss_clip": 0.01309956,
      "auxiliary_loss_mlp": 0.01051402,
      "balance_loss_clip": 1.08537078,
      "balance_loss_mlp": 1.03527308,
      "epoch": 0.04316719773943366,
      "flos": 19646297475840.0,
      "grad_norm": 2.2657939925781188,
      "language_loss": 0.84906995,
      "learning_rate": 3.998197932409363e-06,
      "loss": 0.87268353,
      "num_input_tokens_seen": 7582390,
      "step": 359,
      "time_per_iteration": 2.501049041748047
    },
    {
      "auxiliary_loss_clip": 0.01303976,
      "auxiliary_loss_mlp": 0.01058092,
      "balance_loss_clip": 1.08091342,
      "balance_loss_mlp": 1.04223669,
      "epoch": 0.04328744063007275,
      "flos": 22452320966400.0,
      "grad_norm": 2.193100691262511,
      "language_loss": 0.86434275,
      "learning_rate": 3.9981647203537125e-06,
      "loss": 0.88796341,
      "num_input_tokens_seen": 7599890,
      "step": 360,
      "time_per_iteration": 2.4809176921844482
    },
    {
      "auxiliary_loss_clip": 0.01307359,
      "auxiliary_loss_mlp": 0.01062598,
      "balance_loss_clip": 1.08094239,
      "balance_loss_mlp": 1.04727912,
      "epoch": 0.04340768352071184,
      "flos": 21283033530240.0,
      "grad_norm": 2.0729731843907726,
      "language_loss": 0.96005911,
      "learning_rate": 3.998131205178063e-06,
      "loss": 0.98375863,
      "num_input_tokens_seen": 7618360,
      "step": 361,
      "time_per_iteration": 2.465364694595337
    },
    {
      "auxiliary_loss_clip": 0.01307294,
      "auxiliary_loss_mlp": 0.01057581,
      "balance_loss_clip": 1.08154821,
      "balance_loss_mlp": 1.04179764,
      "epoch": 0.04352792641135093,
      "flos": 11583705951360.0,
      "grad_norm": 4.862272840517238,
      "language_loss": 0.76420879,
      "learning_rate": 3.998097386887498e-06,
      "loss": 0.78785759,
      "num_input_tokens_seen": 7635435,
      "step": 362,
      "time_per_iteration": 2.4743475914001465
    },
    {
      "auxiliary_loss_clip": 0.01304918,
      "auxiliary_loss_mlp": 0.01067394,
      "balance_loss_clip": 1.08116436,
      "balance_loss_mlp": 1.05105066,
      "epoch": 0.04364816930199002,
      "flos": 23623547736960.0,
      "grad_norm": 1.9716536780055414,
      "language_loss": 0.85103494,
      "learning_rate": 3.998063265487148e-06,
      "loss": 0.87475806,
      "num_input_tokens_seen": 7656485,
      "step": 363,
      "time_per_iteration": 2.5508499145507812
    },
    {
      "auxiliary_loss_clip": 0.01308042,
      "auxiliary_loss_mlp": 0.01056062,
      "balance_loss_clip": 1.08364844,
      "balance_loss_mlp": 1.04043388,
      "epoch": 0.043768412192629114,
      "flos": 14429734214400.0,
      "grad_norm": 2.209335022683545,
      "language_loss": 0.80973411,
      "learning_rate": 3.99802884098219e-06,
      "loss": 0.8333751,
      "num_input_tokens_seen": 7674595,
      "step": 364,
      "time_per_iteration": 2.4609665870666504
    },
    {
      "auxiliary_loss_clip": 0.01307313,
      "auxiliary_loss_mlp": 0.01047458,
      "balance_loss_clip": 1.08088434,
      "balance_loss_mlp": 1.03095937,
      "epoch": 0.043888655083268203,
      "flos": 26468893641600.0,
      "grad_norm": 2.1858082866538378,
      "language_loss": 0.82345247,
      "learning_rate": 3.997994113377845e-06,
      "loss": 0.84700024,
      "num_input_tokens_seen": 7693495,
      "step": 365,
      "time_per_iteration": 2.530149459838867
    },
    {
      "auxiliary_loss_clip": 0.01306787,
      "auxiliary_loss_mlp": 0.01048853,
      "balance_loss_clip": 1.08145404,
      "balance_loss_mlp": 1.03215146,
      "epoch": 0.04400889797390729,
      "flos": 27235263242880.0,
      "grad_norm": 2.3087204289123235,
      "language_loss": 0.83136582,
      "learning_rate": 3.9979590826793815e-06,
      "loss": 0.85492224,
      "num_input_tokens_seen": 7714685,
      "step": 366,
      "time_per_iteration": 2.5117695331573486
    },
    {
      "auxiliary_loss_clip": 0.01310481,
      "auxiliary_loss_mlp": 0.01053061,
      "balance_loss_clip": 1.08402312,
      "balance_loss_mlp": 1.03628802,
      "epoch": 0.04412914086454638,
      "flos": 20119528183680.0,
      "grad_norm": 2.37421501344759,
      "language_loss": 0.80873638,
      "learning_rate": 3.997923748892113e-06,
      "loss": 0.83237183,
      "num_input_tokens_seen": 7734005,
      "step": 367,
      "time_per_iteration": 3.252614974975586
    },
    {
      "auxiliary_loss_clip": 0.01305054,
      "auxiliary_loss_mlp": 0.01051596,
      "balance_loss_clip": 1.08323932,
      "balance_loss_mlp": 1.03655171,
      "epoch": 0.04424938375518547,
      "flos": 22604618632320.0,
      "grad_norm": 2.060065623354843,
      "language_loss": 0.88556302,
      "learning_rate": 3.9978881120214015e-06,
      "loss": 0.90912956,
      "num_input_tokens_seen": 7755525,
      "step": 368,
      "time_per_iteration": 2.515429973602295
    },
    {
      "auxiliary_loss_clip": 0.01307272,
      "auxiliary_loss_mlp": 0.01050379,
      "balance_loss_clip": 1.08134317,
      "balance_loss_mlp": 1.03366542,
      "epoch": 0.04436962664582456,
      "flos": 24132365844480.0,
      "grad_norm": 1.9910126897436693,
      "language_loss": 0.79411018,
      "learning_rate": 3.997852172072652e-06,
      "loss": 0.81768668,
      "num_input_tokens_seen": 7776740,
      "step": 369,
      "time_per_iteration": 2.5036027431488037
    },
    {
      "auxiliary_loss_clip": 0.0130762,
      "auxiliary_loss_mlp": 0.01061772,
      "balance_loss_clip": 1.08186865,
      "balance_loss_mlp": 1.04572678,
      "epoch": 0.04448986953646366,
      "flos": 18222906251520.0,
      "grad_norm": 4.881557466520376,
      "language_loss": 0.8943032,
      "learning_rate": 3.9978159290513155e-06,
      "loss": 0.91799712,
      "num_input_tokens_seen": 7794820,
      "step": 370,
      "time_per_iteration": 3.3193178176879883
    },
    {
      "auxiliary_loss_clip": 0.01307958,
      "auxiliary_loss_mlp": 0.01065044,
      "balance_loss_clip": 1.0822525,
      "balance_loss_mlp": 1.04829526,
      "epoch": 0.04461011242710275,
      "flos": 30117920400000.0,
      "grad_norm": 1.6630904649713403,
      "language_loss": 0.80331135,
      "learning_rate": 3.997779382962892e-06,
      "loss": 0.82704139,
      "num_input_tokens_seen": 7817705,
      "step": 371,
      "time_per_iteration": 2.547529458999634
    },
    {
      "auxiliary_loss_clip": 0.0130169,
      "auxiliary_loss_mlp": 0.01055319,
      "balance_loss_clip": 1.0793817,
      "balance_loss_mlp": 1.03841484,
      "epoch": 0.04473035531774184,
      "flos": 29752529299200.0,
      "grad_norm": 2.1486411377727923,
      "language_loss": 0.73549217,
      "learning_rate": 3.997742533812924e-06,
      "loss": 0.75906223,
      "num_input_tokens_seen": 7840970,
      "step": 372,
      "time_per_iteration": 3.419358730316162
    },
    {
      "auxiliary_loss_clip": 0.01307919,
      "auxiliary_loss_mlp": 0.01067117,
      "balance_loss_clip": 1.08419704,
      "balance_loss_mlp": 1.05107129,
      "epoch": 0.04485059820838093,
      "flos": 13151565676800.0,
      "grad_norm": 2.450927137963643,
      "language_loss": 0.92658639,
      "learning_rate": 3.997705381607001e-06,
      "loss": 0.95033681,
      "num_input_tokens_seen": 7857785,
      "step": 373,
      "time_per_iteration": 2.4556984901428223
    },
    {
      "auxiliary_loss_clip": 0.01213861,
      "auxiliary_loss_mlp": 0.01050382,
      "balance_loss_clip": 1.06126595,
      "balance_loss_mlp": 1.04227602,
      "epoch": 0.04497084109902002,
      "flos": 68094209548800.0,
      "grad_norm": 0.9790704418458376,
      "language_loss": 0.60295194,
      "learning_rate": 3.997667926350761e-06,
      "loss": 0.62559438,
      "num_input_tokens_seen": 7916115,
      "step": 374,
      "time_per_iteration": 2.9758031368255615
    },
    {
      "auxiliary_loss_clip": 0.01212173,
      "auxiliary_loss_mlp": 0.01044617,
      "balance_loss_clip": 1.06006777,
      "balance_loss_mlp": 1.036129,
      "epoch": 0.04509108398965911,
      "flos": 64342263346560.0,
      "grad_norm": 0.9080632105557515,
      "language_loss": 0.57768965,
      "learning_rate": 3.997630168049886e-06,
      "loss": 0.60025764,
      "num_input_tokens_seen": 7974480,
      "step": 375,
      "time_per_iteration": 3.054396629333496
    },
    {
      "auxiliary_loss_clip": 0.01308536,
      "auxiliary_loss_mlp": 0.01060521,
      "balance_loss_clip": 1.08260226,
      "balance_loss_mlp": 1.04392719,
      "epoch": 0.045211326880298205,
      "flos": 22271115830400.0,
      "grad_norm": 1.8107792463460122,
      "language_loss": 0.77470303,
      "learning_rate": 3.997592106710101e-06,
      "loss": 0.79839367,
      "num_input_tokens_seen": 7993940,
      "step": 376,
      "time_per_iteration": 2.5062403678894043
    },
    {
      "auxiliary_loss_clip": 0.01301868,
      "auxiliary_loss_mlp": 0.01051322,
      "balance_loss_clip": 1.08013248,
      "balance_loss_mlp": 1.03538346,
      "epoch": 0.045331569770937295,
      "flos": 32159441796480.0,
      "grad_norm": 2.8265780724162384,
      "language_loss": 0.65737206,
      "learning_rate": 3.997553742337182e-06,
      "loss": 0.68090397,
      "num_input_tokens_seen": 8013365,
      "step": 377,
      "time_per_iteration": 2.5672574043273926
    },
    {
      "auxiliary_loss_clip": 0.01304414,
      "auxiliary_loss_mlp": 0.01055704,
      "balance_loss_clip": 1.08102703,
      "balance_loss_mlp": 1.03939605,
      "epoch": 0.045451812661576385,
      "flos": 22163455791360.0,
      "grad_norm": 1.8365341766960068,
      "language_loss": 0.912081,
      "learning_rate": 3.997515074936949e-06,
      "loss": 0.93568218,
      "num_input_tokens_seen": 8034240,
      "step": 378,
      "time_per_iteration": 2.5110716819763184
    },
    {
      "auxiliary_loss_clip": 0.01304302,
      "auxiliary_loss_mlp": 0.01059454,
      "balance_loss_clip": 1.0807929,
      "balance_loss_mlp": 1.0435034,
      "epoch": 0.045572055552215475,
      "flos": 16581968305920.0,
      "grad_norm": 3.127611100388022,
      "language_loss": 0.86878395,
      "learning_rate": 3.997476104515268e-06,
      "loss": 0.89242154,
      "num_input_tokens_seen": 8052430,
      "step": 379,
      "time_per_iteration": 2.490489959716797
    },
    {
      "auxiliary_loss_clip": 0.01303696,
      "auxiliary_loss_mlp": 0.01058352,
      "balance_loss_clip": 1.08355606,
      "balance_loss_mlp": 1.04277122,
      "epoch": 0.045692298442854565,
      "flos": 17603375448960.0,
      "grad_norm": 3.5862771205039743,
      "language_loss": 0.77378345,
      "learning_rate": 3.9974368310780485e-06,
      "loss": 0.79740387,
      "num_input_tokens_seen": 8069605,
      "step": 380,
      "time_per_iteration": 2.4650542736053467
    },
    {
      "auxiliary_loss_clip": 0.01313976,
      "auxiliary_loss_mlp": 0.01059389,
      "balance_loss_clip": 1.0852859,
      "balance_loss_mlp": 1.04187727,
      "epoch": 0.045812541333493655,
      "flos": 26761098781440.0,
      "grad_norm": 2.969493865473316,
      "language_loss": 0.74292934,
      "learning_rate": 3.997397254631251e-06,
      "loss": 0.76666296,
      "num_input_tokens_seen": 8090225,
      "step": 381,
      "time_per_iteration": 2.5366361141204834
    },
    {
      "auxiliary_loss_clip": 0.01201911,
      "auxiliary_loss_mlp": 0.01041464,
      "balance_loss_clip": 1.05471504,
      "balance_loss_mlp": 1.03407311,
      "epoch": 0.04593278422413275,
      "flos": 60250349894400.0,
      "grad_norm": 0.826749808815751,
      "language_loss": 0.60024762,
      "learning_rate": 3.997357375180878e-06,
      "loss": 0.62268138,
      "num_input_tokens_seen": 8154505,
      "step": 382,
      "time_per_iteration": 3.1870150566101074
    },
    {
      "auxiliary_loss_clip": 0.01308093,
      "auxiliary_loss_mlp": 0.01050148,
      "balance_loss_clip": 1.08306384,
      "balance_loss_mlp": 1.03308892,
      "epoch": 0.04605302711477184,
      "flos": 21799249839360.0,
      "grad_norm": 1.9575818561244382,
      "language_loss": 0.75227451,
      "learning_rate": 3.997317192732979e-06,
      "loss": 0.77585691,
      "num_input_tokens_seen": 8173285,
      "step": 383,
      "time_per_iteration": 2.5041611194610596
    },
    {
      "auxiliary_loss_clip": 0.01307153,
      "auxiliary_loss_mlp": 0.01062596,
      "balance_loss_clip": 1.08269322,
      "balance_loss_mlp": 1.04579926,
      "epoch": 0.04617327000541093,
      "flos": 19459705299840.0,
      "grad_norm": 15.082697588225773,
      "language_loss": 0.82512033,
      "learning_rate": 3.99727670729365e-06,
      "loss": 0.84881777,
      "num_input_tokens_seen": 8191845,
      "step": 384,
      "time_per_iteration": 2.483057975769043
    },
    {
      "auxiliary_loss_clip": 0.01305237,
      "auxiliary_loss_mlp": 0.01059073,
      "balance_loss_clip": 1.08576119,
      "balance_loss_mlp": 1.0434922,
      "epoch": 0.04629351289605002,
      "flos": 25411468135680.0,
      "grad_norm": 1.8832234573273792,
      "language_loss": 0.77701378,
      "learning_rate": 3.997235918869033e-06,
      "loss": 0.80065686,
      "num_input_tokens_seen": 8212880,
      "step": 385,
      "time_per_iteration": 2.584770679473877
    },
    {
      "auxiliary_loss_clip": 0.01307316,
      "auxiliary_loss_mlp": 0.0104868,
      "balance_loss_clip": 1.08620346,
      "balance_loss_mlp": 1.03336191,
      "epoch": 0.04641375578668911,
      "flos": 20558284813440.0,
      "grad_norm": 2.1588079040337074,
      "language_loss": 0.82640243,
      "learning_rate": 3.997194827465315e-06,
      "loss": 0.84996235,
      "num_input_tokens_seen": 8231475,
      "step": 386,
      "time_per_iteration": 2.5023181438446045
    },
    {
      "auxiliary_loss_clip": 0.01305475,
      "auxiliary_loss_mlp": 0.01048771,
      "balance_loss_clip": 1.08197701,
      "balance_loss_mlp": 1.03318989,
      "epoch": 0.0465339986773282,
      "flos": 13188661447680.0,
      "grad_norm": 3.255443470714345,
      "language_loss": 0.91170299,
      "learning_rate": 3.997153433088728e-06,
      "loss": 0.93524545,
      "num_input_tokens_seen": 8248600,
      "step": 387,
      "time_per_iteration": 2.462891101837158
    },
    {
      "auxiliary_loss_clip": 0.01306422,
      "auxiliary_loss_mlp": 0.01054352,
      "balance_loss_clip": 1.08443904,
      "balance_loss_mlp": 1.03742421,
      "epoch": 0.0466542415679673,
      "flos": 25556547168000.0,
      "grad_norm": 2.0165489000311903,
      "language_loss": 0.8136313,
      "learning_rate": 3.997111735745554e-06,
      "loss": 0.83723903,
      "num_input_tokens_seen": 8271570,
      "step": 388,
      "time_per_iteration": 2.5356483459472656
    },
    {
      "auxiliary_loss_clip": 0.01302901,
      "auxiliary_loss_mlp": 0.01060489,
      "balance_loss_clip": 1.08219302,
      "balance_loss_mlp": 1.0427146,
      "epoch": 0.04677448445860639,
      "flos": 22236749493120.0,
      "grad_norm": 1.849563944640564,
      "language_loss": 0.82464463,
      "learning_rate": 3.997069735442118e-06,
      "loss": 0.84827852,
      "num_input_tokens_seen": 8291265,
      "step": 389,
      "time_per_iteration": 2.526449680328369
    },
    {
      "auxiliary_loss_clip": 0.0130254,
      "auxiliary_loss_mlp": 0.0105639,
      "balance_loss_clip": 1.08192551,
      "balance_loss_mlp": 1.04064286,
      "epoch": 0.04689472734924548,
      "flos": 28147825198080.0,
      "grad_norm": 1.4598805329766036,
      "language_loss": 0.80379522,
      "learning_rate": 3.997027432184792e-06,
      "loss": 0.82738447,
      "num_input_tokens_seen": 8315925,
      "step": 390,
      "time_per_iteration": 2.6069769859313965
    },
    {
      "auxiliary_loss_clip": 0.0130467,
      "auxiliary_loss_mlp": 0.010543,
      "balance_loss_clip": 1.08359206,
      "balance_loss_mlp": 1.03882658,
      "epoch": 0.04701497023988457,
      "flos": 23148952312320.0,
      "grad_norm": 2.013483234248443,
      "language_loss": 0.89315629,
      "learning_rate": 3.99698482597999e-06,
      "loss": 0.91674602,
      "num_input_tokens_seen": 8333605,
      "step": 391,
      "time_per_iteration": 2.4999606609344482
    },
    {
      "auxiliary_loss_clip": 0.01194121,
      "auxiliary_loss_mlp": 0.01014802,
      "balance_loss_clip": 1.05286396,
      "balance_loss_mlp": 1.00836456,
      "epoch": 0.04713521313052366,
      "flos": 64827668764800.0,
      "grad_norm": 0.8743215866590295,
      "language_loss": 0.63898301,
      "learning_rate": 3.99694191683418e-06,
      "loss": 0.66107219,
      "num_input_tokens_seen": 8394405,
      "step": 392,
      "time_per_iteration": 3.077955484390259
    },
    {
      "auxiliary_loss_clip": 0.0130794,
      "auxiliary_loss_mlp": 0.01055877,
      "balance_loss_clip": 1.08684731,
      "balance_loss_mlp": 1.03896117,
      "epoch": 0.047255456021162746,
      "flos": 18771585477120.0,
      "grad_norm": 1.9766125883336993,
      "language_loss": 0.81709909,
      "learning_rate": 3.996898704753867e-06,
      "loss": 0.84073722,
      "num_input_tokens_seen": 8412355,
      "step": 393,
      "time_per_iteration": 2.467668056488037
    },
    {
      "auxiliary_loss_clip": 0.01301119,
      "auxiliary_loss_mlp": 0.0105127,
      "balance_loss_clip": 1.0808208,
      "balance_loss_mlp": 1.03555799,
      "epoch": 0.04737569891180184,
      "flos": 22053820504320.0,
      "grad_norm": 2.1282780018454623,
      "language_loss": 0.87553906,
      "learning_rate": 3.996855189745609e-06,
      "loss": 0.89906299,
      "num_input_tokens_seen": 8431620,
      "step": 394,
      "time_per_iteration": 3.3214595317840576
    },
    {
      "auxiliary_loss_clip": 0.01301429,
      "auxiliary_loss_mlp": 0.01058369,
      "balance_loss_clip": 1.08062744,
      "balance_loss_mlp": 1.04173899,
      "epoch": 0.04749594180244093,
      "flos": 29057370410880.0,
      "grad_norm": 1.7742712768962188,
      "language_loss": 0.92587304,
      "learning_rate": 3.996811371816007e-06,
      "loss": 0.949471,
      "num_input_tokens_seen": 8454045,
      "step": 395,
      "time_per_iteration": 2.5340468883514404
    },
    {
      "auxiliary_loss_clip": 0.01304723,
      "auxiliary_loss_mlp": 0.01062,
      "balance_loss_clip": 1.08485949,
      "balance_loss_mlp": 1.04672945,
      "epoch": 0.04761618469308002,
      "flos": 35112268172160.0,
      "grad_norm": 1.9140049090450078,
      "language_loss": 0.78198045,
      "learning_rate": 3.996767250971707e-06,
      "loss": 0.80564767,
      "num_input_tokens_seen": 8476785,
      "step": 396,
      "time_per_iteration": 2.604710340499878
    },
    {
      "auxiliary_loss_clip": 0.01306915,
      "auxiliary_loss_mlp": 0.01052324,
      "balance_loss_clip": 1.08561552,
      "balance_loss_mlp": 1.03587341,
      "epoch": 0.04773642758371911,
      "flos": 25630702796160.0,
      "grad_norm": 1.8965995587752826,
      "language_loss": 0.86975163,
      "learning_rate": 3.996722827219403e-06,
      "loss": 0.89334404,
      "num_input_tokens_seen": 8498400,
      "step": 397,
      "time_per_iteration": 3.320405960083008
    },
    {
      "auxiliary_loss_clip": 0.01310238,
      "auxiliary_loss_mlp": 0.01061852,
      "balance_loss_clip": 1.08772373,
      "balance_loss_mlp": 1.0454371,
      "epoch": 0.0478566704743582,
      "flos": 20631506688000.0,
      "grad_norm": 2.7339659543636303,
      "language_loss": 0.82932961,
      "learning_rate": 3.996678100565833e-06,
      "loss": 0.85305053,
      "num_input_tokens_seen": 8517455,
      "step": 398,
      "time_per_iteration": 2.5041348934173584
    },
    {
      "auxiliary_loss_clip": 0.0129915,
      "auxiliary_loss_mlp": 0.01055837,
      "balance_loss_clip": 1.08027911,
      "balance_loss_mlp": 1.03838503,
      "epoch": 0.04797691336499729,
      "flos": 18835721210880.0,
      "grad_norm": 2.2288814645322828,
      "language_loss": 0.88556743,
      "learning_rate": 3.996633071017783e-06,
      "loss": 0.90911722,
      "num_input_tokens_seen": 8534085,
      "step": 399,
      "time_per_iteration": 4.04193902015686
    },
    {
      "auxiliary_loss_clip": 0.01300727,
      "auxiliary_loss_mlp": 0.01055607,
      "balance_loss_clip": 1.08265805,
      "balance_loss_mlp": 1.0391798,
      "epoch": 0.04809715625563638,
      "flos": 21099673578240.0,
      "grad_norm": 2.366517592869167,
      "language_loss": 0.81760651,
      "learning_rate": 3.996587738582084e-06,
      "loss": 0.84116983,
      "num_input_tokens_seen": 8550885,
      "step": 400,
      "time_per_iteration": 2.4908688068389893
    },
    {
      "auxiliary_loss_clip": 0.01298287,
      "auxiliary_loss_mlp": 0.01045888,
      "balance_loss_clip": 1.07882285,
      "balance_loss_mlp": 1.03060579,
      "epoch": 0.04821739914627548,
      "flos": 23805650712960.0,
      "grad_norm": 13.411460199123866,
      "language_loss": 0.86008286,
      "learning_rate": 3.9965421032656115e-06,
      "loss": 0.88352466,
      "num_input_tokens_seen": 8570815,
      "step": 401,
      "time_per_iteration": 2.5356838703155518
    },
    {
      "auxiliary_loss_clip": 0.0130131,
      "auxiliary_loss_mlp": 0.01051442,
      "balance_loss_clip": 1.08075023,
      "balance_loss_mlp": 1.03390598,
      "epoch": 0.04833764203691457,
      "flos": 22200587475840.0,
      "grad_norm": 2.7015538418090874,
      "language_loss": 0.94228053,
      "learning_rate": 3.99649616507529e-06,
      "loss": 0.96580803,
      "num_input_tokens_seen": 8589910,
      "step": 402,
      "time_per_iteration": 2.515148639678955
    },
    {
      "auxiliary_loss_clip": 0.01191461,
      "auxiliary_loss_mlp": 0.01014412,
      "balance_loss_clip": 1.05255461,
      "balance_loss_mlp": 1.00852287,
      "epoch": 0.04845788492755366,
      "flos": 65904376896000.0,
      "grad_norm": 0.8948925615385385,
      "language_loss": 0.63133913,
      "learning_rate": 3.996449924018088e-06,
      "loss": 0.65339786,
      "num_input_tokens_seen": 8650370,
      "step": 403,
      "time_per_iteration": 3.001668930053711
    },
    {
      "auxiliary_loss_clip": 0.01297493,
      "auxiliary_loss_mlp": 0.01055731,
      "balance_loss_clip": 1.08124352,
      "balance_loss_mlp": 1.04083014,
      "epoch": 0.04857812781819275,
      "flos": 19281301424640.0,
      "grad_norm": 3.3128949042104976,
      "language_loss": 0.79501402,
      "learning_rate": 3.99640338010102e-06,
      "loss": 0.81854618,
      "num_input_tokens_seen": 8669475,
      "step": 404,
      "time_per_iteration": 2.5077438354492188
    },
    {
      "auxiliary_loss_clip": 0.01296881,
      "auxiliary_loss_mlp": 0.01050455,
      "balance_loss_clip": 1.07876885,
      "balance_loss_mlp": 1.03388453,
      "epoch": 0.04869837070883184,
      "flos": 24062376193920.0,
      "grad_norm": 1.8264005464572708,
      "language_loss": 0.78732586,
      "learning_rate": 3.996356533331146e-06,
      "loss": 0.81079918,
      "num_input_tokens_seen": 8691345,
      "step": 405,
      "time_per_iteration": 2.5813164710998535
    },
    {
      "auxiliary_loss_clip": 0.01309015,
      "auxiliary_loss_mlp": 0.01047016,
      "balance_loss_clip": 1.0822556,
      "balance_loss_mlp": 1.03093433,
      "epoch": 0.04881861359947093,
      "flos": 25187169657600.0,
      "grad_norm": 2.299877226344788,
      "language_loss": 0.61913782,
      "learning_rate": 3.996309383715573e-06,
      "loss": 0.64269817,
      "num_input_tokens_seen": 8710125,
      "step": 406,
      "time_per_iteration": 2.5506973266601562
    },
    {
      "auxiliary_loss_clip": 0.01305136,
      "auxiliary_loss_mlp": 0.0104492,
      "balance_loss_clip": 1.08353484,
      "balance_loss_mlp": 1.02932703,
      "epoch": 0.048938856490110025,
      "flos": 16362913213440.0,
      "grad_norm": 4.542308600645671,
      "language_loss": 0.73766047,
      "learning_rate": 3.996261931261454e-06,
      "loss": 0.76116097,
      "num_input_tokens_seen": 8728705,
      "step": 407,
      "time_per_iteration": 2.4768905639648438
    },
    {
      "auxiliary_loss_clip": 0.01301929,
      "auxiliary_loss_mlp": 0.01050798,
      "balance_loss_clip": 1.08295596,
      "balance_loss_mlp": 1.03481197,
      "epoch": 0.049059099380749115,
      "flos": 29895094379520.0,
      "grad_norm": 1.69931226147769,
      "language_loss": 0.86397779,
      "learning_rate": 3.996214175975987e-06,
      "loss": 0.88750505,
      "num_input_tokens_seen": 8749225,
      "step": 408,
      "time_per_iteration": 2.541062831878662
    },
    {
      "auxiliary_loss_clip": 0.01305683,
      "auxiliary_loss_mlp": 0.01056994,
      "balance_loss_clip": 1.08446574,
      "balance_loss_mlp": 1.04096055,
      "epoch": 0.049179342271388204,
      "flos": 35918858027520.0,
      "grad_norm": 16.323599984219786,
      "language_loss": 0.79003918,
      "learning_rate": 3.996166117866417e-06,
      "loss": 0.81366599,
      "num_input_tokens_seen": 8771160,
      "step": 409,
      "time_per_iteration": 2.6045706272125244
    },
    {
      "auxiliary_loss_clip": 0.01296232,
      "auxiliary_loss_mlp": 0.01049345,
      "balance_loss_clip": 1.07948112,
      "balance_loss_mlp": 1.03385949,
      "epoch": 0.049299585162027294,
      "flos": 14611226659200.0,
      "grad_norm": 1.9343237973452194,
      "language_loss": 0.86551613,
      "learning_rate": 3.996117756940035e-06,
      "loss": 0.88897181,
      "num_input_tokens_seen": 8787845,
      "step": 410,
      "time_per_iteration": 2.472079277038574
    },
    {
      "auxiliary_loss_clip": 0.0130136,
      "auxiliary_loss_mlp": 0.01048057,
      "balance_loss_clip": 1.08308649,
      "balance_loss_mlp": 1.03274441,
      "epoch": 0.049419828052666384,
      "flos": 19567939956480.0,
      "grad_norm": 2.5200190517528114,
      "language_loss": 0.97680199,
      "learning_rate": 3.996069093204175e-06,
      "loss": 1.00029612,
      "num_input_tokens_seen": 8803805,
      "step": 411,
      "time_per_iteration": 2.5612008571624756
    },
    {
      "auxiliary_loss_clip": 0.01307525,
      "auxiliary_loss_mlp": 0.01053742,
      "balance_loss_clip": 1.08575988,
      "balance_loss_mlp": 1.0374223,
      "epoch": 0.049540070943305474,
      "flos": 13659916907520.0,
      "grad_norm": 2.5097680591367753,
      "language_loss": 0.88016677,
      "learning_rate": 3.996020126666221e-06,
      "loss": 0.90377945,
      "num_input_tokens_seen": 8820785,
      "step": 412,
      "time_per_iteration": 2.5755655765533447
    },
    {
      "auxiliary_loss_clip": 0.01299896,
      "auxiliary_loss_mlp": 0.01047871,
      "balance_loss_clip": 1.08220887,
      "balance_loss_mlp": 1.03295815,
      "epoch": 0.04966031383394457,
      "flos": 21832035978240.0,
      "grad_norm": 1.955007345608373,
      "language_loss": 0.81952274,
      "learning_rate": 3.995970857333601e-06,
      "loss": 0.84300047,
      "num_input_tokens_seen": 8841195,
      "step": 413,
      "time_per_iteration": 2.533642530441284
    },
    {
      "auxiliary_loss_clip": 0.01301584,
      "auxiliary_loss_mlp": 0.01051497,
      "balance_loss_clip": 1.08071661,
      "balance_loss_mlp": 1.03554714,
      "epoch": 0.04978055672458366,
      "flos": 28618793349120.0,
      "grad_norm": 1.8510990305408832,
      "language_loss": 0.79738045,
      "learning_rate": 3.995921285213789e-06,
      "loss": 0.82091129,
      "num_input_tokens_seen": 8861455,
      "step": 414,
      "time_per_iteration": 2.533730983734131
    },
    {
      "auxiliary_loss_clip": 0.01296805,
      "auxiliary_loss_mlp": 0.0104877,
      "balance_loss_clip": 1.08063483,
      "balance_loss_mlp": 1.034024,
      "epoch": 0.04990079961522275,
      "flos": 19828220883840.0,
      "grad_norm": 2.488698298117378,
      "language_loss": 0.80567664,
      "learning_rate": 3.995871410314305e-06,
      "loss": 0.82913238,
      "num_input_tokens_seen": 8880015,
      "step": 415,
      "time_per_iteration": 2.4749958515167236
    },
    {
      "auxiliary_loss_clip": 0.01173741,
      "auxiliary_loss_mlp": 0.01007018,
      "balance_loss_clip": 1.04972339,
      "balance_loss_mlp": 1.0009625,
      "epoch": 0.05002104250586184,
      "flos": 62735045293440.0,
      "grad_norm": 0.9054442068518652,
      "language_loss": 0.59635961,
      "learning_rate": 3.995821232642714e-06,
      "loss": 0.61816722,
      "num_input_tokens_seen": 8938420,
      "step": 416,
      "time_per_iteration": 3.158643960952759
    },
    {
      "auxiliary_loss_clip": 0.01281183,
      "auxiliary_loss_mlp": 0.01051766,
      "balance_loss_clip": 1.08201373,
      "balance_loss_mlp": 1.03673339,
      "epoch": 0.05014128539650093,
      "flos": 27928518710400.0,
      "grad_norm": 2.1831508310459284,
      "language_loss": 0.82651401,
      "learning_rate": 3.995770752206629e-06,
      "loss": 0.8498435,
      "num_input_tokens_seen": 8959495,
      "step": 417,
      "time_per_iteration": 2.61844539642334
    },
    {
      "auxiliary_loss_clip": 0.01300244,
      "auxiliary_loss_mlp": 0.01045056,
      "balance_loss_clip": 1.08212149,
      "balance_loss_mlp": 1.02893901,
      "epoch": 0.05026152828714002,
      "flos": 17705576620800.0,
      "grad_norm": 2.3843534208319035,
      "language_loss": 0.9731673,
      "learning_rate": 3.995719969013709e-06,
      "loss": 0.99662036,
      "num_input_tokens_seen": 8976675,
      "step": 418,
      "time_per_iteration": 2.510394811630249
    },
    {
      "auxiliary_loss_clip": 0.01264372,
      "auxiliary_loss_mlp": 0.010525,
      "balance_loss_clip": 1.07839465,
      "balance_loss_mlp": 1.0364666,
      "epoch": 0.05038177117777912,
      "flos": 19133277477120.0,
      "grad_norm": 3.010087020556891,
      "language_loss": 0.85372448,
      "learning_rate": 3.995668883071655e-06,
      "loss": 0.87689316,
      "num_input_tokens_seen": 8992900,
      "step": 419,
      "time_per_iteration": 2.5459470748901367
    },
    {
      "auxiliary_loss_clip": 0.01300471,
      "auxiliary_loss_mlp": 0.01051948,
      "balance_loss_clip": 1.08223248,
      "balance_loss_mlp": 1.03637969,
      "epoch": 0.050502014068418206,
      "flos": 20667704618880.0,
      "grad_norm": 2.3014059850181887,
      "language_loss": 0.9099651,
      "learning_rate": 3.995617494388219e-06,
      "loss": 0.93348932,
      "num_input_tokens_seen": 9011020,
      "step": 420,
      "time_per_iteration": 2.4808101654052734
    },
    {
      "auxiliary_loss_clip": 0.01261402,
      "auxiliary_loss_mlp": 0.01044376,
      "balance_loss_clip": 1.07446694,
      "balance_loss_mlp": 1.02864623,
      "epoch": 0.050622256959057296,
      "flos": 21361103740800.0,
      "grad_norm": 1.989831026765851,
      "language_loss": 0.80376565,
      "learning_rate": 3.995565802971196e-06,
      "loss": 0.82682347,
      "num_input_tokens_seen": 9030995,
      "step": 421,
      "time_per_iteration": 3.3927552700042725
    },
    {
      "auxiliary_loss_clip": 0.01258431,
      "auxiliary_loss_mlp": 0.01052933,
      "balance_loss_clip": 1.07435346,
      "balance_loss_mlp": 1.03863978,
      "epoch": 0.050742499849696386,
      "flos": 27673588909440.0,
      "grad_norm": 1.925790732415257,
      "language_loss": 0.67646933,
      "learning_rate": 3.995513808828427e-06,
      "loss": 0.69958293,
      "num_input_tokens_seen": 9053790,
      "step": 422,
      "time_per_iteration": 2.646677255630493
    },
    {
      "auxiliary_loss_clip": 0.01260708,
      "auxiliary_loss_mlp": 0.01048243,
      "balance_loss_clip": 1.07567978,
      "balance_loss_mlp": 1.03291273,
      "epoch": 0.050862742740335476,
      "flos": 19865999013120.0,
      "grad_norm": 1.8116825290558,
      "language_loss": 0.76393795,
      "learning_rate": 3.9954615119678e-06,
      "loss": 0.78702748,
      "num_input_tokens_seen": 9072345,
      "step": 423,
      "time_per_iteration": 2.561293363571167
    },
    {
      "auxiliary_loss_clip": 0.01269033,
      "auxiliary_loss_mlp": 0.01056699,
      "balance_loss_clip": 1.07558346,
      "balance_loss_mlp": 1.04059434,
      "epoch": 0.050982985630974566,
      "flos": 22085098272000.0,
      "grad_norm": 2.03521907066593,
      "language_loss": 0.80695105,
      "learning_rate": 3.995408912397248e-06,
      "loss": 0.83020842,
      "num_input_tokens_seen": 9090240,
      "step": 424,
      "time_per_iteration": 2.5259456634521484
    },
    {
      "auxiliary_loss_clip": 0.01266463,
      "auxiliary_loss_mlp": 0.01051782,
      "balance_loss_clip": 1.07880783,
      "balance_loss_mlp": 1.03574848,
      "epoch": 0.05110322852161366,
      "flos": 20740962407040.0,
      "grad_norm": 2.153202176186237,
      "language_loss": 0.93311572,
      "learning_rate": 3.99535601012475e-06,
      "loss": 0.95629811,
      "num_input_tokens_seen": 9105570,
      "step": 425,
      "time_per_iteration": 4.150558233261108
    },
    {
      "auxiliary_loss_clip": 0.01245075,
      "auxiliary_loss_mlp": 0.00766993,
      "balance_loss_clip": 1.07676184,
      "balance_loss_mlp": 1.00095129,
      "epoch": 0.05122347141225275,
      "flos": 28547295327360.0,
      "grad_norm": 1.7456692394336903,
      "language_loss": 0.75426227,
      "learning_rate": 3.995302805158333e-06,
      "loss": 0.77438295,
      "num_input_tokens_seen": 9128225,
      "step": 426,
      "time_per_iteration": 2.6686456203460693
    },
    {
      "auxiliary_loss_clip": 0.01255958,
      "auxiliary_loss_mlp": 0.01051416,
      "balance_loss_clip": 1.07547879,
      "balance_loss_mlp": 1.03403497,
      "epoch": 0.05134371430289184,
      "flos": 19722679747200.0,
      "grad_norm": 1.952348852847772,
      "language_loss": 0.83475518,
      "learning_rate": 3.9952492975060665e-06,
      "loss": 0.85782886,
      "num_input_tokens_seen": 9148295,
      "step": 427,
      "time_per_iteration": 3.3942737579345703
    },
    {
      "auxiliary_loss_clip": 0.01276961,
      "auxiliary_loss_mlp": 0.01041478,
      "balance_loss_clip": 1.07808042,
      "balance_loss_mlp": 1.02670765,
      "epoch": 0.05146395719353093,
      "flos": 34458945649920.0,
      "grad_norm": 2.593341049968763,
      "language_loss": 0.84843683,
      "learning_rate": 3.995195487176067e-06,
      "loss": 0.87162125,
      "num_input_tokens_seen": 9168525,
      "step": 428,
      "time_per_iteration": 2.6617722511291504
    },
    {
      "auxiliary_loss_clip": 0.01295977,
      "auxiliary_loss_mlp": 0.01050163,
      "balance_loss_clip": 1.07999504,
      "balance_loss_mlp": 1.03482068,
      "epoch": 0.05158420008417002,
      "flos": 21760286561280.0,
      "grad_norm": 1.9204342854519867,
      "language_loss": 0.85410869,
      "learning_rate": 3.995141374176499e-06,
      "loss": 0.87757009,
      "num_input_tokens_seen": 9186920,
      "step": 429,
      "time_per_iteration": 2.5337750911712646
    },
    {
      "auxiliary_loss_clip": 0.01143837,
      "auxiliary_loss_mlp": 0.00756563,
      "balance_loss_clip": 1.04659235,
      "balance_loss_mlp": 1.00036418,
      "epoch": 0.05170444297480911,
      "flos": 72553956226560.0,
      "grad_norm": 0.8703592539281128,
      "language_loss": 0.63139963,
      "learning_rate": 3.995086958515572e-06,
      "loss": 0.65040362,
      "num_input_tokens_seen": 9244940,
      "step": 430,
      "time_per_iteration": 3.196502208709717
    },
    {
      "auxiliary_loss_clip": 0.0118113,
      "auxiliary_loss_mlp": 0.007568,
      "balance_loss_clip": 1.04612827,
      "balance_loss_mlp": 1.00034904,
      "epoch": 0.05182468586544821,
      "flos": 62416159326720.0,
      "grad_norm": 0.8555098447788403,
      "language_loss": 0.59934545,
      "learning_rate": 3.995032240201538e-06,
      "loss": 0.61872476,
      "num_input_tokens_seen": 9307335,
      "step": 431,
      "time_per_iteration": 3.0434532165527344
    },
    {
      "auxiliary_loss_clip": 0.01153987,
      "auxiliary_loss_mlp": 0.01007454,
      "balance_loss_clip": 1.03965926,
      "balance_loss_mlp": 1.00194609,
      "epoch": 0.0519449287560873,
      "flos": 41225989432320.0,
      "grad_norm": 0.9454745863448065,
      "language_loss": 0.63146764,
      "learning_rate": 3.9949772192427e-06,
      "loss": 0.65308207,
      "num_input_tokens_seen": 9353960,
      "step": 432,
      "time_per_iteration": 2.7855050563812256
    },
    {
      "auxiliary_loss_clip": 0.01259463,
      "auxiliary_loss_mlp": 0.01047102,
      "balance_loss_clip": 1.07326114,
      "balance_loss_mlp": 1.03160453,
      "epoch": 0.05206517164672639,
      "flos": 17494530261120.0,
      "grad_norm": 1.966456443919025,
      "language_loss": 0.79672104,
      "learning_rate": 3.994921895647405e-06,
      "loss": 0.81978667,
      "num_input_tokens_seen": 9372130,
      "step": 433,
      "time_per_iteration": 2.5267980098724365
    },
    {
      "auxiliary_loss_clip": 0.01176159,
      "auxiliary_loss_mlp": 0.01007939,
      "balance_loss_clip": 1.0428772,
      "balance_loss_mlp": 1.00245512,
      "epoch": 0.05218541453736548,
      "flos": 64002762973440.0,
      "grad_norm": 0.8373059307137443,
      "language_loss": 0.55360562,
      "learning_rate": 3.994866269424043e-06,
      "loss": 0.57544661,
      "num_input_tokens_seen": 9428500,
      "step": 434,
      "time_per_iteration": 2.9598982334136963
    },
    {
      "auxiliary_loss_clip": 0.01201427,
      "auxiliary_loss_mlp": 0.01053205,
      "balance_loss_clip": 1.05995917,
      "balance_loss_mlp": 1.0375762,
      "epoch": 0.05230565742800457,
      "flos": 19317319787520.0,
      "grad_norm": 2.507600517771012,
      "language_loss": 0.78426898,
      "learning_rate": 3.9948103405810545e-06,
      "loss": 0.80681527,
      "num_input_tokens_seen": 9447450,
      "step": 435,
      "time_per_iteration": 2.6395328044891357
    },
    {
      "auxiliary_loss_clip": 0.01230608,
      "auxiliary_loss_mlp": 0.01053478,
      "balance_loss_clip": 1.06978583,
      "balance_loss_mlp": 1.03942299,
      "epoch": 0.05242590031864366,
      "flos": 25298636538240.0,
      "grad_norm": 4.149633701889057,
      "language_loss": 0.86005664,
      "learning_rate": 3.994754109126923e-06,
      "loss": 0.88289744,
      "num_input_tokens_seen": 9468945,
      "step": 436,
      "time_per_iteration": 2.645141839981079
    },
    {
      "auxiliary_loss_clip": 0.0120372,
      "auxiliary_loss_mlp": 0.01043157,
      "balance_loss_clip": 1.06939352,
      "balance_loss_mlp": 1.02864957,
      "epoch": 0.052546143209282754,
      "flos": 26211629456640.0,
      "grad_norm": 1.6682583084824458,
      "language_loss": 0.93285549,
      "learning_rate": 3.994697575070181e-06,
      "loss": 0.95532429,
      "num_input_tokens_seen": 9488405,
      "step": 437,
      "time_per_iteration": 2.8471429347991943
    },
    {
      "auxiliary_loss_clip": 0.01259901,
      "auxiliary_loss_mlp": 0.01054144,
      "balance_loss_clip": 1.07766271,
      "balance_loss_mlp": 1.03888524,
      "epoch": 0.052666386099921844,
      "flos": 22158140578560.0,
      "grad_norm": 1.7520408156355585,
      "language_loss": 0.91556644,
      "learning_rate": 3.994640738419402e-06,
      "loss": 0.93870687,
      "num_input_tokens_seen": 9507780,
      "step": 438,
      "time_per_iteration": 2.873599052429199
    },
    {
      "auxiliary_loss_clip": 0.01274003,
      "auxiliary_loss_mlp": 0.01043856,
      "balance_loss_clip": 1.07774472,
      "balance_loss_mlp": 1.02959228,
      "epoch": 0.052786628990560934,
      "flos": 23881817502720.0,
      "grad_norm": 2.007857097295353,
      "language_loss": 0.80960095,
      "learning_rate": 3.9945835991832075e-06,
      "loss": 0.83277953,
      "num_input_tokens_seen": 9529665,
      "step": 439,
      "time_per_iteration": 2.573204517364502
    },
    {
      "auxiliary_loss_clip": 0.01294579,
      "auxiliary_loss_mlp": 0.01057563,
      "balance_loss_clip": 1.08350658,
      "balance_loss_mlp": 1.04297209,
      "epoch": 0.052906871881200024,
      "flos": 24605021934720.0,
      "grad_norm": 2.7126775079446124,
      "language_loss": 0.92883289,
      "learning_rate": 3.994526157370268e-06,
      "loss": 0.95235425,
      "num_input_tokens_seen": 9548280,
      "step": 440,
      "time_per_iteration": 2.5361196994781494
    },
    {
      "auxiliary_loss_clip": 0.01149993,
      "auxiliary_loss_mlp": 0.01006307,
      "balance_loss_clip": 1.03666568,
      "balance_loss_mlp": 1.00089502,
      "epoch": 0.053027114771839114,
      "flos": 56461631143680.0,
      "grad_norm": 0.8956927470746533,
      "language_loss": 0.59263247,
      "learning_rate": 3.994468412989296e-06,
      "loss": 0.61419547,
      "num_input_tokens_seen": 9609690,
      "step": 441,
      "time_per_iteration": 3.219245672225952
    },
    {
      "auxiliary_loss_clip": 0.01233992,
      "auxiliary_loss_mlp": 0.01049743,
      "balance_loss_clip": 1.06908166,
      "balance_loss_mlp": 1.03502071,
      "epoch": 0.053147357662478203,
      "flos": 17311098481920.0,
      "grad_norm": 2.0762146149015854,
      "language_loss": 0.92690444,
      "learning_rate": 3.994410366049052e-06,
      "loss": 0.94974178,
      "num_input_tokens_seen": 9627550,
      "step": 442,
      "time_per_iteration": 2.536608934402466
    },
    {
      "auxiliary_loss_clip": 0.01272897,
      "auxiliary_loss_mlp": 0.0104361,
      "balance_loss_clip": 1.07576668,
      "balance_loss_mlp": 1.02873874,
      "epoch": 0.0532676005531173,
      "flos": 17164977955200.0,
      "grad_norm": 2.1637109291866286,
      "language_loss": 0.83134085,
      "learning_rate": 3.994352016558341e-06,
      "loss": 0.8545059,
      "num_input_tokens_seen": 9644855,
      "step": 443,
      "time_per_iteration": 2.4811205863952637
    },
    {
      "auxiliary_loss_clip": 0.01274636,
      "auxiliary_loss_mlp": 0.01051717,
      "balance_loss_clip": 1.07829463,
      "balance_loss_mlp": 1.03706574,
      "epoch": 0.05338784344375639,
      "flos": 27819960831360.0,
      "grad_norm": 1.9838096726617758,
      "language_loss": 0.73960245,
      "learning_rate": 3.994293364526014e-06,
      "loss": 0.76286602,
      "num_input_tokens_seen": 9665740,
      "step": 444,
      "time_per_iteration": 2.6801724433898926
    },
    {
      "auxiliary_loss_clip": 0.01249816,
      "auxiliary_loss_mlp": 0.01047337,
      "balance_loss_clip": 1.07561362,
      "balance_loss_mlp": 1.03107703,
      "epoch": 0.05350808633439548,
      "flos": 21507691144320.0,
      "grad_norm": 2.3305621249999358,
      "language_loss": 0.84785217,
      "learning_rate": 3.99423440996097e-06,
      "loss": 0.87082368,
      "num_input_tokens_seen": 9685280,
      "step": 445,
      "time_per_iteration": 2.5926566123962402
    },
    {
      "auxiliary_loss_clip": 0.01259055,
      "auxiliary_loss_mlp": 0.01051048,
      "balance_loss_clip": 1.07913065,
      "balance_loss_mlp": 1.03587234,
      "epoch": 0.05362832922503457,
      "flos": 20084299920000.0,
      "grad_norm": 6.29055922851008,
      "language_loss": 0.81651968,
      "learning_rate": 3.994175152872152e-06,
      "loss": 0.83962071,
      "num_input_tokens_seen": 9704365,
      "step": 446,
      "time_per_iteration": 2.569978713989258
    },
    {
      "auxiliary_loss_clip": 0.01275756,
      "auxiliary_loss_mlp": 0.01041242,
      "balance_loss_clip": 1.07585716,
      "balance_loss_mlp": 1.02719927,
      "epoch": 0.05374857211567366,
      "flos": 26137222433280.0,
      "grad_norm": 2.196225724100749,
      "language_loss": 0.79002571,
      "learning_rate": 3.994115593268548e-06,
      "loss": 0.81319571,
      "num_input_tokens_seen": 9724145,
      "step": 447,
      "time_per_iteration": 3.436283826828003
    },
    {
      "auxiliary_loss_clip": 0.01291874,
      "auxiliary_loss_mlp": 0.01053269,
      "balance_loss_clip": 1.07938385,
      "balance_loss_mlp": 1.03897595,
      "epoch": 0.05386881500631275,
      "flos": 27486817165440.0,
      "grad_norm": 1.9615877816766543,
      "language_loss": 0.82201803,
      "learning_rate": 3.994055731159195e-06,
      "loss": 0.84546947,
      "num_input_tokens_seen": 9741615,
      "step": 448,
      "time_per_iteration": 2.530015230178833
    },
    {
      "auxiliary_loss_clip": 0.01277408,
      "auxiliary_loss_mlp": 0.01060105,
      "balance_loss_clip": 1.08094954,
      "balance_loss_mlp": 1.0459249,
      "epoch": 0.053989057896951846,
      "flos": 23585087249280.0,
      "grad_norm": 1.8797501118212996,
      "language_loss": 0.8712545,
      "learning_rate": 3.993995566553172e-06,
      "loss": 0.8946296,
      "num_input_tokens_seen": 9760580,
      "step": 449,
      "time_per_iteration": 2.6010842323303223
    },
    {
      "auxiliary_loss_clip": 0.01235954,
      "auxiliary_loss_mlp": 0.01046144,
      "balance_loss_clip": 1.06595063,
      "balance_loss_mlp": 1.03150511,
      "epoch": 0.054109300787590936,
      "flos": 25228862369280.0,
      "grad_norm": 1.6976254695350672,
      "language_loss": 0.772946,
      "learning_rate": 3.993935099459607e-06,
      "loss": 0.79576695,
      "num_input_tokens_seen": 9782195,
      "step": 450,
      "time_per_iteration": 2.612295389175415
    },
    {
      "auxiliary_loss_clip": 0.01284598,
      "auxiliary_loss_mlp": 0.01048673,
      "balance_loss_clip": 1.07867634,
      "balance_loss_mlp": 1.03505337,
      "epoch": 0.054229543678230026,
      "flos": 23841525421440.0,
      "grad_norm": 2.1527376349446117,
      "language_loss": 0.74068034,
      "learning_rate": 3.993874329887673e-06,
      "loss": 0.76401305,
      "num_input_tokens_seen": 9800850,
      "step": 451,
      "time_per_iteration": 2.5610320568084717
    },
    {
      "auxiliary_loss_clip": 0.01275216,
      "auxiliary_loss_mlp": 0.01055472,
      "balance_loss_clip": 1.07793713,
      "balance_loss_mlp": 1.04037964,
      "epoch": 0.054349786568869116,
      "flos": 16320933192960.0,
      "grad_norm": 2.43952839340982,
      "language_loss": 0.86216784,
      "learning_rate": 3.993813257846589e-06,
      "loss": 0.88547456,
      "num_input_tokens_seen": 9817605,
      "step": 452,
      "time_per_iteration": 3.3321056365966797
    },
    {
      "auxiliary_loss_clip": 0.01273227,
      "auxiliary_loss_mlp": 0.01045703,
      "balance_loss_clip": 1.07825232,
      "balance_loss_mlp": 1.03110576,
      "epoch": 0.054470029459508205,
      "flos": 18660729127680.0,
      "grad_norm": 2.4540345738589604,
      "language_loss": 0.92340803,
      "learning_rate": 3.993751883345619e-06,
      "loss": 0.94659734,
      "num_input_tokens_seen": 9835965,
      "step": 453,
      "time_per_iteration": 3.3967556953430176
    },
    {
      "auxiliary_loss_clip": 0.01253371,
      "auxiliary_loss_mlp": 0.0105064,
      "balance_loss_clip": 1.0768224,
      "balance_loss_mlp": 1.03570938,
      "epoch": 0.054590272350147295,
      "flos": 17785298856960.0,
      "grad_norm": 2.4492687650873255,
      "language_loss": 0.87209421,
      "learning_rate": 3.993690206394073e-06,
      "loss": 0.89513433,
      "num_input_tokens_seen": 9852265,
      "step": 454,
      "time_per_iteration": 3.352182626724243
    },
    {
      "auxiliary_loss_clip": 0.01260155,
      "auxiliary_loss_mlp": 0.01048789,
      "balance_loss_clip": 1.07582223,
      "balance_loss_mlp": 1.03435254,
      "epoch": 0.054710515240786385,
      "flos": 17785945301760.0,
      "grad_norm": 2.0167613805190956,
      "language_loss": 0.87748712,
      "learning_rate": 3.993628227001307e-06,
      "loss": 0.90057659,
      "num_input_tokens_seen": 9870465,
      "step": 455,
      "time_per_iteration": 2.5466055870056152
    },
    {
      "auxiliary_loss_clip": 0.01255651,
      "auxiliary_loss_mlp": 0.01054282,
      "balance_loss_clip": 1.07518387,
      "balance_loss_mlp": 1.04006076,
      "epoch": 0.05483075813142548,
      "flos": 48210900180480.0,
      "grad_norm": 1.8777234743699365,
      "language_loss": 0.71359921,
      "learning_rate": 3.993565945176726e-06,
      "loss": 0.73669851,
      "num_input_tokens_seen": 9891490,
      "step": 456,
      "time_per_iteration": 2.798192262649536
    },
    {
      "auxiliary_loss_clip": 0.01247937,
      "auxiliary_loss_mlp": 0.01047025,
      "balance_loss_clip": 1.0742749,
      "balance_loss_mlp": 1.03255939,
      "epoch": 0.05495100102206457,
      "flos": 19682244011520.0,
      "grad_norm": 1.958825381402083,
      "language_loss": 0.84441829,
      "learning_rate": 3.993503360929776e-06,
      "loss": 0.86736798,
      "num_input_tokens_seen": 9910375,
      "step": 457,
      "time_per_iteration": 2.5571701526641846
    },
    {
      "auxiliary_loss_clip": 0.01186518,
      "auxiliary_loss_mlp": 0.01048125,
      "balance_loss_clip": 1.06417263,
      "balance_loss_mlp": 1.03300881,
      "epoch": 0.05507124391270366,
      "flos": 26360048453760.0,
      "grad_norm": 1.7567421591039132,
      "language_loss": 0.81142128,
      "learning_rate": 3.99344047426995e-06,
      "loss": 0.83376771,
      "num_input_tokens_seen": 9931635,
      "step": 458,
      "time_per_iteration": 2.8402340412139893
    },
    {
      "auxiliary_loss_clip": 0.01225332,
      "auxiliary_loss_mlp": 0.01049963,
      "balance_loss_clip": 1.06929886,
      "balance_loss_mlp": 1.03465652,
      "epoch": 0.05519148680334275,
      "flos": 22601314581120.0,
      "grad_norm": 2.104397228739993,
      "language_loss": 0.93625522,
      "learning_rate": 3.993377285206789e-06,
      "loss": 0.95900822,
      "num_input_tokens_seen": 9951420,
      "step": 459,
      "time_per_iteration": 2.9285833835601807
    },
    {
      "auxiliary_loss_clip": 0.01216721,
      "auxiliary_loss_mlp": 0.01056765,
      "balance_loss_clip": 1.06795645,
      "balance_loss_mlp": 1.04189968,
      "epoch": 0.05531172969398184,
      "flos": 40552519380480.0,
      "grad_norm": 1.7810400891534426,
      "language_loss": 0.86242628,
      "learning_rate": 3.99331379374988e-06,
      "loss": 0.8851611,
      "num_input_tokens_seen": 9975025,
      "step": 460,
      "time_per_iteration": 2.7733118534088135
    },
    {
      "auxiliary_loss_clip": 0.01260238,
      "auxiliary_loss_mlp": 0.01044803,
      "balance_loss_clip": 1.07015657,
      "balance_loss_mlp": 1.03131461,
      "epoch": 0.05543197258462093,
      "flos": 23477894087040.0,
      "grad_norm": 2.0558160232231293,
      "language_loss": 0.80142117,
      "learning_rate": 3.993249999908852e-06,
      "loss": 0.82447153,
      "num_input_tokens_seen": 9995175,
      "step": 461,
      "time_per_iteration": 2.61435604095459
    },
    {
      "auxiliary_loss_clip": 0.01287158,
      "auxiliary_loss_mlp": 0.01047705,
      "balance_loss_clip": 1.07595766,
      "balance_loss_mlp": 1.03390098,
      "epoch": 0.05555221547526003,
      "flos": 18624603024000.0,
      "grad_norm": 2.0611482450465903,
      "language_loss": 0.87428278,
      "learning_rate": 3.993185903693384e-06,
      "loss": 0.89763141,
      "num_input_tokens_seen": 10011975,
      "step": 462,
      "time_per_iteration": 2.4678704738616943
    },
    {
      "auxiliary_loss_clip": 0.0125292,
      "auxiliary_loss_mlp": 0.01039594,
      "balance_loss_clip": 1.07375169,
      "balance_loss_mlp": 1.02593231,
      "epoch": 0.05567245836589912,
      "flos": 23587098410880.0,
      "grad_norm": 2.2576931871790533,
      "language_loss": 0.82325077,
      "learning_rate": 3.9931215051131995e-06,
      "loss": 0.84617591,
      "num_input_tokens_seen": 10032620,
      "step": 463,
      "time_per_iteration": 2.614413022994995
    },
    {
      "auxiliary_loss_clip": 0.01256273,
      "auxiliary_loss_mlp": 0.01046013,
      "balance_loss_clip": 1.06942487,
      "balance_loss_mlp": 1.03173161,
      "epoch": 0.05579270125653821,
      "flos": 27746667129600.0,
      "grad_norm": 1.5333354564677812,
      "language_loss": 0.80077308,
      "learning_rate": 3.993056804178068e-06,
      "loss": 0.82379591,
      "num_input_tokens_seen": 10054165,
      "step": 464,
      "time_per_iteration": 2.625532865524292
    },
    {
      "auxiliary_loss_clip": 0.01215156,
      "auxiliary_loss_mlp": 0.01045022,
      "balance_loss_clip": 1.06792319,
      "balance_loss_mlp": 1.03006136,
      "epoch": 0.0559129441471773,
      "flos": 27014161075200.0,
      "grad_norm": 2.051909912610453,
      "language_loss": 0.84421498,
      "learning_rate": 3.992991800897803e-06,
      "loss": 0.86681676,
      "num_input_tokens_seen": 10073970,
      "step": 465,
      "time_per_iteration": 2.793992757797241
    },
    {
      "auxiliary_loss_clip": 0.01286037,
      "auxiliary_loss_mlp": 0.01047208,
      "balance_loss_clip": 1.07677484,
      "balance_loss_mlp": 1.03202069,
      "epoch": 0.05603318703781639,
      "flos": 15229787794560.0,
      "grad_norm": 2.3041596877673123,
      "language_loss": 0.89988774,
      "learning_rate": 3.9929264952822665e-06,
      "loss": 0.92322022,
      "num_input_tokens_seen": 10091505,
      "step": 466,
      "time_per_iteration": 2.5002784729003906
    },
    {
      "auxiliary_loss_clip": 0.0127289,
      "auxiliary_loss_mlp": 0.01050952,
      "balance_loss_clip": 1.07353401,
      "balance_loss_mlp": 1.03664696,
      "epoch": 0.05615342992845548,
      "flos": 22266482976000.0,
      "grad_norm": 2.3220236947354196,
      "language_loss": 0.88353026,
      "learning_rate": 3.992860887341366e-06,
      "loss": 0.90676868,
      "num_input_tokens_seen": 10109675,
      "step": 467,
      "time_per_iteration": 2.573352575302124
    },
    {
      "auxiliary_loss_clip": 0.01225384,
      "auxiliary_loss_mlp": 0.01042764,
      "balance_loss_clip": 1.06847656,
      "balance_loss_mlp": 1.02748108,
      "epoch": 0.056273672819094574,
      "flos": 23584979508480.0,
      "grad_norm": 2.2303254393803043,
      "language_loss": 0.81188333,
      "learning_rate": 3.992794977085052e-06,
      "loss": 0.83456481,
      "num_input_tokens_seen": 10127675,
      "step": 468,
      "time_per_iteration": 2.624342679977417
    },
    {
      "auxiliary_loss_clip": 0.01241007,
      "auxiliary_loss_mlp": 0.01053444,
      "balance_loss_clip": 1.0728451,
      "balance_loss_mlp": 1.03932929,
      "epoch": 0.056393915709733664,
      "flos": 19858708552320.0,
      "grad_norm": 2.1551598959605265,
      "language_loss": 0.84982693,
      "learning_rate": 3.992728764523326e-06,
      "loss": 0.87277138,
      "num_input_tokens_seen": 10146620,
      "step": 469,
      "time_per_iteration": 2.6475706100463867
    },
    {
      "auxiliary_loss_clip": 0.01254025,
      "auxiliary_loss_mlp": 0.01047833,
      "balance_loss_clip": 1.07240641,
      "balance_loss_mlp": 1.03345656,
      "epoch": 0.05651415860037275,
      "flos": 22163779013760.0,
      "grad_norm": 2.1054534597155476,
      "language_loss": 0.80992544,
      "learning_rate": 3.99266224966623e-06,
      "loss": 0.8329441,
      "num_input_tokens_seen": 10167535,
      "step": 470,
      "time_per_iteration": 2.619994640350342
    },
    {
      "auxiliary_loss_clip": 0.01244018,
      "auxiliary_loss_mlp": 0.01046113,
      "balance_loss_clip": 1.07288432,
      "balance_loss_mlp": 1.03129482,
      "epoch": 0.05663440149101184,
      "flos": 19463548055040.0,
      "grad_norm": 1.9288716141113582,
      "language_loss": 0.87809944,
      "learning_rate": 3.992595432523855e-06,
      "loss": 0.90100068,
      "num_input_tokens_seen": 10184825,
      "step": 471,
      "time_per_iteration": 2.569845676422119
    },
    {
      "auxiliary_loss_clip": 0.01227736,
      "auxiliary_loss_mlp": 0.01050565,
      "balance_loss_clip": 1.06888604,
      "balance_loss_mlp": 1.03649819,
      "epoch": 0.05675464438165093,
      "flos": 22670226823680.0,
      "grad_norm": 1.9535716613615486,
      "language_loss": 0.86224347,
      "learning_rate": 3.992528313106338e-06,
      "loss": 0.88502645,
      "num_input_tokens_seen": 10203025,
      "step": 472,
      "time_per_iteration": 2.633270502090454
    },
    {
      "auxiliary_loss_clip": 0.0128748,
      "auxiliary_loss_mlp": 0.00766872,
      "balance_loss_clip": 1.08005643,
      "balance_loss_mlp": 1.00075042,
      "epoch": 0.05687488727229002,
      "flos": 16901177495040.0,
      "grad_norm": 2.305118764708682,
      "language_loss": 0.81970888,
      "learning_rate": 3.9924608914238595e-06,
      "loss": 0.8402524,
      "num_input_tokens_seen": 10218020,
      "step": 473,
      "time_per_iteration": 2.520453691482544
    },
    {
      "auxiliary_loss_clip": 0.01270703,
      "auxiliary_loss_mlp": 0.01049679,
      "balance_loss_clip": 1.07682896,
      "balance_loss_mlp": 1.03543305,
      "epoch": 0.05699513016292912,
      "flos": 29168980945920.0,
      "grad_norm": 2.2892678541469316,
      "language_loss": 0.84009999,
      "learning_rate": 3.992393167486648e-06,
      "loss": 0.86330378,
      "num_input_tokens_seen": 10237170,
      "step": 474,
      "time_per_iteration": 3.406655788421631
    },
    {
      "auxiliary_loss_clip": 0.01289301,
      "auxiliary_loss_mlp": 0.01055027,
      "balance_loss_clip": 1.07881534,
      "balance_loss_mlp": 1.03964925,
      "epoch": 0.05711537305356821,
      "flos": 18916197632640.0,
      "grad_norm": 2.5701574061421413,
      "language_loss": 0.80658674,
      "learning_rate": 3.992325141304977e-06,
      "loss": 0.83002996,
      "num_input_tokens_seen": 10255125,
      "step": 475,
      "time_per_iteration": 2.5102367401123047
    },
    {
      "auxiliary_loss_clip": 0.01224152,
      "auxiliary_loss_mlp": 0.01048825,
      "balance_loss_clip": 1.06943631,
      "balance_loss_mlp": 1.03480554,
      "epoch": 0.0572356159442073,
      "flos": 26758979879040.0,
      "grad_norm": 2.231398342184023,
      "language_loss": 0.86522353,
      "learning_rate": 3.992256812889166e-06,
      "loss": 0.88795334,
      "num_input_tokens_seen": 10271230,
      "step": 476,
      "time_per_iteration": 2.6430747509002686
    },
    {
      "auxiliary_loss_clip": 0.01287087,
      "auxiliary_loss_mlp": 0.01047905,
      "balance_loss_clip": 1.07895398,
      "balance_loss_mlp": 1.03381395,
      "epoch": 0.05735585883484639,
      "flos": 35116146840960.0,
      "grad_norm": 4.912103906002928,
      "language_loss": 0.76712555,
      "learning_rate": 3.992188182249582e-06,
      "loss": 0.79047549,
      "num_input_tokens_seen": 10293125,
      "step": 477,
      "time_per_iteration": 2.6121628284454346
    },
    {
      "auxiliary_loss_clip": 0.01252452,
      "auxiliary_loss_mlp": 0.01055691,
      "balance_loss_clip": 1.0760603,
      "balance_loss_mlp": 1.04084325,
      "epoch": 0.05747610172548548,
      "flos": 18734381965440.0,
      "grad_norm": 7.740941948675379,
      "language_loss": 0.90576172,
      "learning_rate": 3.992119249396633e-06,
      "loss": 0.92884308,
      "num_input_tokens_seen": 10311810,
      "step": 478,
      "time_per_iteration": 2.526930332183838
    },
    {
      "auxiliary_loss_clip": 0.01243971,
      "auxiliary_loss_mlp": 0.00766679,
      "balance_loss_clip": 1.06933284,
      "balance_loss_mlp": 1.000741,
      "epoch": 0.05759634461612457,
      "flos": 27964752554880.0,
      "grad_norm": 1.9002993415101694,
      "language_loss": 0.82179916,
      "learning_rate": 3.992050014340778e-06,
      "loss": 0.84190571,
      "num_input_tokens_seen": 10332165,
      "step": 479,
      "time_per_iteration": 3.464747190475464
    },
    {
      "auxiliary_loss_clip": 0.01147847,
      "auxiliary_loss_mlp": 0.01009889,
      "balance_loss_clip": 1.03669572,
      "balance_loss_mlp": 1.00461996,
      "epoch": 0.057716587506763666,
      "flos": 69292009405440.0,
      "grad_norm": 0.8640049590043475,
      "language_loss": 0.55069149,
      "learning_rate": 3.99198047709252e-06,
      "loss": 0.57226884,
      "num_input_tokens_seen": 10393685,
      "step": 480,
      "time_per_iteration": 3.161722183227539
    },
    {
      "auxiliary_loss_clip": 0.01232759,
      "auxiliary_loss_mlp": 0.0105072,
      "balance_loss_clip": 1.06522787,
      "balance_loss_mlp": 1.03571713,
      "epoch": 0.057836830397402755,
      "flos": 25009196745600.0,
      "grad_norm": 1.8616834802833198,
      "language_loss": 0.78794473,
      "learning_rate": 3.991910637662408e-06,
      "loss": 0.81077951,
      "num_input_tokens_seen": 10413975,
      "step": 481,
      "time_per_iteration": 3.4052248001098633
    },
    {
      "auxiliary_loss_clip": 0.01285052,
      "auxiliary_loss_mlp": 0.01040041,
      "balance_loss_clip": 1.07858062,
      "balance_loss_mlp": 1.02562273,
      "epoch": 0.057957073288041845,
      "flos": 25593894334080.0,
      "grad_norm": 1.8138002061911072,
      "language_loss": 0.80608875,
      "learning_rate": 3.9918404960610355e-06,
      "loss": 0.82933968,
      "num_input_tokens_seen": 10433005,
      "step": 482,
      "time_per_iteration": 2.5188021659851074
    },
    {
      "auxiliary_loss_clip": 0.01277709,
      "auxiliary_loss_mlp": 0.01051611,
      "balance_loss_clip": 1.07918572,
      "balance_loss_mlp": 1.03740668,
      "epoch": 0.058077316178680935,
      "flos": 20777411733120.0,
      "grad_norm": 2.261458503024001,
      "language_loss": 0.77691972,
      "learning_rate": 3.991770052299043e-06,
      "loss": 0.80021298,
      "num_input_tokens_seen": 10451235,
      "step": 483,
      "time_per_iteration": 2.569852828979492
    },
    {
      "auxiliary_loss_clip": 0.01250886,
      "auxiliary_loss_mlp": 0.010403,
      "balance_loss_clip": 1.07015252,
      "balance_loss_mlp": 1.0271039,
      "epoch": 0.058197559069320025,
      "flos": 18916484941440.0,
      "grad_norm": 2.1821964473996855,
      "language_loss": 0.87601757,
      "learning_rate": 3.991699306387118e-06,
      "loss": 0.89892948,
      "num_input_tokens_seen": 10469705,
      "step": 484,
      "time_per_iteration": 2.537954330444336
    },
    {
      "auxiliary_loss_clip": 0.01269438,
      "auxiliary_loss_mlp": 0.01052509,
      "balance_loss_clip": 1.07506013,
      "balance_loss_mlp": 1.03837669,
      "epoch": 0.058317801959959115,
      "flos": 24863327614080.0,
      "grad_norm": 1.8435456857959243,
      "language_loss": 0.78118783,
      "learning_rate": 3.991628258335991e-06,
      "loss": 0.8044073,
      "num_input_tokens_seen": 10491910,
      "step": 485,
      "time_per_iteration": 2.5575599670410156
    },
    {
      "auxiliary_loss_clip": 0.01228738,
      "auxiliary_loss_mlp": 0.01045503,
      "balance_loss_clip": 1.06771517,
      "balance_loss_mlp": 1.03119779,
      "epoch": 0.05843804485059821,
      "flos": 23257977068160.0,
      "grad_norm": 3.9403707254477287,
      "language_loss": 0.8729986,
      "learning_rate": 3.991556908156442e-06,
      "loss": 0.89574105,
      "num_input_tokens_seen": 10508435,
      "step": 486,
      "time_per_iteration": 2.5650887489318848
    },
    {
      "auxiliary_loss_clip": 0.01256629,
      "auxiliary_loss_mlp": 0.01053806,
      "balance_loss_clip": 1.07344568,
      "balance_loss_mlp": 1.03997207,
      "epoch": 0.0585582877412373,
      "flos": 23150532510720.0,
      "grad_norm": 2.5836211887872502,
      "language_loss": 0.87835133,
      "learning_rate": 3.9914852558592914e-06,
      "loss": 0.90145564,
      "num_input_tokens_seen": 10529485,
      "step": 487,
      "time_per_iteration": 2.5938074588775635
    },
    {
      "auxiliary_loss_clip": 0.01269577,
      "auxiliary_loss_mlp": 0.01045475,
      "balance_loss_clip": 1.0771538,
      "balance_loss_mlp": 1.03089583,
      "epoch": 0.05867853063187639,
      "flos": 23506406507520.0,
      "grad_norm": 3.245791908158948,
      "language_loss": 0.80608124,
      "learning_rate": 3.991413301455413e-06,
      "loss": 0.82923174,
      "num_input_tokens_seen": 10545935,
      "step": 488,
      "time_per_iteration": 2.5468804836273193
    },
    {
      "auxiliary_loss_clip": 0.01237281,
      "auxiliary_loss_mlp": 0.01043722,
      "balance_loss_clip": 1.06981635,
      "balance_loss_mlp": 1.0305438,
      "epoch": 0.05879877352251548,
      "flos": 29495803818240.0,
      "grad_norm": 2.8241493027910707,
      "language_loss": 0.78130937,
      "learning_rate": 3.991341044955719e-06,
      "loss": 0.80411935,
      "num_input_tokens_seen": 10565690,
      "step": 489,
      "time_per_iteration": 2.7312850952148438
    },
    {
      "auxiliary_loss_clip": 0.01265329,
      "auxiliary_loss_mlp": 0.00767242,
      "balance_loss_clip": 1.0730257,
      "balance_loss_mlp": 1.00077808,
      "epoch": 0.05891901641315457,
      "flos": 20157485880960.0,
      "grad_norm": 1.960062980502636,
      "language_loss": 0.81499922,
      "learning_rate": 3.991268486371172e-06,
      "loss": 0.83532488,
      "num_input_tokens_seen": 10584245,
      "step": 490,
      "time_per_iteration": 2.597562074661255
    },
    {
      "auxiliary_loss_clip": 0.01251835,
      "auxiliary_loss_mlp": 0.01052381,
      "balance_loss_clip": 1.07094288,
      "balance_loss_mlp": 1.03585863,
      "epoch": 0.05903925930379366,
      "flos": 24644200694400.0,
      "grad_norm": 2.4332300998667087,
      "language_loss": 0.87856263,
      "learning_rate": 3.991195625712779e-06,
      "loss": 0.90160477,
      "num_input_tokens_seen": 10601210,
      "step": 491,
      "time_per_iteration": 2.646657943725586
    },
    {
      "auxiliary_loss_clip": 0.01283379,
      "auxiliary_loss_mlp": 0.01044919,
      "balance_loss_clip": 1.07858241,
      "balance_loss_mlp": 1.03049469,
      "epoch": 0.05915950219443276,
      "flos": 21250391045760.0,
      "grad_norm": 1.9560043557368332,
      "language_loss": 0.81519854,
      "learning_rate": 3.991122462991592e-06,
      "loss": 0.83848155,
      "num_input_tokens_seen": 10620730,
      "step": 492,
      "time_per_iteration": 2.55846905708313
    },
    {
      "auxiliary_loss_clip": 0.01288231,
      "auxiliary_loss_mlp": 0.01050067,
      "balance_loss_clip": 1.07664061,
      "balance_loss_mlp": 1.03595257,
      "epoch": 0.05927974508507185,
      "flos": 9902727319680.0,
      "grad_norm": 4.138479272791149,
      "language_loss": 0.81250608,
      "learning_rate": 3.991048998218712e-06,
      "loss": 0.8358891,
      "num_input_tokens_seen": 10634035,
      "step": 493,
      "time_per_iteration": 2.4798851013183594
    },
    {
      "auxiliary_loss_clip": 0.01265389,
      "auxiliary_loss_mlp": 0.01050066,
      "balance_loss_clip": 1.07141185,
      "balance_loss_mlp": 1.0360117,
      "epoch": 0.05939998797571094,
      "flos": 18259499232000.0,
      "grad_norm": 2.6924464145983067,
      "language_loss": 0.76272655,
      "learning_rate": 3.990975231405281e-06,
      "loss": 0.78588104,
      "num_input_tokens_seen": 10652485,
      "step": 494,
      "time_per_iteration": 2.540430784225464
    },
    {
      "auxiliary_loss_clip": 0.01264716,
      "auxiliary_loss_mlp": 0.01046595,
      "balance_loss_clip": 1.0755918,
      "balance_loss_mlp": 1.03237343,
      "epoch": 0.05952023086635003,
      "flos": 28256598558720.0,
      "grad_norm": 1.7629613226632177,
      "language_loss": 0.78670263,
      "learning_rate": 3.990901162562491e-06,
      "loss": 0.80981576,
      "num_input_tokens_seen": 10673175,
      "step": 495,
      "time_per_iteration": 2.63199782371521
    },
    {
      "auxiliary_loss_clip": 0.01227425,
      "auxiliary_loss_mlp": 0.00767761,
      "balance_loss_clip": 1.06459618,
      "balance_loss_mlp": 1.00068176,
      "epoch": 0.05964047375698912,
      "flos": 14902498045440.0,
      "grad_norm": 2.473320868503803,
      "language_loss": 0.90764403,
      "learning_rate": 3.9908267917015765e-06,
      "loss": 0.92759585,
      "num_input_tokens_seen": 10691235,
      "step": 496,
      "time_per_iteration": 2.6073436737060547
    },
    {
      "auxiliary_loss_clip": 0.01254178,
      "auxiliary_loss_mlp": 0.01055666,
      "balance_loss_clip": 1.06982493,
      "balance_loss_mlp": 1.04096711,
      "epoch": 0.059760716647628206,
      "flos": 23185581206400.0,
      "grad_norm": 2.4768764183403613,
      "language_loss": 0.93161809,
      "learning_rate": 3.990752118833821e-06,
      "loss": 0.95471656,
      "num_input_tokens_seen": 10708675,
      "step": 497,
      "time_per_iteration": 2.5714454650878906
    },
    {
      "auxiliary_loss_clip": 0.01283475,
      "auxiliary_loss_mlp": 0.01045611,
      "balance_loss_clip": 1.07747734,
      "balance_loss_mlp": 1.03153205,
      "epoch": 0.0598809595382673,
      "flos": 22746968231040.0,
      "grad_norm": 1.816147997487087,
      "language_loss": 0.78227592,
      "learning_rate": 3.990677143970553e-06,
      "loss": 0.80556679,
      "num_input_tokens_seen": 10729485,
      "step": 498,
      "time_per_iteration": 2.589621067047119
    },
    {
      "auxiliary_loss_clip": 0.01230367,
      "auxiliary_loss_mlp": 0.01055318,
      "balance_loss_clip": 1.07252336,
      "balance_loss_mlp": 1.0402143,
      "epoch": 0.06000120242890639,
      "flos": 22127221946880.0,
      "grad_norm": 2.3668604459595377,
      "language_loss": 0.81221884,
      "learning_rate": 3.990601867123144e-06,
      "loss": 0.83507574,
      "num_input_tokens_seen": 10749210,
      "step": 499,
      "time_per_iteration": 2.6350436210632324
    },
    {
      "auxiliary_loss_clip": 0.0121815,
      "auxiliary_loss_mlp": 0.01049183,
      "balance_loss_clip": 1.06963706,
      "balance_loss_mlp": 1.03477049,
      "epoch": 0.06012144531954548,
      "flos": 19171773878400.0,
      "grad_norm": 2.065752786043113,
      "language_loss": 0.84776658,
      "learning_rate": 3.990526288303014e-06,
      "loss": 0.87043989,
      "num_input_tokens_seen": 10768000,
      "step": 500,
      "time_per_iteration": 2.630920171737671
    },
    {
      "auxiliary_loss_clip": 0.01245623,
      "auxiliary_loss_mlp": 0.00766268,
      "balance_loss_clip": 1.07011294,
      "balance_loss_mlp": 1.00060201,
      "epoch": 0.06024168821018457,
      "flos": 22783345729920.0,
      "grad_norm": 1.711959340986966,
      "language_loss": 0.91168308,
      "learning_rate": 3.9904504075216295e-06,
      "loss": 0.93180203,
      "num_input_tokens_seen": 10788760,
      "step": 501,
      "time_per_iteration": 3.4149577617645264
    },
    {
      "auxiliary_loss_clip": 0.01232014,
      "auxiliary_loss_mlp": 0.01051664,
      "balance_loss_clip": 1.06641507,
      "balance_loss_mlp": 1.03676319,
      "epoch": 0.06036193110082366,
      "flos": 18770687637120.0,
      "grad_norm": 2.0796848827466285,
      "language_loss": 0.94184071,
      "learning_rate": 3.990374224790501e-06,
      "loss": 0.96467751,
      "num_input_tokens_seen": 10806965,
      "step": 502,
      "time_per_iteration": 2.604987621307373
    },
    {
      "auxiliary_loss_clip": 0.01248037,
      "auxiliary_loss_mlp": 0.01052614,
      "balance_loss_clip": 1.07285142,
      "balance_loss_mlp": 1.03800464,
      "epoch": 0.06048217399146275,
      "flos": 17201570935680.0,
      "grad_norm": 1.8303229565552828,
      "language_loss": 0.70841467,
      "learning_rate": 3.990297740121185e-06,
      "loss": 0.73142111,
      "num_input_tokens_seen": 10824900,
      "step": 503,
      "time_per_iteration": 2.5631203651428223
    },
    {
      "auxiliary_loss_clip": 0.0126291,
      "auxiliary_loss_mlp": 0.00766852,
      "balance_loss_clip": 1.07250094,
      "balance_loss_mlp": 1.00057173,
      "epoch": 0.06060241688210185,
      "flos": 24024131187840.0,
      "grad_norm": 1.7632762460295015,
      "language_loss": 0.78359771,
      "learning_rate": 3.990220953525284e-06,
      "loss": 0.80389529,
      "num_input_tokens_seen": 10842010,
      "step": 504,
      "time_per_iteration": 2.5622775554656982
    },
    {
      "auxiliary_loss_clip": 0.01236032,
      "auxiliary_loss_mlp": 0.01049064,
      "balance_loss_clip": 1.06633329,
      "balance_loss_mlp": 1.03552783,
      "epoch": 0.06072265977274094,
      "flos": 14611190745600.0,
      "grad_norm": 3.417936454705504,
      "language_loss": 0.74035621,
      "learning_rate": 3.9901438650144465e-06,
      "loss": 0.76320714,
      "num_input_tokens_seen": 10858260,
      "step": 505,
      "time_per_iteration": 3.3221349716186523
    },
    {
      "auxiliary_loss_clip": 0.01255656,
      "auxiliary_loss_mlp": 0.01044378,
      "balance_loss_clip": 1.07094288,
      "balance_loss_mlp": 1.03138399,
      "epoch": 0.06084290266338003,
      "flos": 20558284813440.0,
      "grad_norm": 4.427473181387212,
      "language_loss": 0.91775775,
      "learning_rate": 3.990066474600367e-06,
      "loss": 0.94075811,
      "num_input_tokens_seen": 10876230,
      "step": 506,
      "time_per_iteration": 3.4471898078918457
    },
    {
      "auxiliary_loss_clip": 0.01248885,
      "auxiliary_loss_mlp": 0.01047921,
      "balance_loss_clip": 1.06613219,
      "balance_loss_mlp": 1.03276944,
      "epoch": 0.06096314555401912,
      "flos": 22309217182080.0,
      "grad_norm": 2.097791928212606,
      "language_loss": 0.67857343,
      "learning_rate": 3.989988782294786e-06,
      "loss": 0.70154148,
      "num_input_tokens_seen": 10896320,
      "step": 507,
      "time_per_iteration": 2.5640347003936768
    },
    {
      "auxiliary_loss_clip": 0.01216106,
      "auxiliary_loss_mlp": 0.01053604,
      "balance_loss_clip": 1.06556737,
      "balance_loss_mlp": 1.03891158,
      "epoch": 0.06108338844465821,
      "flos": 19131374056320.0,
      "grad_norm": 1.7265108818948975,
      "language_loss": 0.95145917,
      "learning_rate": 3.989910788109489e-06,
      "loss": 0.97415626,
      "num_input_tokens_seen": 10912970,
      "step": 508,
      "time_per_iteration": 3.319075584411621
    },
    {
      "auxiliary_loss_clip": 0.01225846,
      "auxiliary_loss_mlp": 0.01045771,
      "balance_loss_clip": 1.06460392,
      "balance_loss_mlp": 1.03205013,
      "epoch": 0.0612036313352973,
      "flos": 33584018169600.0,
      "grad_norm": 4.653434425353272,
      "language_loss": 0.75043541,
      "learning_rate": 3.989832492056307e-06,
      "loss": 0.77315158,
      "num_input_tokens_seen": 10933995,
      "step": 509,
      "time_per_iteration": 2.7116730213165283
    },
    {
      "auxiliary_loss_clip": 0.0126432,
      "auxiliary_loss_mlp": 0.01049766,
      "balance_loss_clip": 1.07376266,
      "balance_loss_mlp": 1.03516316,
      "epoch": 0.06132387422593639,
      "flos": 27490552179840.0,
      "grad_norm": 2.30611056388382,
      "language_loss": 0.8086307,
      "learning_rate": 3.989753894147119e-06,
      "loss": 0.83177161,
      "num_input_tokens_seen": 10954120,
      "step": 510,
      "time_per_iteration": 2.5530357360839844
    },
    {
      "auxiliary_loss_clip": 0.01258622,
      "auxiliary_loss_mlp": 0.01046003,
      "balance_loss_clip": 1.07547843,
      "balance_loss_mlp": 1.03217459,
      "epoch": 0.061444117116575485,
      "flos": 25885057979520.0,
      "grad_norm": 1.8684002224500142,
      "language_loss": 0.79945993,
      "learning_rate": 3.989674994393846e-06,
      "loss": 0.82250619,
      "num_input_tokens_seen": 10973595,
      "step": 511,
      "time_per_iteration": 2.58402943611145
    },
    {
      "auxiliary_loss_clip": 0.01262331,
      "auxiliary_loss_mlp": 0.01041539,
      "balance_loss_clip": 1.07509351,
      "balance_loss_mlp": 1.02762151,
      "epoch": 0.061564360007214575,
      "flos": 28512031150080.0,
      "grad_norm": 2.1365770111365485,
      "language_loss": 0.94072628,
      "learning_rate": 3.98959579280846e-06,
      "loss": 0.96376503,
      "num_input_tokens_seen": 10991995,
      "step": 512,
      "time_per_iteration": 2.5635733604431152
    },
    {
      "auxiliary_loss_clip": 0.01196261,
      "auxiliary_loss_mlp": 0.01045152,
      "balance_loss_clip": 1.06744456,
      "balance_loss_mlp": 1.03076339,
      "epoch": 0.061684602897853665,
      "flos": 12094355652480.0,
      "grad_norm": 2.2795350665917007,
      "language_loss": 0.83073515,
      "learning_rate": 3.989516289402973e-06,
      "loss": 0.85314929,
      "num_input_tokens_seen": 11007625,
      "step": 513,
      "time_per_iteration": 2.649857759475708
    },
    {
      "auxiliary_loss_clip": 0.01177261,
      "auxiliary_loss_mlp": 0.01048162,
      "balance_loss_clip": 1.0571543,
      "balance_loss_mlp": 1.03379762,
      "epoch": 0.061804845788492754,
      "flos": 19532639865600.0,
      "grad_norm": 2.8027560640377995,
      "language_loss": 0.80464005,
      "learning_rate": 3.989436484189447e-06,
      "loss": 0.82689428,
      "num_input_tokens_seen": 11025570,
      "step": 514,
      "time_per_iteration": 2.6532979011535645
    },
    {
      "auxiliary_loss_clip": 0.01263404,
      "auxiliary_loss_mlp": 0.01041708,
      "balance_loss_clip": 1.06956124,
      "balance_loss_mlp": 1.02771306,
      "epoch": 0.061925088679131844,
      "flos": 15341111020800.0,
      "grad_norm": 2.9314761155770612,
      "language_loss": 0.80997372,
      "learning_rate": 3.9893563771799885e-06,
      "loss": 0.83302486,
      "num_input_tokens_seen": 11042045,
      "step": 515,
      "time_per_iteration": 2.526691198348999
    },
    {
      "auxiliary_loss_clip": 0.01279233,
      "auxiliary_loss_mlp": 0.01048377,
      "balance_loss_clip": 1.07494998,
      "balance_loss_mlp": 1.03396511,
      "epoch": 0.062045331569770934,
      "flos": 25919927107200.0,
      "grad_norm": 2.3013100180536035,
      "language_loss": 0.85945821,
      "learning_rate": 3.989275968386749e-06,
      "loss": 0.8827343,
      "num_input_tokens_seen": 11059955,
      "step": 516,
      "time_per_iteration": 2.540703773498535
    },
    {
      "auxiliary_loss_clip": 0.01238818,
      "auxiliary_loss_mlp": 0.01051251,
      "balance_loss_clip": 1.06632495,
      "balance_loss_mlp": 1.03590882,
      "epoch": 0.06216557446041003,
      "flos": 28110621686400.0,
      "grad_norm": 2.2433027139282955,
      "language_loss": 0.76641953,
      "learning_rate": 3.989195257821926e-06,
      "loss": 0.78932023,
      "num_input_tokens_seen": 11078440,
      "step": 517,
      "time_per_iteration": 2.687325954437256
    },
    {
      "auxiliary_loss_clip": 0.01242201,
      "auxiliary_loss_mlp": 0.01050637,
      "balance_loss_clip": 1.07084441,
      "balance_loss_mlp": 1.03628373,
      "epoch": 0.06228581735104912,
      "flos": 23478181395840.0,
      "grad_norm": 2.1961413645515186,
      "language_loss": 0.84141314,
      "learning_rate": 3.989114245497765e-06,
      "loss": 0.86434156,
      "num_input_tokens_seen": 11098240,
      "step": 518,
      "time_per_iteration": 2.602976083755493
    },
    {
      "auxiliary_loss_clip": 0.01261688,
      "auxiliary_loss_mlp": 0.0104308,
      "balance_loss_clip": 1.06707501,
      "balance_loss_mlp": 1.02932966,
      "epoch": 0.06240606024168821,
      "flos": 15195205975680.0,
      "grad_norm": 2.5054984938972042,
      "language_loss": 0.94887328,
      "learning_rate": 3.989032931426554e-06,
      "loss": 0.97192097,
      "num_input_tokens_seen": 11115395,
      "step": 519,
      "time_per_iteration": 2.5378425121307373
    },
    {
      "auxiliary_loss_clip": 0.01237419,
      "auxiliary_loss_mlp": 0.01044545,
      "balance_loss_clip": 1.06759143,
      "balance_loss_mlp": 1.03063893,
      "epoch": 0.06252630313232731,
      "flos": 20631829910400.0,
      "grad_norm": 2.209568759441187,
      "language_loss": 0.86931837,
      "learning_rate": 3.9889513156206295e-06,
      "loss": 0.892138,
      "num_input_tokens_seen": 11134835,
      "step": 520,
      "time_per_iteration": 2.536693572998047
    },
    {
      "auxiliary_loss_clip": 0.01234608,
      "auxiliary_loss_mlp": 0.01047263,
      "balance_loss_clip": 1.06920779,
      "balance_loss_mlp": 1.03227854,
      "epoch": 0.06264654602296639,
      "flos": 20778058177920.0,
      "grad_norm": 2.9212118464256167,
      "language_loss": 0.73774111,
      "learning_rate": 3.988869398092371e-06,
      "loss": 0.7605598,
      "num_input_tokens_seen": 11154745,
      "step": 521,
      "time_per_iteration": 2.6357240676879883
    },
    {
      "auxiliary_loss_clip": 0.01246808,
      "auxiliary_loss_mlp": 0.01047027,
      "balance_loss_clip": 1.07113934,
      "balance_loss_mlp": 1.03257942,
      "epoch": 0.06276678891360549,
      "flos": 29605798241280.0,
      "grad_norm": 2.363192753656143,
      "language_loss": 0.79168439,
      "learning_rate": 3.988787178854206e-06,
      "loss": 0.81462276,
      "num_input_tokens_seen": 11174280,
      "step": 522,
      "time_per_iteration": 2.5789244174957275
    },
    {
      "auxiliary_loss_clip": 0.01278253,
      "auxiliary_loss_mlp": 0.01048427,
      "balance_loss_clip": 1.07506073,
      "balance_loss_mlp": 1.03434825,
      "epoch": 0.06288703180424457,
      "flos": 22126288193280.0,
      "grad_norm": 2.292919083044179,
      "language_loss": 0.8775146,
      "learning_rate": 3.988704657918608e-06,
      "loss": 0.90078139,
      "num_input_tokens_seen": 11193340,
      "step": 523,
      "time_per_iteration": 2.5421876907348633
    },
    {
      "auxiliary_loss_clip": 0.01260593,
      "auxiliary_loss_mlp": 0.01053341,
      "balance_loss_clip": 1.07486653,
      "balance_loss_mlp": 1.03994203,
      "epoch": 0.06300727469488367,
      "flos": 14976689587200.0,
      "grad_norm": 2.8851721886215547,
      "language_loss": 0.80009913,
      "learning_rate": 3.988621835298094e-06,
      "loss": 0.82323843,
      "num_input_tokens_seen": 11210555,
      "step": 524,
      "time_per_iteration": 2.4824345111846924
    },
    {
      "auxiliary_loss_clip": 0.01274082,
      "auxiliary_loss_mlp": 0.01045944,
      "balance_loss_clip": 1.07479048,
      "balance_loss_mlp": 1.0320915,
      "epoch": 0.06312751758552275,
      "flos": 24535391420160.0,
      "grad_norm": 1.9399650971274744,
      "language_loss": 0.91913724,
      "learning_rate": 3.988538711005229e-06,
      "loss": 0.94233757,
      "num_input_tokens_seen": 11230010,
      "step": 525,
      "time_per_iteration": 2.5533370971679688
    },
    {
      "auxiliary_loss_clip": 0.01254739,
      "auxiliary_loss_mlp": 0.01045345,
      "balance_loss_clip": 1.07180393,
      "balance_loss_mlp": 1.03244102,
      "epoch": 0.06324776047616185,
      "flos": 21507008785920.0,
      "grad_norm": 2.257703989251811,
      "language_loss": 0.88241124,
      "learning_rate": 3.988455285052622e-06,
      "loss": 0.90541208,
      "num_input_tokens_seen": 11246190,
      "step": 526,
      "time_per_iteration": 2.486172914505005
    },
    {
      "auxiliary_loss_clip": 0.01257162,
      "auxiliary_loss_mlp": 0.010524,
      "balance_loss_clip": 1.07308877,
      "balance_loss_mlp": 1.03854191,
      "epoch": 0.06336800336680094,
      "flos": 21688034353920.0,
      "grad_norm": 2.0333813020777622,
      "language_loss": 0.84032071,
      "learning_rate": 3.98837155745293e-06,
      "loss": 0.86341631,
      "num_input_tokens_seen": 11264230,
      "step": 527,
      "time_per_iteration": 2.557490825653076
    },
    {
      "auxiliary_loss_clip": 0.01263769,
      "auxiliary_loss_mlp": 0.01045589,
      "balance_loss_clip": 1.07662797,
      "balance_loss_mlp": 1.03129542,
      "epoch": 0.06348824625744003,
      "flos": 19500895221120.0,
      "grad_norm": 2.0420687094624435,
      "language_loss": 0.7600944,
      "learning_rate": 3.988287528218854e-06,
      "loss": 0.78318799,
      "num_input_tokens_seen": 11283015,
      "step": 528,
      "time_per_iteration": 3.268552303314209
    },
    {
      "auxiliary_loss_clip": 0.01259022,
      "auxiliary_loss_mlp": 0.01044304,
      "balance_loss_clip": 1.07505417,
      "balance_loss_mlp": 1.0313158,
      "epoch": 0.06360848914807912,
      "flos": 15481233976320.0,
      "grad_norm": 2.07627932200372,
      "language_loss": 0.90610939,
      "learning_rate": 3.98820319736314e-06,
      "loss": 0.92914271,
      "num_input_tokens_seen": 11299630,
      "step": 529,
      "time_per_iteration": 2.5093588829040527
    },
    {
      "auxiliary_loss_clip": 0.01228938,
      "auxiliary_loss_mlp": 0.01044071,
      "balance_loss_clip": 1.06564808,
      "balance_loss_mlp": 1.03015924,
      "epoch": 0.0637287320387182,
      "flos": 20593369422720.0,
      "grad_norm": 1.8016543805090741,
      "language_loss": 0.85254645,
      "learning_rate": 3.988118564898582e-06,
      "loss": 0.87527651,
      "num_input_tokens_seen": 11319170,
      "step": 530,
      "time_per_iteration": 2.573732852935791
    },
    {
      "auxiliary_loss_clip": 0.01223118,
      "auxiliary_loss_mlp": 0.0076726,
      "balance_loss_clip": 1.07045996,
      "balance_loss_mlp": 1.00039816,
      "epoch": 0.0638489749293573,
      "flos": 17412222245760.0,
      "grad_norm": 2.2854967834248274,
      "language_loss": 0.89324105,
      "learning_rate": 3.988033630838019e-06,
      "loss": 0.91314483,
      "num_input_tokens_seen": 11333210,
      "step": 531,
      "time_per_iteration": 2.605701446533203
    },
    {
      "auxiliary_loss_clip": 0.01263484,
      "auxiliary_loss_mlp": 0.0105179,
      "balance_loss_clip": 1.07501411,
      "balance_loss_mlp": 1.03842676,
      "epoch": 0.0639692178199964,
      "flos": 23807661874560.0,
      "grad_norm": 1.6475820456493724,
      "language_loss": 0.88092899,
      "learning_rate": 3.987948395194334e-06,
      "loss": 0.9040817,
      "num_input_tokens_seen": 11355590,
      "step": 532,
      "time_per_iteration": 4.175787925720215
    },
    {
      "auxiliary_loss_clip": 0.01250721,
      "auxiliary_loss_mlp": 0.01052487,
      "balance_loss_clip": 1.06814122,
      "balance_loss_mlp": 1.03903401,
      "epoch": 0.06408946071063548,
      "flos": 18477225521280.0,
      "grad_norm": 2.1057366672911475,
      "language_loss": 0.76576364,
      "learning_rate": 3.987862857980458e-06,
      "loss": 0.78879577,
      "num_input_tokens_seen": 11371535,
      "step": 533,
      "time_per_iteration": 2.5321125984191895
    },
    {
      "auxiliary_loss_clip": 0.01228782,
      "auxiliary_loss_mlp": 0.01045632,
      "balance_loss_clip": 1.06811428,
      "balance_loss_mlp": 1.03112459,
      "epoch": 0.06420970360127458,
      "flos": 27162220936320.0,
      "grad_norm": 2.634596840346206,
      "language_loss": 0.76803386,
      "learning_rate": 3.987777019209368e-06,
      "loss": 0.79077792,
      "num_input_tokens_seen": 11392050,
      "step": 534,
      "time_per_iteration": 3.3541109561920166
    },
    {
      "auxiliary_loss_clip": 0.01278138,
      "auxiliary_loss_mlp": 0.01042612,
      "balance_loss_clip": 1.07637906,
      "balance_loss_mlp": 1.02886152,
      "epoch": 0.06432994649191366,
      "flos": 23659673840640.0,
      "grad_norm": 4.810119883205886,
      "language_loss": 0.81286293,
      "learning_rate": 3.987690878894084e-06,
      "loss": 0.83607042,
      "num_input_tokens_seen": 11411765,
      "step": 535,
      "time_per_iteration": 2.5653088092803955
    },
    {
      "auxiliary_loss_clip": 0.01250204,
      "auxiliary_loss_mlp": 0.01036321,
      "balance_loss_clip": 1.07265449,
      "balance_loss_mlp": 1.02244473,
      "epoch": 0.06445018938255276,
      "flos": 23403953940480.0,
      "grad_norm": 2.6439322320632006,
      "language_loss": 0.85333127,
      "learning_rate": 3.987604437047673e-06,
      "loss": 0.8761965,
      "num_input_tokens_seen": 11431565,
      "step": 536,
      "time_per_iteration": 2.558048725128174
    },
    {
      "auxiliary_loss_clip": 0.01256634,
      "auxiliary_loss_mlp": 0.01043786,
      "balance_loss_clip": 1.07157469,
      "balance_loss_mlp": 1.03023195,
      "epoch": 0.06457043227319184,
      "flos": 19646692525440.0,
      "grad_norm": 2.2994892071675324,
      "language_loss": 0.77566165,
      "learning_rate": 3.987517693683251e-06,
      "loss": 0.79866588,
      "num_input_tokens_seen": 11450140,
      "step": 537,
      "time_per_iteration": 2.570960760116577
    },
    {
      "auxiliary_loss_clip": 0.01240312,
      "auxiliary_loss_mlp": 0.01056201,
      "balance_loss_clip": 1.07161307,
      "balance_loss_mlp": 1.04209828,
      "epoch": 0.06469067516383094,
      "flos": 16978744915200.0,
      "grad_norm": 3.178924399386921,
      "language_loss": 0.9600358,
      "learning_rate": 3.9874306488139745e-06,
      "loss": 0.98300099,
      "num_input_tokens_seen": 11465400,
      "step": 538,
      "time_per_iteration": 2.5246057510375977
    },
    {
      "auxiliary_loss_clip": 0.01225697,
      "auxiliary_loss_mlp": 0.01047211,
      "balance_loss_clip": 1.06984317,
      "balance_loss_mlp": 1.03340697,
      "epoch": 0.06481091805447003,
      "flos": 23296401642240.0,
      "grad_norm": 2.022463506911064,
      "language_loss": 0.87789249,
      "learning_rate": 3.987343302453049e-06,
      "loss": 0.90062153,
      "num_input_tokens_seen": 11486675,
      "step": 539,
      "time_per_iteration": 2.6185922622680664
    },
    {
      "auxiliary_loss_clip": 0.01242706,
      "auxiliary_loss_mlp": 0.0104951,
      "balance_loss_clip": 1.07214081,
      "balance_loss_mlp": 1.03572941,
      "epoch": 0.06493116094510912,
      "flos": 29172356824320.0,
      "grad_norm": 1.646683429518235,
      "language_loss": 0.82634234,
      "learning_rate": 3.987255654613724e-06,
      "loss": 0.8492645,
      "num_input_tokens_seen": 11510440,
      "step": 540,
      "time_per_iteration": 2.6307239532470703
    },
    {
      "auxiliary_loss_clip": 0.01220481,
      "auxiliary_loss_mlp": 0.01046343,
      "balance_loss_clip": 1.06522894,
      "balance_loss_mlp": 1.03256226,
      "epoch": 0.06505140383574821,
      "flos": 19865065259520.0,
      "grad_norm": 2.8280155344709574,
      "language_loss": 0.70145231,
      "learning_rate": 3.987167705309296e-06,
      "loss": 0.72412056,
      "num_input_tokens_seen": 11529715,
      "step": 541,
      "time_per_iteration": 2.5936086177825928
    },
    {
      "auxiliary_loss_clip": 0.01259629,
      "auxiliary_loss_mlp": 0.00766089,
      "balance_loss_clip": 1.07191896,
      "balance_loss_mlp": 1.00040436,
      "epoch": 0.0651716467263873,
      "flos": 17924703540480.0,
      "grad_norm": 2.014998082808893,
      "language_loss": 0.95142359,
      "learning_rate": 3.987079454553108e-06,
      "loss": 0.97168076,
      "num_input_tokens_seen": 11547665,
      "step": 542,
      "time_per_iteration": 2.5201218128204346
    },
    {
      "auxiliary_loss_clip": 0.01223939,
      "auxiliary_loss_mlp": 0.01043385,
      "balance_loss_clip": 1.07018447,
      "balance_loss_mlp": 1.02989042,
      "epoch": 0.0652918896170264,
      "flos": 20842840356480.0,
      "grad_norm": 1.8811402526990344,
      "language_loss": 0.91133386,
      "learning_rate": 3.986990902358546e-06,
      "loss": 0.93400711,
      "num_input_tokens_seen": 11564605,
      "step": 543,
      "time_per_iteration": 2.5884342193603516
    },
    {
      "auxiliary_loss_clip": 0.01259024,
      "auxiliary_loss_mlp": 0.01048271,
      "balance_loss_clip": 1.07132578,
      "balance_loss_mlp": 1.03453207,
      "epoch": 0.06541213250766549,
      "flos": 21872507627520.0,
      "grad_norm": 2.4480662023516198,
      "language_loss": 0.93326581,
      "learning_rate": 3.986902048739045e-06,
      "loss": 0.95633876,
      "num_input_tokens_seen": 11584550,
      "step": 544,
      "time_per_iteration": 2.540523052215576
    },
    {
      "auxiliary_loss_clip": 0.01244425,
      "auxiliary_loss_mlp": 0.0104968,
      "balance_loss_clip": 1.07057416,
      "balance_loss_mlp": 1.03486204,
      "epoch": 0.06553237539830457,
      "flos": 23110743219840.0,
      "grad_norm": 2.685322106617403,
      "language_loss": 0.79816675,
      "learning_rate": 3.986812893708082e-06,
      "loss": 0.82110775,
      "num_input_tokens_seen": 11600740,
      "step": 545,
      "time_per_iteration": 2.569758415222168
    },
    {
      "auxiliary_loss_clip": 0.01242421,
      "auxiliary_loss_mlp": 0.01048732,
      "balance_loss_clip": 1.06779075,
      "balance_loss_mlp": 1.03405702,
      "epoch": 0.06565261828894367,
      "flos": 17923769786880.0,
      "grad_norm": 2.3005544598330743,
      "language_loss": 0.81333226,
      "learning_rate": 3.9867234372791826e-06,
      "loss": 0.83624381,
      "num_input_tokens_seen": 11618695,
      "step": 546,
      "time_per_iteration": 2.511117935180664
    },
    {
      "auxiliary_loss_clip": 0.01255258,
      "auxiliary_loss_mlp": 0.0104612,
      "balance_loss_clip": 1.07118583,
      "balance_loss_mlp": 1.03232718,
      "epoch": 0.06577286117958275,
      "flos": 22783058421120.0,
      "grad_norm": 3.6692112474506358,
      "language_loss": 0.87165922,
      "learning_rate": 3.986633679465918e-06,
      "loss": 0.89467299,
      "num_input_tokens_seen": 11638850,
      "step": 547,
      "time_per_iteration": 2.5443801879882812
    },
    {
      "auxiliary_loss_clip": 0.01211543,
      "auxiliary_loss_mlp": 0.01052541,
      "balance_loss_clip": 1.06769454,
      "balance_loss_mlp": 1.038939,
      "epoch": 0.06589310407022185,
      "flos": 23696194993920.0,
      "grad_norm": 2.111359443670224,
      "language_loss": 0.80513996,
      "learning_rate": 3.986543620281904e-06,
      "loss": 0.82778072,
      "num_input_tokens_seen": 11658500,
      "step": 548,
      "time_per_iteration": 2.636155366897583
    },
    {
      "auxiliary_loss_clip": 0.01223896,
      "auxiliary_loss_mlp": 0.01034971,
      "balance_loss_clip": 1.06502867,
      "balance_loss_mlp": 1.02123213,
      "epoch": 0.06601334696086093,
      "flos": 26864772410880.0,
      "grad_norm": 1.927825081736006,
      "language_loss": 0.91351926,
      "learning_rate": 3.986453259740802e-06,
      "loss": 0.93610793,
      "num_input_tokens_seen": 11676670,
      "step": 549,
      "time_per_iteration": 2.6165995597839355
    },
    {
      "auxiliary_loss_clip": 0.01240177,
      "auxiliary_loss_mlp": 0.01049228,
      "balance_loss_clip": 1.07355571,
      "balance_loss_mlp": 1.03520274,
      "epoch": 0.06613358985150003,
      "flos": 12567694101120.0,
      "grad_norm": 2.7127872046687806,
      "language_loss": 0.79120648,
      "learning_rate": 3.986362597856319e-06,
      "loss": 0.8141005,
      "num_input_tokens_seen": 11693170,
      "step": 550,
      "time_per_iteration": 2.559262990951538
    },
    {
      "auxiliary_loss_clip": 0.01237732,
      "auxiliary_loss_mlp": 0.00767832,
      "balance_loss_clip": 1.06790137,
      "balance_loss_mlp": 1.00048375,
      "epoch": 0.06625383274213913,
      "flos": 18332505624960.0,
      "grad_norm": 2.8655993041631063,
      "language_loss": 0.81506467,
      "learning_rate": 3.986271634642211e-06,
      "loss": 0.83512026,
      "num_input_tokens_seen": 11710150,
      "step": 551,
      "time_per_iteration": 2.5514111518859863
    },
    {
      "auxiliary_loss_clip": 0.01271234,
      "auxiliary_loss_mlp": 0.01046563,
      "balance_loss_clip": 1.07411563,
      "balance_loss_mlp": 1.03225207,
      "epoch": 0.06637407563277821,
      "flos": 15375585098880.0,
      "grad_norm": 1.99877330424272,
      "language_loss": 0.81640399,
      "learning_rate": 3.986180370112274e-06,
      "loss": 0.83958191,
      "num_input_tokens_seen": 11726670,
      "step": 552,
      "time_per_iteration": 2.4699618816375732
    },
    {
      "auxiliary_loss_clip": 0.01257554,
      "auxiliary_loss_mlp": 0.00767611,
      "balance_loss_clip": 1.07208824,
      "balance_loss_mlp": 1.00044155,
      "epoch": 0.0664943185234173,
      "flos": 24025244509440.0,
      "grad_norm": 1.748967635523427,
      "language_loss": 0.74675941,
      "learning_rate": 3.986088804280354e-06,
      "loss": 0.76701099,
      "num_input_tokens_seen": 11746400,
      "step": 553,
      "time_per_iteration": 2.5647308826446533
    },
    {
      "auxiliary_loss_clip": 0.012432,
      "auxiliary_loss_mlp": 0.01046937,
      "balance_loss_clip": 1.07146311,
      "balance_loss_mlp": 1.03258979,
      "epoch": 0.06661456141405639,
      "flos": 20957503547520.0,
      "grad_norm": 2.530207107619296,
      "language_loss": 0.94020468,
      "learning_rate": 3.985996937160342e-06,
      "loss": 0.96310604,
      "num_input_tokens_seen": 11765590,
      "step": 554,
      "time_per_iteration": 3.3341805934906006
    },
    {
      "auxiliary_loss_clip": 0.01251465,
      "auxiliary_loss_mlp": 0.01051324,
      "balance_loss_clip": 1.07013559,
      "balance_loss_mlp": 1.03773403,
      "epoch": 0.06673480430469549,
      "flos": 52223953322880.0,
      "grad_norm": 1.9353637869000901,
      "language_loss": 0.68715233,
      "learning_rate": 3.985904768766173e-06,
      "loss": 0.71018022,
      "num_input_tokens_seen": 11788365,
      "step": 555,
      "time_per_iteration": 2.8013808727264404
    },
    {
      "auxiliary_loss_clip": 0.0122862,
      "auxiliary_loss_mlp": 0.01047409,
      "balance_loss_clip": 1.06846809,
      "balance_loss_mlp": 1.03291917,
      "epoch": 0.06685504719533458,
      "flos": 16217079995520.0,
      "grad_norm": 2.907012445313646,
      "language_loss": 0.75966603,
      "learning_rate": 3.98581229911183e-06,
      "loss": 0.7824263,
      "num_input_tokens_seen": 11807285,
      "step": 556,
      "time_per_iteration": 2.5596134662628174
    },
    {
      "auxiliary_loss_clip": 0.01257191,
      "auxiliary_loss_mlp": 0.01045985,
      "balance_loss_clip": 1.06949735,
      "balance_loss_mlp": 1.03170407,
      "epoch": 0.06697529008597367,
      "flos": 22491535639680.0,
      "grad_norm": 1.7139576351672947,
      "language_loss": 0.91908765,
      "learning_rate": 3.985719528211341e-06,
      "loss": 0.94211942,
      "num_input_tokens_seen": 11826655,
      "step": 557,
      "time_per_iteration": 2.5499160289764404
    },
    {
      "auxiliary_loss_clip": 0.01143971,
      "auxiliary_loss_mlp": 0.0100841,
      "balance_loss_clip": 1.03919816,
      "balance_loss_mlp": 1.00252092,
      "epoch": 0.06709553297661276,
      "flos": 62688216936960.0,
      "grad_norm": 0.8447410932211391,
      "language_loss": 0.63014245,
      "learning_rate": 3.985626456078777e-06,
      "loss": 0.65166628,
      "num_input_tokens_seen": 11891310,
      "step": 558,
      "time_per_iteration": 4.009221076965332
    },
    {
      "auxiliary_loss_clip": 0.01228934,
      "auxiliary_loss_mlp": 0.01045226,
      "balance_loss_clip": 1.07019997,
      "balance_loss_mlp": 1.03140962,
      "epoch": 0.06721577586725185,
      "flos": 11216590997760.0,
      "grad_norm": 2.1453007329263425,
      "language_loss": 0.86267483,
      "learning_rate": 3.985533082728259e-06,
      "loss": 0.88541645,
      "num_input_tokens_seen": 11906965,
      "step": 559,
      "time_per_iteration": 3.4730141162872314
    },
    {
      "auxiliary_loss_clip": 0.01277107,
      "auxiliary_loss_mlp": 0.01037639,
      "balance_loss_clip": 1.07436895,
      "balance_loss_mlp": 1.02359629,
      "epoch": 0.06733601875789094,
      "flos": 25922189664000.0,
      "grad_norm": 1.9090270022415432,
      "language_loss": 0.75014746,
      "learning_rate": 3.985439408173951e-06,
      "loss": 0.77329493,
      "num_input_tokens_seen": 11927190,
      "step": 560,
      "time_per_iteration": 2.5410356521606445
    },
    {
      "auxiliary_loss_clip": 0.0127515,
      "auxiliary_loss_mlp": 0.0105379,
      "balance_loss_clip": 1.0754025,
      "balance_loss_mlp": 1.03944969,
      "epoch": 0.06745626164853002,
      "flos": 20813645577600.0,
      "grad_norm": 2.430005953821239,
      "language_loss": 0.70951903,
      "learning_rate": 3.9853454324300634e-06,
      "loss": 0.73280847,
      "num_input_tokens_seen": 11946400,
      "step": 561,
      "time_per_iteration": 3.2623889446258545
    },
    {
      "auxiliary_loss_clip": 0.01199154,
      "auxiliary_loss_mlp": 0.01042931,
      "balance_loss_clip": 1.06361532,
      "balance_loss_mlp": 1.0283339,
      "epoch": 0.06757650453916912,
      "flos": 19829262378240.0,
      "grad_norm": 2.075310094903047,
      "language_loss": 0.77732491,
      "learning_rate": 3.985251155510852e-06,
      "loss": 0.7997458,
      "num_input_tokens_seen": 11965430,
      "step": 562,
      "time_per_iteration": 2.6828272342681885
    },
    {
      "auxiliary_loss_clip": 0.01209868,
      "auxiliary_loss_mlp": 0.01043947,
      "balance_loss_clip": 1.07043219,
      "balance_loss_mlp": 1.02976108,
      "epoch": 0.06769674742980822,
      "flos": 25739224761600.0,
      "grad_norm": 1.775925135573773,
      "language_loss": 0.80364096,
      "learning_rate": 3.98515657743062e-06,
      "loss": 0.82617909,
      "num_input_tokens_seen": 11984895,
      "step": 563,
      "time_per_iteration": 2.674323320388794
    },
    {
      "auxiliary_loss_clip": 0.01235705,
      "auxiliary_loss_mlp": 0.01054189,
      "balance_loss_clip": 1.06648397,
      "balance_loss_mlp": 1.04071283,
      "epoch": 0.0678169903204473,
      "flos": 13074788355840.0,
      "grad_norm": 2.2398220849929085,
      "language_loss": 0.77869093,
      "learning_rate": 3.985061698203711e-06,
      "loss": 0.80158991,
      "num_input_tokens_seen": 12002010,
      "step": 564,
      "time_per_iteration": 2.5196056365966797
    },
    {
      "auxiliary_loss_clip": 0.01161138,
      "auxiliary_loss_mlp": 0.01009265,
      "balance_loss_clip": 1.03621054,
      "balance_loss_mlp": 1.00354278,
      "epoch": 0.0679372332110864,
      "flos": 70865830788480.0,
      "grad_norm": 0.8904670521634256,
      "language_loss": 0.63779444,
      "learning_rate": 3.984966517844523e-06,
      "loss": 0.65949851,
      "num_input_tokens_seen": 12057255,
      "step": 565,
      "time_per_iteration": 3.0515527725219727
    },
    {
      "auxiliary_loss_clip": 0.01274531,
      "auxiliary_loss_mlp": 0.01049362,
      "balance_loss_clip": 1.07415497,
      "balance_loss_mlp": 1.03561151,
      "epoch": 0.06805747610172548,
      "flos": 28256418990720.0,
      "grad_norm": 2.4938047860429164,
      "language_loss": 0.80811733,
      "learning_rate": 3.984871036367492e-06,
      "loss": 0.83135623,
      "num_input_tokens_seen": 12077280,
      "step": 566,
      "time_per_iteration": 2.541935920715332
    },
    {
      "auxiliary_loss_clip": 0.01254022,
      "auxiliary_loss_mlp": 0.00766778,
      "balance_loss_clip": 1.07211626,
      "balance_loss_mlp": 1.00041664,
      "epoch": 0.06817771899236458,
      "flos": 20120533764480.0,
      "grad_norm": 1.9718423772549256,
      "language_loss": 0.83261657,
      "learning_rate": 3.984775253787102e-06,
      "loss": 0.85282451,
      "num_input_tokens_seen": 12095570,
      "step": 567,
      "time_per_iteration": 2.543840169906616
    },
    {
      "auxiliary_loss_clip": 0.01259013,
      "auxiliary_loss_mlp": 0.01040242,
      "balance_loss_clip": 1.0702976,
      "balance_loss_mlp": 1.02662253,
      "epoch": 0.06829796188300366,
      "flos": 17930629284480.0,
      "grad_norm": 2.9638526082362917,
      "language_loss": 0.87789202,
      "learning_rate": 3.984679170117885e-06,
      "loss": 0.90088463,
      "num_input_tokens_seen": 12111775,
      "step": 568,
      "time_per_iteration": 2.4911394119262695
    },
    {
      "auxiliary_loss_clip": 0.01253339,
      "auxiliary_loss_mlp": 0.01041753,
      "balance_loss_clip": 1.06978703,
      "balance_loss_mlp": 1.02701879,
      "epoch": 0.06841820477364276,
      "flos": 14501627285760.0,
      "grad_norm": 2.599977270728226,
      "language_loss": 0.78607631,
      "learning_rate": 3.984582785374415e-06,
      "loss": 0.80902725,
      "num_input_tokens_seen": 12129215,
      "step": 569,
      "time_per_iteration": 2.512500524520874
    },
    {
      "auxiliary_loss_clip": 0.01240847,
      "auxiliary_loss_mlp": 0.00766995,
      "balance_loss_clip": 1.07188869,
      "balance_loss_mlp": 1.00038803,
      "epoch": 0.06853844766428185,
      "flos": 21938474954880.0,
      "grad_norm": 2.106959947043695,
      "language_loss": 0.80572426,
      "learning_rate": 3.9844860995713155e-06,
      "loss": 0.82580268,
      "num_input_tokens_seen": 12148755,
      "step": 570,
      "time_per_iteration": 2.5807104110717773
    },
    {
      "auxiliary_loss_clip": 0.01257449,
      "auxiliary_loss_mlp": 0.01043734,
      "balance_loss_clip": 1.07771003,
      "balance_loss_mlp": 1.03032315,
      "epoch": 0.06865869055492094,
      "flos": 16800628348800.0,
      "grad_norm": 2.3690490243003826,
      "language_loss": 0.83040488,
      "learning_rate": 3.9843891127232524e-06,
      "loss": 0.8534168,
      "num_input_tokens_seen": 12166290,
      "step": 571,
      "time_per_iteration": 2.5020804405212402
    },
    {
      "auxiliary_loss_clip": 0.0119598,
      "auxiliary_loss_mlp": 0.0104394,
      "balance_loss_clip": 1.06299663,
      "balance_loss_mlp": 1.02921748,
      "epoch": 0.06877893344556003,
      "flos": 19937281553280.0,
      "grad_norm": 2.2342563898718804,
      "language_loss": 0.66607964,
      "learning_rate": 3.984291824844938e-06,
      "loss": 0.68847883,
      "num_input_tokens_seen": 12181385,
      "step": 572,
      "time_per_iteration": 2.6187117099761963
    },
    {
      "auxiliary_loss_clip": 0.0127414,
      "auxiliary_loss_mlp": 0.01042764,
      "balance_loss_clip": 1.07464194,
      "balance_loss_mlp": 1.02866173,
      "epoch": 0.06889917633619912,
      "flos": 23039388852480.0,
      "grad_norm": 2.5559593446396294,
      "language_loss": 0.85313123,
      "learning_rate": 3.984194235951132e-06,
      "loss": 0.87630028,
      "num_input_tokens_seen": 12197530,
      "step": 573,
      "time_per_iteration": 2.500725746154785
    },
    {
      "auxiliary_loss_clip": 0.01277281,
      "auxiliary_loss_mlp": 0.01057823,
      "balance_loss_clip": 1.07881522,
      "balance_loss_mlp": 1.04467988,
      "epoch": 0.06901941922683821,
      "flos": 20960556203520.0,
      "grad_norm": 2.6776634110256503,
      "language_loss": 0.84668672,
      "learning_rate": 3.9840963460566375e-06,
      "loss": 0.87003773,
      "num_input_tokens_seen": 12216310,
      "step": 574,
      "time_per_iteration": 2.4887170791625977
    },
    {
      "auxiliary_loss_clip": 0.01180166,
      "auxiliary_loss_mlp": 0.01040193,
      "balance_loss_clip": 1.06158209,
      "balance_loss_mlp": 1.02635908,
      "epoch": 0.06913966211747731,
      "flos": 24821850384000.0,
      "grad_norm": 2.8095923957813396,
      "language_loss": 0.89605409,
      "learning_rate": 3.983998155176305e-06,
      "loss": 0.91825771,
      "num_input_tokens_seen": 12236670,
      "step": 575,
      "time_per_iteration": 2.8775432109832764
    },
    {
      "auxiliary_loss_clip": 0.01160083,
      "auxiliary_loss_mlp": 0.01004491,
      "balance_loss_clip": 1.03653765,
      "balance_loss_mlp": 0.99853033,
      "epoch": 0.06925990500811639,
      "flos": 58367446957440.0,
      "grad_norm": 0.8249431166627818,
      "language_loss": 0.57010722,
      "learning_rate": 3.9838996633250305e-06,
      "loss": 0.59175301,
      "num_input_tokens_seen": 12297185,
      "step": 576,
      "time_per_iteration": 3.157111883163452
    },
    {
      "auxiliary_loss_clip": 0.01255641,
      "auxiliary_loss_mlp": 0.01048447,
      "balance_loss_clip": 1.07015228,
      "balance_loss_mlp": 1.03566158,
      "epoch": 0.06938014789875549,
      "flos": 12749940731520.0,
      "grad_norm": 2.202914150551851,
      "language_loss": 0.88080609,
      "learning_rate": 3.983800870517753e-06,
      "loss": 0.90384698,
      "num_input_tokens_seen": 12313975,
      "step": 577,
      "time_per_iteration": 2.5130162239074707
    },
    {
      "auxiliary_loss_clip": 0.01252016,
      "auxiliary_loss_mlp": 0.01045524,
      "balance_loss_clip": 1.07390738,
      "balance_loss_mlp": 1.03321528,
      "epoch": 0.06950039078939457,
      "flos": 22820226019200.0,
      "grad_norm": 2.539474317481431,
      "language_loss": 0.78479326,
      "learning_rate": 3.983701776769463e-06,
      "loss": 0.80776858,
      "num_input_tokens_seen": 12331385,
      "step": 578,
      "time_per_iteration": 2.5052263736724854
    },
    {
      "auxiliary_loss_clip": 0.01250283,
      "auxiliary_loss_mlp": 0.01043673,
      "balance_loss_clip": 1.07302582,
      "balance_loss_mlp": 1.02979124,
      "epoch": 0.06962063368003367,
      "flos": 21941348042880.0,
      "grad_norm": 1.8980668948412203,
      "language_loss": 0.85984695,
      "learning_rate": 3.9836023820951885e-06,
      "loss": 0.88278651,
      "num_input_tokens_seen": 12350600,
      "step": 579,
      "time_per_iteration": 2.526423692703247
    },
    {
      "auxiliary_loss_clip": 0.01216759,
      "auxiliary_loss_mlp": 0.01049247,
      "balance_loss_clip": 1.06326473,
      "balance_loss_mlp": 1.03671849,
      "epoch": 0.06974087657067275,
      "flos": 20706021452160.0,
      "grad_norm": 1.789721971271885,
      "language_loss": 0.68311691,
      "learning_rate": 3.983502686510011e-06,
      "loss": 0.70577699,
      "num_input_tokens_seen": 12371430,
      "step": 580,
      "time_per_iteration": 2.6056437492370605
    },
    {
      "auxiliary_loss_clip": 0.01258208,
      "auxiliary_loss_mlp": 0.00766454,
      "balance_loss_clip": 1.06969035,
      "balance_loss_mlp": 1.00043762,
      "epoch": 0.06986111946131185,
      "flos": 22638230784000.0,
      "grad_norm": 2.542945349135428,
      "language_loss": 0.73738778,
      "learning_rate": 3.9834026900290525e-06,
      "loss": 0.75763446,
      "num_input_tokens_seen": 12390825,
      "step": 581,
      "time_per_iteration": 3.3825559616088867
    },
    {
      "auxiliary_loss_clip": 0.01270935,
      "auxiliary_loss_mlp": 0.01044763,
      "balance_loss_clip": 1.07250535,
      "balance_loss_mlp": 1.03175092,
      "epoch": 0.06998136235195095,
      "flos": 26943453152640.0,
      "grad_norm": 2.0011242746998557,
      "language_loss": 1.00358772,
      "learning_rate": 3.983302392667482e-06,
      "loss": 1.0267446,
      "num_input_tokens_seen": 12411670,
      "step": 582,
      "time_per_iteration": 2.5829575061798096
    },
    {
      "auxiliary_loss_clip": 0.01254151,
      "auxiliary_loss_mlp": 0.01043282,
      "balance_loss_clip": 1.07271314,
      "balance_loss_mlp": 1.03006136,
      "epoch": 0.07010160524259003,
      "flos": 22492505306880.0,
      "grad_norm": 1.8228352045585985,
      "language_loss": 0.93882,
      "learning_rate": 3.983201794440517e-06,
      "loss": 0.96179426,
      "num_input_tokens_seen": 12431245,
      "step": 583,
      "time_per_iteration": 2.4973533153533936
    },
    {
      "auxiliary_loss_clip": 0.01227027,
      "auxiliary_loss_mlp": 0.01040701,
      "balance_loss_clip": 1.06721604,
      "balance_loss_mlp": 1.02745128,
      "epoch": 0.07022184813322913,
      "flos": 18332541538560.0,
      "grad_norm": 1.6934672667998625,
      "language_loss": 0.67659652,
      "learning_rate": 3.9831008953634165e-06,
      "loss": 0.69927382,
      "num_input_tokens_seen": 12450535,
      "step": 584,
      "time_per_iteration": 3.3078415393829346
    },
    {
      "auxiliary_loss_clip": 0.01189039,
      "auxiliary_loss_mlp": 0.0104552,
      "balance_loss_clip": 1.05903637,
      "balance_loss_mlp": 1.03095865,
      "epoch": 0.07034209102386821,
      "flos": 24675550289280.0,
      "grad_norm": 2.2925334378134368,
      "language_loss": 0.8127383,
      "learning_rate": 3.9829996954514864e-06,
      "loss": 0.83508384,
      "num_input_tokens_seen": 12469675,
      "step": 585,
      "time_per_iteration": 2.6435909271240234
    },
    {
      "auxiliary_loss_clip": 0.01244677,
      "auxiliary_loss_mlp": 0.01047812,
      "balance_loss_clip": 1.06925857,
      "balance_loss_mlp": 1.03337574,
      "epoch": 0.0704623339145073,
      "flos": 25995878415360.0,
      "grad_norm": 1.8228520911126989,
      "language_loss": 0.84309483,
      "learning_rate": 3.982898194720079e-06,
      "loss": 0.86601979,
      "num_input_tokens_seen": 12490405,
      "step": 586,
      "time_per_iteration": 3.4164116382598877
    },
    {
      "auxiliary_loss_clip": 0.01236621,
      "auxiliary_loss_mlp": 0.00767169,
      "balance_loss_clip": 1.07208943,
      "balance_loss_mlp": 1.00044227,
      "epoch": 0.0705825768051464,
      "flos": 25338318088320.0,
      "grad_norm": 2.1328378285480496,
      "language_loss": 0.82623172,
      "learning_rate": 3.982796393184592e-06,
      "loss": 0.84626967,
      "num_input_tokens_seen": 12509485,
      "step": 587,
      "time_per_iteration": 3.375338554382324
    },
    {
      "auxiliary_loss_clip": 0.0114264,
      "auxiliary_loss_mlp": 0.01018892,
      "balance_loss_clip": 1.03358495,
      "balance_loss_mlp": 1.01264572,
      "epoch": 0.07070281969578548,
      "flos": 66047552507520.0,
      "grad_norm": 0.7932328357860626,
      "language_loss": 0.62630105,
      "learning_rate": 3.98269429086047e-06,
      "loss": 0.64791638,
      "num_input_tokens_seen": 12567325,
      "step": 588,
      "time_per_iteration": 2.972320318222046
    },
    {
      "auxiliary_loss_clip": 0.01227739,
      "auxiliary_loss_mlp": 0.01047875,
      "balance_loss_clip": 1.06882036,
      "balance_loss_mlp": 1.03336143,
      "epoch": 0.07082306258642458,
      "flos": 23653568528640.0,
      "grad_norm": 3.1409108826991807,
      "language_loss": 0.8650409,
      "learning_rate": 3.982591887763199e-06,
      "loss": 0.887797,
      "num_input_tokens_seen": 12584785,
      "step": 589,
      "time_per_iteration": 2.5690934658050537
    },
    {
      "auxiliary_loss_clip": 0.01199029,
      "auxiliary_loss_mlp": 0.01041068,
      "balance_loss_clip": 1.05703413,
      "balance_loss_mlp": 1.02635217,
      "epoch": 0.07094330547706366,
      "flos": 13880049408000.0,
      "grad_norm": 2.259644323531196,
      "language_loss": 0.81569409,
      "learning_rate": 3.982489183908316e-06,
      "loss": 0.83809513,
      "num_input_tokens_seen": 12601205,
      "step": 590,
      "time_per_iteration": 2.539745807647705
    },
    {
      "auxiliary_loss_clip": 0.01162121,
      "auxiliary_loss_mlp": 0.01044046,
      "balance_loss_clip": 1.05206561,
      "balance_loss_mlp": 1.03160095,
      "epoch": 0.07106354836770276,
      "flos": 24645098534400.0,
      "grad_norm": 2.361266753145197,
      "language_loss": 0.84465516,
      "learning_rate": 3.982386179311399e-06,
      "loss": 0.86671686,
      "num_input_tokens_seen": 12621725,
      "step": 591,
      "time_per_iteration": 2.6699905395507812
    },
    {
      "auxiliary_loss_clip": 0.01258631,
      "auxiliary_loss_mlp": 0.01047805,
      "balance_loss_clip": 1.07177889,
      "balance_loss_mlp": 1.03220034,
      "epoch": 0.07118379125834184,
      "flos": 16217223649920.0,
      "grad_norm": 2.277611624363638,
      "language_loss": 0.87452149,
      "learning_rate": 3.982282873988075e-06,
      "loss": 0.89758587,
      "num_input_tokens_seen": 12639600,
      "step": 592,
      "time_per_iteration": 2.4876914024353027
    },
    {
      "auxiliary_loss_clip": 0.01238607,
      "auxiliary_loss_mlp": 0.01042214,
      "balance_loss_clip": 1.07014549,
      "balance_loss_mlp": 1.02961385,
      "epoch": 0.07130403414898094,
      "flos": 19719986227200.0,
      "grad_norm": 1.6624626919673313,
      "language_loss": 0.87099338,
      "learning_rate": 3.982179267954016e-06,
      "loss": 0.89380157,
      "num_input_tokens_seen": 12660030,
      "step": 593,
      "time_per_iteration": 2.58138108253479
    },
    {
      "auxiliary_loss_clip": 0.0126898,
      "auxiliary_loss_mlp": 0.01038798,
      "balance_loss_clip": 1.07175934,
      "balance_loss_mlp": 1.02510643,
      "epoch": 0.07142427703962004,
      "flos": 21871933009920.0,
      "grad_norm": 2.132711499028081,
      "language_loss": 0.96150446,
      "learning_rate": 3.982075361224937e-06,
      "loss": 0.98458219,
      "num_input_tokens_seen": 12678395,
      "step": 594,
      "time_per_iteration": 2.519207239151001
    },
    {
      "auxiliary_loss_clip": 0.0124978,
      "auxiliary_loss_mlp": 0.00766412,
      "balance_loss_clip": 1.07181573,
      "balance_loss_mlp": 1.00041747,
      "epoch": 0.07154451993025912,
      "flos": 18296595002880.0,
      "grad_norm": 2.029373387923085,
      "language_loss": 0.87810063,
      "learning_rate": 3.981971153816602e-06,
      "loss": 0.89826262,
      "num_input_tokens_seen": 12696000,
      "step": 595,
      "time_per_iteration": 2.512521743774414
    },
    {
      "auxiliary_loss_clip": 0.01269919,
      "auxiliary_loss_mlp": 0.01041849,
      "balance_loss_clip": 1.07637262,
      "balance_loss_mlp": 1.02937961,
      "epoch": 0.07166476282089822,
      "flos": 22160690444160.0,
      "grad_norm": 1.6195154270940286,
      "language_loss": 0.96278137,
      "learning_rate": 3.981866645744819e-06,
      "loss": 0.98589909,
      "num_input_tokens_seen": 12716715,
      "step": 596,
      "time_per_iteration": 2.5581467151641846
    },
    {
      "auxiliary_loss_clip": 0.01272034,
      "auxiliary_loss_mlp": 0.00767275,
      "balance_loss_clip": 1.07455778,
      "balance_loss_mlp": 1.00038826,
      "epoch": 0.0717850057115373,
      "flos": 14136343925760.0,
      "grad_norm": 2.0429279469758113,
      "language_loss": 0.81561875,
      "learning_rate": 3.9817618370254416e-06,
      "loss": 0.83601189,
      "num_input_tokens_seen": 12733370,
      "step": 597,
      "time_per_iteration": 2.490833282470703
    },
    {
      "auxiliary_loss_clip": 0.01270889,
      "auxiliary_loss_mlp": 0.01050715,
      "balance_loss_clip": 1.07419133,
      "balance_loss_mlp": 1.03690469,
      "epoch": 0.0719052486021764,
      "flos": 30917794412160.0,
      "grad_norm": 2.035687491896235,
      "language_loss": 0.87465435,
      "learning_rate": 3.9816567276743684e-06,
      "loss": 0.89787042,
      "num_input_tokens_seen": 12753235,
      "step": 598,
      "time_per_iteration": 2.560598134994507
    },
    {
      "auxiliary_loss_clip": 0.01233461,
      "auxiliary_loss_mlp": 0.01042264,
      "balance_loss_clip": 1.06928658,
      "balance_loss_mlp": 1.02846587,
      "epoch": 0.0720254914928155,
      "flos": 21287019939840.0,
      "grad_norm": 2.1583285288588976,
      "language_loss": 0.77268213,
      "learning_rate": 3.9815513177075466e-06,
      "loss": 0.79543942,
      "num_input_tokens_seen": 12772020,
      "step": 599,
      "time_per_iteration": 2.5347886085510254
    },
    {
      "auxiliary_loss_clip": 0.01243331,
      "auxiliary_loss_mlp": 0.01043086,
      "balance_loss_clip": 1.06968951,
      "balance_loss_mlp": 1.0309267,
      "epoch": 0.07214573438345458,
      "flos": 27819170732160.0,
      "grad_norm": 1.6157611217976484,
      "language_loss": 0.70185947,
      "learning_rate": 3.9814456071409646e-06,
      "loss": 0.72472358,
      "num_input_tokens_seen": 12792555,
      "step": 600,
      "time_per_iteration": 2.6127421855926514
    },
    {
      "auxiliary_loss_clip": 0.01207243,
      "auxiliary_loss_mlp": 0.01054259,
      "balance_loss_clip": 1.06441975,
      "balance_loss_mlp": 1.03988266,
      "epoch": 0.07226597727409367,
      "flos": 25483576688640.0,
      "grad_norm": 3.3255513682903772,
      "language_loss": 0.85243708,
      "learning_rate": 3.981339595990659e-06,
      "loss": 0.87505209,
      "num_input_tokens_seen": 12811085,
      "step": 601,
      "time_per_iteration": 2.6214487552642822
    },
    {
      "auxiliary_loss_clip": 0.01252778,
      "auxiliary_loss_mlp": 0.01051634,
      "balance_loss_clip": 1.07165539,
      "balance_loss_mlp": 1.03695917,
      "epoch": 0.07238622016473276,
      "flos": 23513840622720.0,
      "grad_norm": 2.047304202507383,
      "language_loss": 0.80958486,
      "learning_rate": 3.981233284272713e-06,
      "loss": 0.83262897,
      "num_input_tokens_seen": 12830830,
      "step": 602,
      "time_per_iteration": 2.525503396987915
    },
    {
      "auxiliary_loss_clip": 0.0122027,
      "auxiliary_loss_mlp": 0.01042446,
      "balance_loss_clip": 1.06605244,
      "balance_loss_mlp": 1.03034663,
      "epoch": 0.07250646305537185,
      "flos": 25453519983360.0,
      "grad_norm": 2.1507699758599914,
      "language_loss": 0.90146458,
      "learning_rate": 3.981126672003253e-06,
      "loss": 0.92409182,
      "num_input_tokens_seen": 12853505,
      "step": 603,
      "time_per_iteration": 2.6561989784240723
    },
    {
      "auxiliary_loss_clip": 0.01239124,
      "auxiliary_loss_mlp": 0.01045948,
      "balance_loss_clip": 1.06450915,
      "balance_loss_mlp": 1.03307962,
      "epoch": 0.07262670594601094,
      "flos": 27155038216320.0,
      "grad_norm": 4.26484111265464,
      "language_loss": 0.78068703,
      "learning_rate": 3.981019759198451e-06,
      "loss": 0.80353779,
      "num_input_tokens_seen": 12872455,
      "step": 604,
      "time_per_iteration": 2.6214585304260254
    },
    {
      "auxiliary_loss_clip": 0.012373,
      "auxiliary_loss_mlp": 0.01048402,
      "balance_loss_clip": 1.06864572,
      "balance_loss_mlp": 1.03468108,
      "epoch": 0.07274694883665003,
      "flos": 26651607148800.0,
      "grad_norm": 2.466745686178935,
      "language_loss": 0.84249026,
      "learning_rate": 3.980912545874528e-06,
      "loss": 0.86534727,
      "num_input_tokens_seen": 12892620,
      "step": 605,
      "time_per_iteration": 2.586289405822754
    },
    {
      "auxiliary_loss_clip": 0.01248823,
      "auxiliary_loss_mlp": 0.007667,
      "balance_loss_clip": 1.07069409,
      "balance_loss_mlp": 1.00038338,
      "epoch": 0.07286719172728913,
      "flos": 29862344154240.0,
      "grad_norm": 1.8941259392666263,
      "language_loss": 0.85113156,
      "learning_rate": 3.980805032047746e-06,
      "loss": 0.87128687,
      "num_input_tokens_seen": 12914090,
      "step": 606,
      "time_per_iteration": 2.5893375873565674
    },
    {
      "auxiliary_loss_clip": 0.01230261,
      "auxiliary_loss_mlp": 0.01045049,
      "balance_loss_clip": 1.06495249,
      "balance_loss_mlp": 1.02938533,
      "epoch": 0.07298743461792821,
      "flos": 17382057799680.0,
      "grad_norm": 2.07065328759031,
      "language_loss": 0.81085038,
      "learning_rate": 3.980697217734415e-06,
      "loss": 0.8336035,
      "num_input_tokens_seen": 12931830,
      "step": 607,
      "time_per_iteration": 3.341517925262451
    },
    {
      "auxiliary_loss_clip": 0.01208733,
      "auxiliary_loss_mlp": 0.00766282,
      "balance_loss_clip": 1.06585383,
      "balance_loss_mlp": 1.00038433,
      "epoch": 0.07310767750856731,
      "flos": 19498201701120.0,
      "grad_norm": 3.847156902932048,
      "language_loss": 0.91369158,
      "learning_rate": 3.980589102950891e-06,
      "loss": 0.93344176,
      "num_input_tokens_seen": 12949995,
      "step": 608,
      "time_per_iteration": 2.586209535598755
    },
    {
      "auxiliary_loss_clip": 0.01237563,
      "auxiliary_loss_mlp": 0.01045981,
      "balance_loss_clip": 1.07175493,
      "balance_loss_mlp": 1.03175354,
      "epoch": 0.07322792039920639,
      "flos": 29168693637120.0,
      "grad_norm": 2.4805693067264927,
      "language_loss": 0.76412082,
      "learning_rate": 3.9804806877135755e-06,
      "loss": 0.78695619,
      "num_input_tokens_seen": 12968040,
      "step": 609,
      "time_per_iteration": 2.6467506885528564
    },
    {
      "auxiliary_loss_clip": 0.01258454,
      "auxiliary_loss_mlp": 0.00767288,
      "balance_loss_clip": 1.07046413,
      "balance_loss_mlp": 1.00043476,
      "epoch": 0.07334816328984549,
      "flos": 23477822259840.0,
      "grad_norm": 2.431415471061829,
      "language_loss": 0.85902739,
      "learning_rate": 3.980371972038915e-06,
      "loss": 0.87928486,
      "num_input_tokens_seen": 12988530,
      "step": 610,
      "time_per_iteration": 2.5906574726104736
    },
    {
      "auxiliary_loss_clip": 0.01270548,
      "auxiliary_loss_mlp": 0.01048316,
      "balance_loss_clip": 1.0740211,
      "balance_loss_mlp": 1.03436267,
      "epoch": 0.07346840618048459,
      "flos": 22962467877120.0,
      "grad_norm": 1.8416321857646867,
      "language_loss": 0.84427935,
      "learning_rate": 3.980262955943399e-06,
      "loss": 0.867468,
      "num_input_tokens_seen": 13008195,
      "step": 611,
      "time_per_iteration": 3.3923685550689697
    },
    {
      "auxiliary_loss_clip": 0.01229167,
      "auxiliary_loss_mlp": 0.01045747,
      "balance_loss_clip": 1.07031846,
      "balance_loss_mlp": 1.03315806,
      "epoch": 0.07358864907112367,
      "flos": 17673903803520.0,
      "grad_norm": 2.529981481700748,
      "language_loss": 0.86986494,
      "learning_rate": 3.980153639443569e-06,
      "loss": 0.89261407,
      "num_input_tokens_seen": 13024180,
      "step": 612,
      "time_per_iteration": 3.392343759536743
    },
    {
      "auxiliary_loss_clip": 0.01241957,
      "auxiliary_loss_mlp": 0.01045824,
      "balance_loss_clip": 1.07091379,
      "balance_loss_mlp": 1.03235972,
      "epoch": 0.07370889196176277,
      "flos": 24097029840000.0,
      "grad_norm": 2.0301275125406475,
      "language_loss": 0.8016212,
      "learning_rate": 3.980044022556005e-06,
      "loss": 0.82449901,
      "num_input_tokens_seen": 13043865,
      "step": 613,
      "time_per_iteration": 2.5738940238952637
    },
    {
      "auxiliary_loss_clip": 0.01253469,
      "auxiliary_loss_mlp": 0.01054136,
      "balance_loss_clip": 1.07109725,
      "balance_loss_mlp": 1.04085612,
      "epoch": 0.07382913485240185,
      "flos": 25885919905920.0,
      "grad_norm": 2.2365504979392195,
      "language_loss": 0.72833198,
      "learning_rate": 3.9799341052973375e-06,
      "loss": 0.75140798,
      "num_input_tokens_seen": 13063700,
      "step": 614,
      "time_per_iteration": 3.3543448448181152
    },
    {
      "auxiliary_loss_clip": 0.01238887,
      "auxiliary_loss_mlp": 0.01041745,
      "balance_loss_clip": 1.07326174,
      "balance_loss_mlp": 1.02729702,
      "epoch": 0.07394937774304094,
      "flos": 16873850223360.0,
      "grad_norm": 2.370760727714136,
      "language_loss": 0.74629426,
      "learning_rate": 3.979823887684241e-06,
      "loss": 0.76910055,
      "num_input_tokens_seen": 13082640,
      "step": 615,
      "time_per_iteration": 2.532843828201294
    },
    {
      "auxiliary_loss_clip": 0.01268693,
      "auxiliary_loss_mlp": 0.01048816,
      "balance_loss_clip": 1.07419467,
      "balance_loss_mlp": 1.03496361,
      "epoch": 0.07406962063368003,
      "flos": 20703471586560.0,
      "grad_norm": 6.031963824455091,
      "language_loss": 0.84616119,
      "learning_rate": 3.979713369733434e-06,
      "loss": 0.86933625,
      "num_input_tokens_seen": 13100505,
      "step": 616,
      "time_per_iteration": 2.4737789630889893
    },
    {
      "auxiliary_loss_clip": 0.01249145,
      "auxiliary_loss_mlp": 0.01053868,
      "balance_loss_clip": 1.07168019,
      "balance_loss_mlp": 1.04010522,
      "epoch": 0.07418986352431912,
      "flos": 21430985650560.0,
      "grad_norm": 2.1586248099889898,
      "language_loss": 0.84475207,
      "learning_rate": 3.979602551461683e-06,
      "loss": 0.86778224,
      "num_input_tokens_seen": 13121285,
      "step": 617,
      "time_per_iteration": 2.545459747314453
    },
    {
      "auxiliary_loss_clip": 0.01232963,
      "auxiliary_loss_mlp": 0.01045402,
      "balance_loss_clip": 1.06911814,
      "balance_loss_mlp": 1.03206229,
      "epoch": 0.07431010641495822,
      "flos": 12021133777920.0,
      "grad_norm": 2.211790457161869,
      "language_loss": 0.91610944,
      "learning_rate": 3.979491432885799e-06,
      "loss": 0.93889308,
      "num_input_tokens_seen": 13137550,
      "step": 618,
      "time_per_iteration": 2.505262613296509
    },
    {
      "auxiliary_loss_clip": 0.01199363,
      "auxiliary_loss_mlp": 0.00766176,
      "balance_loss_clip": 1.06141984,
      "balance_loss_mlp": 1.00033629,
      "epoch": 0.0744303493055973,
      "flos": 20957575374720.0,
      "grad_norm": 3.1984461305388243,
      "language_loss": 0.83165497,
      "learning_rate": 3.97938001402264e-06,
      "loss": 0.85131025,
      "num_input_tokens_seen": 13156675,
      "step": 619,
      "time_per_iteration": 2.5636367797851562
    },
    {
      "auxiliary_loss_clip": 0.01214358,
      "auxiliary_loss_mlp": 0.01044044,
      "balance_loss_clip": 1.06799436,
      "balance_loss_mlp": 1.0309844,
      "epoch": 0.0745505921962364,
      "flos": 16253134272000.0,
      "grad_norm": 2.762055545839595,
      "language_loss": 0.79720354,
      "learning_rate": 3.979268294889105e-06,
      "loss": 0.81978762,
      "num_input_tokens_seen": 13172225,
      "step": 620,
      "time_per_iteration": 2.5558083057403564
    },
    {
      "auxiliary_loss_clip": 0.01268026,
      "auxiliary_loss_mlp": 0.01045921,
      "balance_loss_clip": 1.07303298,
      "balance_loss_mlp": 1.03282547,
      "epoch": 0.07467083508687548,
      "flos": 50944635550080.0,
      "grad_norm": 1.8737530166268968,
      "language_loss": 0.7385307,
      "learning_rate": 3.979156275502143e-06,
      "loss": 0.76167017,
      "num_input_tokens_seen": 13195885,
      "step": 621,
      "time_per_iteration": 2.7360219955444336
    },
    {
      "auxiliary_loss_clip": 0.0122435,
      "auxiliary_loss_mlp": 0.01049868,
      "balance_loss_clip": 1.06858814,
      "balance_loss_mlp": 1.03562284,
      "epoch": 0.07479107797751458,
      "flos": 17529686697600.0,
      "grad_norm": 2.2542555759247103,
      "language_loss": 0.91447878,
      "learning_rate": 3.979043955878749e-06,
      "loss": 0.93722099,
      "num_input_tokens_seen": 13213730,
      "step": 622,
      "time_per_iteration": 2.5534796714782715
    },
    {
      "auxiliary_loss_clip": 0.0123374,
      "auxiliary_loss_mlp": 0.01041248,
      "balance_loss_clip": 1.06887007,
      "balance_loss_mlp": 1.02829587,
      "epoch": 0.07491132086815366,
      "flos": 23473943591040.0,
      "grad_norm": 2.323915351861144,
      "language_loss": 0.83450806,
      "learning_rate": 3.978931336035959e-06,
      "loss": 0.85725796,
      "num_input_tokens_seen": 13232540,
      "step": 623,
      "time_per_iteration": 2.5568387508392334
    },
    {
      "auxiliary_loss_clip": 0.0125447,
      "auxiliary_loss_mlp": 0.01048807,
      "balance_loss_clip": 1.07419753,
      "balance_loss_mlp": 1.03507972,
      "epoch": 0.07503156375879276,
      "flos": 20157557708160.0,
      "grad_norm": 2.1762620619161925,
      "language_loss": 0.82393086,
      "learning_rate": 3.9788184159908595e-06,
      "loss": 0.84696364,
      "num_input_tokens_seen": 13249670,
      "step": 624,
      "time_per_iteration": 2.513934373855591
    },
    {
      "auxiliary_loss_clip": 0.01230364,
      "auxiliary_loss_mlp": 0.01049586,
      "balance_loss_clip": 1.06845868,
      "balance_loss_mlp": 1.03695536,
      "epoch": 0.07515180664943186,
      "flos": 15115519653120.0,
      "grad_norm": 2.3595351152660653,
      "language_loss": 0.81973308,
      "learning_rate": 3.97870519576058e-06,
      "loss": 0.84253263,
      "num_input_tokens_seen": 13266095,
      "step": 625,
      "time_per_iteration": 2.5559334754943848
    },
    {
      "auxiliary_loss_clip": 0.01218141,
      "auxiliary_loss_mlp": 0.00767158,
      "balance_loss_clip": 1.06659019,
      "balance_loss_mlp": 1.0003103,
      "epoch": 0.07527204954007094,
      "flos": 21287702298240.0,
      "grad_norm": 2.801635387591967,
      "language_loss": 0.80702305,
      "learning_rate": 3.978591675362295e-06,
      "loss": 0.82687598,
      "num_input_tokens_seen": 13284810,
      "step": 626,
      "time_per_iteration": 2.595956802368164
    },
    {
      "auxiliary_loss_clip": 0.01201154,
      "auxiliary_loss_mlp": 0.01044592,
      "balance_loss_clip": 1.06840372,
      "balance_loss_mlp": 1.03134155,
      "epoch": 0.07539229243071004,
      "flos": 21324187537920.0,
      "grad_norm": 1.6730813388501025,
      "language_loss": 0.87549746,
      "learning_rate": 3.978477854813226e-06,
      "loss": 0.89795494,
      "num_input_tokens_seen": 13304150,
      "step": 627,
      "time_per_iteration": 2.600982666015625
    },
    {
      "auxiliary_loss_clip": 0.01253248,
      "auxiliary_loss_mlp": 0.01048684,
      "balance_loss_clip": 1.07085347,
      "balance_loss_mlp": 1.03615475,
      "epoch": 0.07551253532134912,
      "flos": 13042540920960.0,
      "grad_norm": 1.9558530616412129,
      "language_loss": 0.82543725,
      "learning_rate": 3.97836373413064e-06,
      "loss": 0.8484565,
      "num_input_tokens_seen": 13322205,
      "step": 628,
      "time_per_iteration": 2.5177395343780518
    },
    {
      "auxiliary_loss_clip": 0.01266088,
      "auxiliary_loss_mlp": 0.01043854,
      "balance_loss_clip": 1.0708847,
      "balance_loss_mlp": 1.03028774,
      "epoch": 0.07563277821198822,
      "flos": 19208761908480.0,
      "grad_norm": 1.9345176139540943,
      "language_loss": 0.74608195,
      "learning_rate": 3.978249313331848e-06,
      "loss": 0.76918143,
      "num_input_tokens_seen": 13340435,
      "step": 629,
      "time_per_iteration": 2.465623378753662
    },
    {
      "auxiliary_loss_clip": 0.01258161,
      "auxiliary_loss_mlp": 0.00766786,
      "balance_loss_clip": 1.06967878,
      "balance_loss_mlp": 1.00033998,
      "epoch": 0.07575302110262731,
      "flos": 19537200892800.0,
      "grad_norm": 3.378119891573478,
      "language_loss": 0.61919975,
      "learning_rate": 3.978134592434208e-06,
      "loss": 0.63944924,
      "num_input_tokens_seen": 13358185,
      "step": 630,
      "time_per_iteration": 2.564579486846924
    },
    {
      "auxiliary_loss_clip": 0.01114498,
      "auxiliary_loss_mlp": 0.01025124,
      "balance_loss_clip": 1.04549789,
      "balance_loss_mlp": 1.01830506,
      "epoch": 0.0758732639932664,
      "flos": 67961808017280.0,
      "grad_norm": 1.0213358593015807,
      "language_loss": 0.59412086,
      "learning_rate": 3.978019571455123e-06,
      "loss": 0.61551702,
      "num_input_tokens_seen": 13410130,
      "step": 631,
      "time_per_iteration": 3.167510509490967
    },
    {
      "auxiliary_loss_clip": 0.01265159,
      "auxiliary_loss_mlp": 0.01043403,
      "balance_loss_clip": 1.07312822,
      "balance_loss_mlp": 1.0312674,
      "epoch": 0.07599350688390549,
      "flos": 18989204025600.0,
      "grad_norm": 2.048854635764491,
      "language_loss": 0.84041762,
      "learning_rate": 3.977904250412042e-06,
      "loss": 0.86350322,
      "num_input_tokens_seen": 13429085,
      "step": 632,
      "time_per_iteration": 2.4932351112365723
    },
    {
      "auxiliary_loss_clip": 0.0121114,
      "auxiliary_loss_mlp": 0.01043917,
      "balance_loss_clip": 1.06574404,
      "balance_loss_mlp": 1.03105998,
      "epoch": 0.07611374977454458,
      "flos": 21069006341760.0,
      "grad_norm": 2.2991946161603662,
      "language_loss": 0.85610855,
      "learning_rate": 3.97778862932246e-06,
      "loss": 0.87865913,
      "num_input_tokens_seen": 13446250,
      "step": 633,
      "time_per_iteration": 2.5709784030914307
    },
    {
      "auxiliary_loss_clip": 0.01163959,
      "auxiliary_loss_mlp": 0.01039104,
      "balance_loss_clip": 1.05389297,
      "balance_loss_mlp": 1.02615809,
      "epoch": 0.07623399266518367,
      "flos": 18514536773760.0,
      "grad_norm": 2.2726817816136817,
      "language_loss": 0.94089937,
      "learning_rate": 3.9776727082039144e-06,
      "loss": 0.96293008,
      "num_input_tokens_seen": 13463220,
      "step": 634,
      "time_per_iteration": 3.648066520690918
    },
    {
      "auxiliary_loss_clip": 0.01176437,
      "auxiliary_loss_mlp": 0.01020059,
      "balance_loss_clip": 1.05035758,
      "balance_loss_mlp": 1.01281059,
      "epoch": 0.07635423555582276,
      "flos": 44663036077440.0,
      "grad_norm": 0.82104065496255,
      "language_loss": 0.55479491,
      "learning_rate": 3.977556487073991e-06,
      "loss": 0.57675982,
      "num_input_tokens_seen": 13517775,
      "step": 635,
      "time_per_iteration": 3.402374505996704
    },
    {
      "auxiliary_loss_clip": 0.01223577,
      "auxiliary_loss_mlp": 0.0104852,
      "balance_loss_clip": 1.06167388,
      "balance_loss_mlp": 1.03644371,
      "epoch": 0.07647447844646185,
      "flos": 21761148487680.0,
      "grad_norm": 1.946885550848919,
      "language_loss": 0.81662524,
      "learning_rate": 3.97743996595032e-06,
      "loss": 0.83934617,
      "num_input_tokens_seen": 13537815,
      "step": 636,
      "time_per_iteration": 2.5996227264404297
    },
    {
      "auxiliary_loss_clip": 0.01265604,
      "auxiliary_loss_mlp": 0.01049177,
      "balance_loss_clip": 1.0724721,
      "balance_loss_mlp": 1.03498518,
      "epoch": 0.07659472133710095,
      "flos": 23806799948160.0,
      "grad_norm": 1.6803141202601504,
      "language_loss": 0.8170054,
      "learning_rate": 3.9773231448505804e-06,
      "loss": 0.84015316,
      "num_input_tokens_seen": 13559605,
      "step": 637,
      "time_per_iteration": 2.537360191345215
    },
    {
      "auxiliary_loss_clip": 0.01230394,
      "auxiliary_loss_mlp": 0.0076715,
      "balance_loss_clip": 1.06851327,
      "balance_loss_mlp": 1.00029206,
      "epoch": 0.07671496422774003,
      "flos": 21469984842240.0,
      "grad_norm": 2.2124122697110415,
      "language_loss": 0.78051829,
      "learning_rate": 3.977206023792491e-06,
      "loss": 0.80049372,
      "num_input_tokens_seen": 13579495,
      "step": 638,
      "time_per_iteration": 3.310861349105835
    },
    {
      "auxiliary_loss_clip": 0.0125072,
      "auxiliary_loss_mlp": 0.01058308,
      "balance_loss_clip": 1.07406616,
      "balance_loss_mlp": 1.04571342,
      "epoch": 0.07683520711837913,
      "flos": 16980971558400.0,
      "grad_norm": 2.2335488895141045,
      "language_loss": 0.81255984,
      "learning_rate": 3.97708860279382e-06,
      "loss": 0.83565015,
      "num_input_tokens_seen": 13597605,
      "step": 639,
      "time_per_iteration": 3.3759865760803223
    },
    {
      "auxiliary_loss_clip": 0.01214848,
      "auxiliary_loss_mlp": 0.01053252,
      "balance_loss_clip": 1.06397736,
      "balance_loss_mlp": 1.03928065,
      "epoch": 0.07695545000901821,
      "flos": 23476744851840.0,
      "grad_norm": 1.8175020274815743,
      "language_loss": 0.78140259,
      "learning_rate": 3.97697088187238e-06,
      "loss": 0.80408359,
      "num_input_tokens_seen": 13618120,
      "step": 640,
      "time_per_iteration": 2.6446046829223633
    },
    {
      "auxiliary_loss_clip": 0.01230928,
      "auxiliary_loss_mlp": 0.01050001,
      "balance_loss_clip": 1.07108378,
      "balance_loss_mlp": 1.03768706,
      "epoch": 0.07707569289965731,
      "flos": 17634258167040.0,
      "grad_norm": 2.831044167807043,
      "language_loss": 0.92103654,
      "learning_rate": 3.976852861046029e-06,
      "loss": 0.94384587,
      "num_input_tokens_seen": 13634735,
      "step": 641,
      "time_per_iteration": 3.3709909915924072
    },
    {
      "auxiliary_loss_clip": 0.01201538,
      "auxiliary_loss_mlp": 0.01045002,
      "balance_loss_clip": 1.06465745,
      "balance_loss_mlp": 1.03225219,
      "epoch": 0.0771959357902964,
      "flos": 25775674087680.0,
      "grad_norm": 1.579703201189408,
      "language_loss": 0.80200726,
      "learning_rate": 3.97673454033267e-06,
      "loss": 0.82447267,
      "num_input_tokens_seen": 13656835,
      "step": 642,
      "time_per_iteration": 2.7125182151794434
    },
    {
      "auxiliary_loss_clip": 0.01229893,
      "auxiliary_loss_mlp": 0.01050755,
      "balance_loss_clip": 1.06527793,
      "balance_loss_mlp": 1.03763044,
      "epoch": 0.07731617868093549,
      "flos": 19828651847040.0,
      "grad_norm": 1.9451512243597169,
      "language_loss": 0.82321537,
      "learning_rate": 3.976615919750254e-06,
      "loss": 0.84602189,
      "num_input_tokens_seen": 13674535,
      "step": 643,
      "time_per_iteration": 2.5751240253448486
    },
    {
      "auxiliary_loss_clip": 0.01246034,
      "auxiliary_loss_mlp": 0.01049205,
      "balance_loss_clip": 1.06955707,
      "balance_loss_mlp": 1.03494167,
      "epoch": 0.07743642157157458,
      "flos": 21324654414720.0,
      "grad_norm": 2.091017236888172,
      "language_loss": 0.86677492,
      "learning_rate": 3.976496999316775e-06,
      "loss": 0.88972723,
      "num_input_tokens_seen": 13693290,
      "step": 644,
      "time_per_iteration": 2.5353384017944336
    },
    {
      "auxiliary_loss_clip": 0.01229732,
      "auxiliary_loss_mlp": 0.01049438,
      "balance_loss_clip": 1.07174492,
      "balance_loss_mlp": 1.03572929,
      "epoch": 0.07755666446221367,
      "flos": 19969133938560.0,
      "grad_norm": 2.1687205140379144,
      "language_loss": 0.84138197,
      "learning_rate": 3.976377779050271e-06,
      "loss": 0.86417365,
      "num_input_tokens_seen": 13711420,
      "step": 645,
      "time_per_iteration": 2.529776096343994
    },
    {
      "auxiliary_loss_clip": 0.01238387,
      "auxiliary_loss_mlp": 0.01052681,
      "balance_loss_clip": 1.06626809,
      "balance_loss_mlp": 1.03955626,
      "epoch": 0.07767690735285276,
      "flos": 23623224514560.0,
      "grad_norm": 2.204351951587358,
      "language_loss": 0.84668046,
      "learning_rate": 3.976258258968831e-06,
      "loss": 0.86959118,
      "num_input_tokens_seen": 13729965,
      "step": 646,
      "time_per_iteration": 2.5297060012817383
    },
    {
      "auxiliary_loss_clip": 0.01214031,
      "auxiliary_loss_mlp": 0.01049297,
      "balance_loss_clip": 1.06744981,
      "balance_loss_mlp": 1.03694677,
      "epoch": 0.07779715024349185,
      "flos": 22236246702720.0,
      "grad_norm": 2.2872897961186633,
      "language_loss": 0.7417531,
      "learning_rate": 3.976138439090583e-06,
      "loss": 0.76438642,
      "num_input_tokens_seen": 13748045,
      "step": 647,
      "time_per_iteration": 2.563839912414551
    },
    {
      "auxiliary_loss_clip": 0.01218976,
      "auxiliary_loss_mlp": 0.01041613,
      "balance_loss_clip": 1.06795955,
      "balance_loss_mlp": 1.02790403,
      "epoch": 0.07791739313413094,
      "flos": 20955097336320.0,
      "grad_norm": 2.080131721883903,
      "language_loss": 0.85294449,
      "learning_rate": 3.976018319433706e-06,
      "loss": 0.87555039,
      "num_input_tokens_seen": 13765590,
      "step": 648,
      "time_per_iteration": 2.6100010871887207
    },
    {
      "auxiliary_loss_clip": 0.01244642,
      "auxiliary_loss_mlp": 0.01045595,
      "balance_loss_clip": 1.06948924,
      "balance_loss_mlp": 1.03258371,
      "epoch": 0.07803763602477004,
      "flos": 19312327797120.0,
      "grad_norm": 2.919402372990991,
      "language_loss": 0.90911055,
      "learning_rate": 3.9758979000164205e-06,
      "loss": 0.93201292,
      "num_input_tokens_seen": 13782410,
      "step": 649,
      "time_per_iteration": 2.493290424346924
    },
    {
      "auxiliary_loss_clip": 0.01222736,
      "auxiliary_loss_mlp": 0.01038453,
      "balance_loss_clip": 1.06792569,
      "balance_loss_mlp": 1.02435088,
      "epoch": 0.07815787891540912,
      "flos": 22710806213760.0,
      "grad_norm": 1.9905593002458517,
      "language_loss": 0.71715081,
      "learning_rate": 3.975777180856995e-06,
      "loss": 0.73976272,
      "num_input_tokens_seen": 13801530,
      "step": 650,
      "time_per_iteration": 2.5844192504882812
    },
    {
      "auxiliary_loss_clip": 0.01269164,
      "auxiliary_loss_mlp": 0.01052243,
      "balance_loss_clip": 1.07210231,
      "balance_loss_mlp": 1.03818274,
      "epoch": 0.07827812180604822,
      "flos": 22711129436160.0,
      "grad_norm": 1.9965005916549772,
      "language_loss": 0.86122298,
      "learning_rate": 3.975656161973742e-06,
      "loss": 0.88443708,
      "num_input_tokens_seen": 13820615,
      "step": 651,
      "time_per_iteration": 2.4871466159820557
    },
    {
      "auxiliary_loss_clip": 0.01266259,
      "auxiliary_loss_mlp": 0.01048124,
      "balance_loss_clip": 1.0703311,
      "balance_loss_mlp": 1.03427768,
      "epoch": 0.0783983646966873,
      "flos": 21725597001600.0,
      "grad_norm": 2.5839022706032084,
      "language_loss": 0.89153659,
      "learning_rate": 3.9755348433850194e-06,
      "loss": 0.91468048,
      "num_input_tokens_seen": 13835955,
      "step": 652,
      "time_per_iteration": 2.4620301723480225
    },
    {
      "auxiliary_loss_clip": 0.01129248,
      "auxiliary_loss_mlp": 0.0102131,
      "balance_loss_clip": 1.03415215,
      "balance_loss_mlp": 1.01501548,
      "epoch": 0.0785186075873264,
      "flos": 60640877537280.0,
      "grad_norm": 0.9887619338597119,
      "language_loss": 0.63607579,
      "learning_rate": 3.975413225109232e-06,
      "loss": 0.65758133,
      "num_input_tokens_seen": 13896505,
      "step": 653,
      "time_per_iteration": 3.128957748413086
    },
    {
      "auxiliary_loss_clip": 0.01247449,
      "auxiliary_loss_mlp": 0.01043969,
      "balance_loss_clip": 1.06966519,
      "balance_loss_mlp": 1.03072464,
      "epoch": 0.0786388504779655,
      "flos": 23877902920320.0,
      "grad_norm": 3.090177528024159,
      "language_loss": 0.93495977,
      "learning_rate": 3.975291307164829e-06,
      "loss": 0.95787394,
      "num_input_tokens_seen": 13915150,
      "step": 654,
      "time_per_iteration": 2.518588066101074
    },
    {
      "auxiliary_loss_clip": 0.01203009,
      "auxiliary_loss_mlp": 0.01042327,
      "balance_loss_clip": 1.06245482,
      "balance_loss_mlp": 1.03022122,
      "epoch": 0.07875909336860458,
      "flos": 15158684822400.0,
      "grad_norm": 2.3712801821022276,
      "language_loss": 0.85193503,
      "learning_rate": 3.975169089570306e-06,
      "loss": 0.8743884,
      "num_input_tokens_seen": 13933525,
      "step": 655,
      "time_per_iteration": 2.5760207176208496
    },
    {
      "auxiliary_loss_clip": 0.01231263,
      "auxiliary_loss_mlp": 0.01044512,
      "balance_loss_clip": 1.06589484,
      "balance_loss_mlp": 1.03122044,
      "epoch": 0.07887933625924368,
      "flos": 22236857233920.0,
      "grad_norm": 2.017975083997081,
      "language_loss": 0.91828609,
      "learning_rate": 3.975046572344202e-06,
      "loss": 0.94104385,
      "num_input_tokens_seen": 13949985,
      "step": 656,
      "time_per_iteration": 2.5114076137542725
    },
    {
      "auxiliary_loss_clip": 0.01208942,
      "auxiliary_loss_mlp": 0.0105043,
      "balance_loss_clip": 1.06128097,
      "balance_loss_mlp": 1.0372932,
      "epoch": 0.07899957914988276,
      "flos": 20777734955520.0,
      "grad_norm": 1.9462314733049724,
      "language_loss": 0.71072853,
      "learning_rate": 3.974923755505103e-06,
      "loss": 0.73332226,
      "num_input_tokens_seen": 13969215,
      "step": 657,
      "time_per_iteration": 2.5772500038146973
    },
    {
      "auxiliary_loss_clip": 0.01204601,
      "auxiliary_loss_mlp": 0.01043775,
      "balance_loss_clip": 1.06278753,
      "balance_loss_mlp": 1.03082943,
      "epoch": 0.07911982204052186,
      "flos": 23003047267200.0,
      "grad_norm": 1.7194952265079781,
      "language_loss": 0.91309035,
      "learning_rate": 3.974800639071641e-06,
      "loss": 0.93557405,
      "num_input_tokens_seen": 13989935,
      "step": 658,
      "time_per_iteration": 2.632525682449341
    },
    {
      "auxiliary_loss_clip": 0.01170711,
      "auxiliary_loss_mlp": 0.0076629,
      "balance_loss_clip": 1.05856371,
      "balance_loss_mlp": 1.00033998,
      "epoch": 0.07924006493116094,
      "flos": 23111389664640.0,
      "grad_norm": 2.136139036446059,
      "language_loss": 1.00600171,
      "learning_rate": 3.974677223062492e-06,
      "loss": 1.02537167,
      "num_input_tokens_seen": 14007150,
      "step": 659,
      "time_per_iteration": 2.723534107208252
    },
    {
      "auxiliary_loss_clip": 0.0122892,
      "auxiliary_loss_mlp": 0.01038667,
      "balance_loss_clip": 1.06812882,
      "balance_loss_mlp": 1.02604234,
      "epoch": 0.07936030782180004,
      "flos": 16472153450880.0,
      "grad_norm": 1.9440289711396954,
      "language_loss": 0.74405819,
      "learning_rate": 3.974553507496378e-06,
      "loss": 0.76673406,
      "num_input_tokens_seen": 14025725,
      "step": 660,
      "time_per_iteration": 3.3443922996520996
    },
    {
      "auxiliary_loss_clip": 0.01222159,
      "auxiliary_loss_mlp": 0.01042804,
      "balance_loss_clip": 1.0670166,
      "balance_loss_mlp": 1.02789712,
      "epoch": 0.07948055071243913,
      "flos": 23733290764800.0,
      "grad_norm": 2.3643717075132424,
      "language_loss": 0.88975132,
      "learning_rate": 3.974429492392068e-06,
      "loss": 0.91240096,
      "num_input_tokens_seen": 14045750,
      "step": 661,
      "time_per_iteration": 2.5975606441497803
    },
    {
      "auxiliary_loss_clip": 0.01260346,
      "auxiliary_loss_mlp": 0.00766076,
      "balance_loss_clip": 1.07208467,
      "balance_loss_mlp": 1.0003016,
      "epoch": 0.07960079360307822,
      "flos": 19573326996480.0,
      "grad_norm": 2.0034913793044646,
      "language_loss": 0.91186309,
      "learning_rate": 3.974305177768373e-06,
      "loss": 0.93212724,
      "num_input_tokens_seen": 14063960,
      "step": 662,
      "time_per_iteration": 2.5311028957366943
    },
    {
      "auxiliary_loss_clip": 0.0120549,
      "auxiliary_loss_mlp": 0.01049673,
      "balance_loss_clip": 1.06485152,
      "balance_loss_mlp": 1.03619099,
      "epoch": 0.07972103649371731,
      "flos": 23513409659520.0,
      "grad_norm": 3.2858681107034564,
      "language_loss": 0.86359072,
      "learning_rate": 3.974180563644152e-06,
      "loss": 0.88614237,
      "num_input_tokens_seen": 14082525,
      "step": 663,
      "time_per_iteration": 2.5958173274993896
    },
    {
      "auxiliary_loss_clip": 0.01234112,
      "auxiliary_loss_mlp": 0.01045875,
      "balance_loss_clip": 1.06768823,
      "balance_loss_mlp": 1.03302395,
      "epoch": 0.0798412793843564,
      "flos": 16726867770240.0,
      "grad_norm": 2.529548067039062,
      "language_loss": 0.89450055,
      "learning_rate": 3.97405565003831e-06,
      "loss": 0.91730046,
      "num_input_tokens_seen": 14098610,
      "step": 664,
      "time_per_iteration": 3.256565570831299
    },
    {
      "auxiliary_loss_clip": 0.01213461,
      "auxiliary_loss_mlp": 0.01037467,
      "balance_loss_clip": 1.06377316,
      "balance_loss_mlp": 1.02453303,
      "epoch": 0.07996152227499549,
      "flos": 18223337214720.0,
      "grad_norm": 1.9961695254687806,
      "language_loss": 0.78604126,
      "learning_rate": 3.973930436969794e-06,
      "loss": 0.80855048,
      "num_input_tokens_seen": 14117065,
      "step": 665,
      "time_per_iteration": 2.5939548015594482
    },
    {
      "auxiliary_loss_clip": 0.01220187,
      "auxiliary_loss_mlp": 0.01050269,
      "balance_loss_clip": 1.06355524,
      "balance_loss_mlp": 1.03709602,
      "epoch": 0.08008176516563459,
      "flos": 20594877793920.0,
      "grad_norm": 1.9619162619694925,
      "language_loss": 0.85930985,
      "learning_rate": 3.973804924457602e-06,
      "loss": 0.88201445,
      "num_input_tokens_seen": 14135145,
      "step": 666,
      "time_per_iteration": 3.353271484375
    },
    {
      "auxiliary_loss_clip": 0.01221835,
      "auxiliary_loss_mlp": 0.01058227,
      "balance_loss_clip": 1.06583881,
      "balance_loss_mlp": 1.04523349,
      "epoch": 0.08020200805627367,
      "flos": 31834306863360.0,
      "grad_norm": 1.9799605433536214,
      "language_loss": 0.85940409,
      "learning_rate": 3.973679112520771e-06,
      "loss": 0.88220471,
      "num_input_tokens_seen": 14156860,
      "step": 667,
      "time_per_iteration": 2.6472995281219482
    },
    {
      "auxiliary_loss_clip": 0.01203249,
      "auxiliary_loss_mlp": 0.01038871,
      "balance_loss_clip": 1.06058013,
      "balance_loss_mlp": 1.02604365,
      "epoch": 0.08032225094691277,
      "flos": 17783503176960.0,
      "grad_norm": 1.9834034320635612,
      "language_loss": 0.99088603,
      "learning_rate": 3.973553001178389e-06,
      "loss": 1.01330733,
      "num_input_tokens_seen": 14174365,
      "step": 668,
      "time_per_iteration": 3.356459379196167
    },
    {
      "auxiliary_loss_clip": 0.01215875,
      "auxiliary_loss_mlp": 0.01037173,
      "balance_loss_clip": 1.06615376,
      "balance_loss_mlp": 1.02455425,
      "epoch": 0.08044249383755185,
      "flos": 24061693835520.0,
      "grad_norm": 2.304705174710886,
      "language_loss": 0.75877726,
      "learning_rate": 3.973426590449585e-06,
      "loss": 0.7813077,
      "num_input_tokens_seen": 14192320,
      "step": 669,
      "time_per_iteration": 2.5939626693725586
    },
    {
      "auxiliary_loss_clip": 0.01198574,
      "auxiliary_loss_mlp": 0.01039915,
      "balance_loss_clip": 1.06314945,
      "balance_loss_mlp": 1.02717161,
      "epoch": 0.08056273672819095,
      "flos": 18223624523520.0,
      "grad_norm": 1.8725881784779344,
      "language_loss": 0.75441134,
      "learning_rate": 3.9732998803535364e-06,
      "loss": 0.77679622,
      "num_input_tokens_seen": 14210380,
      "step": 670,
      "time_per_iteration": 2.5770716667175293
    },
    {
      "auxiliary_loss_clip": 0.01262434,
      "auxiliary_loss_mlp": 0.01044146,
      "balance_loss_clip": 1.0697782,
      "balance_loss_mlp": 1.03130746,
      "epoch": 0.08068297961883003,
      "flos": 19676856971520.0,
      "grad_norm": 2.540563615216749,
      "language_loss": 0.85565472,
      "learning_rate": 3.973172870909465e-06,
      "loss": 0.87872052,
      "num_input_tokens_seen": 14225145,
      "step": 671,
      "time_per_iteration": 2.4702680110931396
    },
    {
      "auxiliary_loss_clip": 0.01235763,
      "auxiliary_loss_mlp": 0.01040664,
      "balance_loss_clip": 1.06584334,
      "balance_loss_mlp": 1.02686596,
      "epoch": 0.08080322250946913,
      "flos": 23148736830720.0,
      "grad_norm": 2.9818086536384993,
      "language_loss": 0.80207551,
      "learning_rate": 3.973045562136638e-06,
      "loss": 0.82483977,
      "num_input_tokens_seen": 14241960,
      "step": 672,
      "time_per_iteration": 2.5467593669891357
    },
    {
      "auxiliary_loss_clip": 0.01249875,
      "auxiliary_loss_mlp": 0.01041783,
      "balance_loss_clip": 1.06946349,
      "balance_loss_mlp": 1.02912903,
      "epoch": 0.08092346540010822,
      "flos": 21763626526080.0,
      "grad_norm": 2.2014684530543898,
      "language_loss": 0.91550684,
      "learning_rate": 3.972917954054368e-06,
      "loss": 0.93842345,
      "num_input_tokens_seen": 14260515,
      "step": 673,
      "time_per_iteration": 2.5179567337036133
    },
    {
      "auxiliary_loss_clip": 0.01229938,
      "auxiliary_loss_mlp": 0.01047212,
      "balance_loss_clip": 1.07049966,
      "balance_loss_mlp": 1.03244793,
      "epoch": 0.08104370829074731,
      "flos": 21032485188480.0,
      "grad_norm": 2.1421099713565432,
      "language_loss": 0.81740284,
      "learning_rate": 3.972790046682013e-06,
      "loss": 0.84017432,
      "num_input_tokens_seen": 14279190,
      "step": 674,
      "time_per_iteration": 2.5228404998779297
    },
    {
      "auxiliary_loss_clip": 0.01214516,
      "auxiliary_loss_mlp": 0.01041553,
      "balance_loss_clip": 1.06217706,
      "balance_loss_mlp": 1.02849913,
      "epoch": 0.0811639511813864,
      "flos": 20083186598400.0,
      "grad_norm": 1.803704715733776,
      "language_loss": 0.78986847,
      "learning_rate": 3.972661840038977e-06,
      "loss": 0.81242919,
      "num_input_tokens_seen": 14299480,
      "step": 675,
      "time_per_iteration": 2.58609676361084
    },
    {
      "auxiliary_loss_clip": 0.01248848,
      "auxiliary_loss_mlp": 0.01040152,
      "balance_loss_clip": 1.07150388,
      "balance_loss_mlp": 1.02758765,
      "epoch": 0.08128419407202549,
      "flos": 16836718538880.0,
      "grad_norm": 2.1729115301359716,
      "language_loss": 0.83291638,
      "learning_rate": 3.972533334144707e-06,
      "loss": 0.85580635,
      "num_input_tokens_seen": 14316405,
      "step": 676,
      "time_per_iteration": 2.4742016792297363
    },
    {
      "auxiliary_loss_clip": 0.01250297,
      "auxiliary_loss_mlp": 0.01043874,
      "balance_loss_clip": 1.06717348,
      "balance_loss_mlp": 1.03067183,
      "epoch": 0.08140443696266458,
      "flos": 23769273214080.0,
      "grad_norm": 2.422506621434802,
      "language_loss": 0.783234,
      "learning_rate": 3.972404529018699e-06,
      "loss": 0.80617571,
      "num_input_tokens_seen": 14336265,
      "step": 677,
      "time_per_iteration": 2.538540840148926
    },
    {
      "auxiliary_loss_clip": 0.01223953,
      "auxiliary_loss_mlp": 0.01035828,
      "balance_loss_clip": 1.06134868,
      "balance_loss_mlp": 1.02343643,
      "epoch": 0.08152467985330367,
      "flos": 24390132819840.0,
      "grad_norm": 1.984735276641255,
      "language_loss": 0.85500354,
      "learning_rate": 3.972275424680493e-06,
      "loss": 0.87760133,
      "num_input_tokens_seen": 14356375,
      "step": 678,
      "time_per_iteration": 2.5549771785736084
    },
    {
      "auxiliary_loss_clip": 0.01259167,
      "auxiliary_loss_mlp": 0.01037917,
      "balance_loss_clip": 1.06892157,
      "balance_loss_mlp": 1.0253942,
      "epoch": 0.08164492274394276,
      "flos": 19317750750720.0,
      "grad_norm": 2.101815434680312,
      "language_loss": 0.91848153,
      "learning_rate": 3.972146021149673e-06,
      "loss": 0.94145238,
      "num_input_tokens_seen": 14374650,
      "step": 679,
      "time_per_iteration": 2.4407427310943604
    },
    {
      "auxiliary_loss_clip": 0.01213828,
      "auxiliary_loss_mlp": 0.01040496,
      "balance_loss_clip": 1.06540799,
      "balance_loss_mlp": 1.02911127,
      "epoch": 0.08176516563458186,
      "flos": 14830461319680.0,
      "grad_norm": 3.178083159867089,
      "language_loss": 0.7836771,
      "learning_rate": 3.972016318445868e-06,
      "loss": 0.80622035,
      "num_input_tokens_seen": 14392650,
      "step": 680,
      "time_per_iteration": 2.5333454608917236
    },
    {
      "auxiliary_loss_clip": 0.01243865,
      "auxiliary_loss_mlp": 0.01046605,
      "balance_loss_clip": 1.0676018,
      "balance_loss_mlp": 1.03423095,
      "epoch": 0.08188540852522094,
      "flos": 22602320161920.0,
      "grad_norm": 2.239080391683848,
      "language_loss": 0.92591882,
      "learning_rate": 3.971886316588757e-06,
      "loss": 0.94882351,
      "num_input_tokens_seen": 14413155,
      "step": 681,
      "time_per_iteration": 2.5087900161743164
    },
    {
      "auxiliary_loss_clip": 0.01204447,
      "auxiliary_loss_mlp": 0.01049683,
      "balance_loss_clip": 1.06499052,
      "balance_loss_mlp": 1.03581882,
      "epoch": 0.08200565141586004,
      "flos": 19463727623040.0,
      "grad_norm": 2.4840795439774612,
      "language_loss": 0.73761892,
      "learning_rate": 3.9717560155980595e-06,
      "loss": 0.76016021,
      "num_input_tokens_seen": 14428805,
      "step": 682,
      "time_per_iteration": 2.557297706604004
    },
    {
      "auxiliary_loss_clip": 0.01245373,
      "auxiliary_loss_mlp": 0.01045271,
      "balance_loss_clip": 1.06866574,
      "balance_loss_mlp": 1.03249168,
      "epoch": 0.08212589430649912,
      "flos": 20594662312320.0,
      "grad_norm": 2.6635650868617953,
      "language_loss": 0.92139184,
      "learning_rate": 3.971625415493542e-06,
      "loss": 0.94429833,
      "num_input_tokens_seen": 14447125,
      "step": 683,
      "time_per_iteration": 2.495816469192505
    },
    {
      "auxiliary_loss_clip": 0.0120769,
      "auxiliary_loss_mlp": 0.01042311,
      "balance_loss_clip": 1.06395745,
      "balance_loss_mlp": 1.02907932,
      "epoch": 0.08224613719713822,
      "flos": 25953611086080.0,
      "grad_norm": 1.8735624448812276,
      "language_loss": 0.87464941,
      "learning_rate": 3.971494516295017e-06,
      "loss": 0.89714932,
      "num_input_tokens_seen": 14466575,
      "step": 684,
      "time_per_iteration": 2.5975239276885986
    },
    {
      "auxiliary_loss_clip": 0.01216489,
      "auxiliary_loss_mlp": 0.0104531,
      "balance_loss_clip": 1.06358957,
      "balance_loss_mlp": 1.03220904,
      "epoch": 0.08236638008777732,
      "flos": 23768734510080.0,
      "grad_norm": 1.977049678289977,
      "language_loss": 0.85380328,
      "learning_rate": 3.971363318022341e-06,
      "loss": 0.87642121,
      "num_input_tokens_seen": 14487915,
      "step": 685,
      "time_per_iteration": 2.5909695625305176
    },
    {
      "auxiliary_loss_clip": 0.0123084,
      "auxiliary_loss_mlp": 0.01047554,
      "balance_loss_clip": 1.06277061,
      "balance_loss_mlp": 1.03398764,
      "epoch": 0.0824866229784164,
      "flos": 38799144887040.0,
      "grad_norm": 2.320995925729637,
      "language_loss": 0.68715346,
      "learning_rate": 3.971231820695417e-06,
      "loss": 0.70993733,
      "num_input_tokens_seen": 14511530,
      "step": 686,
      "time_per_iteration": 2.6729483604431152
    },
    {
      "auxiliary_loss_clip": 0.01236806,
      "auxiliary_loss_mlp": 0.01045352,
      "balance_loss_clip": 1.06897497,
      "balance_loss_mlp": 1.03226888,
      "epoch": 0.0826068658690555,
      "flos": 23107762391040.0,
      "grad_norm": 1.7689594334367413,
      "language_loss": 0.81249619,
      "learning_rate": 3.971100024334193e-06,
      "loss": 0.83531773,
      "num_input_tokens_seen": 14529050,
      "step": 687,
      "time_per_iteration": 3.4973368644714355
    },
    {
      "auxiliary_loss_clip": 0.01195503,
      "auxiliary_loss_mlp": 0.01045408,
      "balance_loss_clip": 1.05743408,
      "balance_loss_mlp": 1.03322434,
      "epoch": 0.08272710875969458,
      "flos": 21136374299520.0,
      "grad_norm": 2.556138440075114,
      "language_loss": 0.86235952,
      "learning_rate": 3.970967928958663e-06,
      "loss": 0.88476861,
      "num_input_tokens_seen": 14546165,
      "step": 688,
      "time_per_iteration": 2.646243095397949
    },
    {
      "auxiliary_loss_clip": 0.01202286,
      "auxiliary_loss_mlp": 0.01048323,
      "balance_loss_clip": 1.06436276,
      "balance_loss_mlp": 1.03584182,
      "epoch": 0.08284735165033368,
      "flos": 19063000517760.0,
      "grad_norm": 1.6641429196113793,
      "language_loss": 0.83409357,
      "learning_rate": 3.970835534588865e-06,
      "loss": 0.85659969,
      "num_input_tokens_seen": 14563660,
      "step": 689,
      "time_per_iteration": 2.590881109237671
    },
    {
      "auxiliary_loss_clip": 0.01231876,
      "auxiliary_loss_mlp": 0.0104943,
      "balance_loss_clip": 1.06994677,
      "balance_loss_mlp": 1.03721166,
      "epoch": 0.08296759454097276,
      "flos": 16727442387840.0,
      "grad_norm": 1.9299228935292398,
      "language_loss": 0.85633647,
      "learning_rate": 3.970702841244883e-06,
      "loss": 0.87914944,
      "num_input_tokens_seen": 14581980,
      "step": 690,
      "time_per_iteration": 2.5300991535186768
    },
    {
      "auxiliary_loss_clip": 0.01249433,
      "auxiliary_loss_mlp": 0.01044814,
      "balance_loss_clip": 1.07060182,
      "balance_loss_mlp": 1.03250003,
      "epoch": 0.08308783743161186,
      "flos": 18004928567040.0,
      "grad_norm": 2.3234322801547704,
      "language_loss": 0.82538581,
      "learning_rate": 3.970569848946847e-06,
      "loss": 0.84832823,
      "num_input_tokens_seen": 14601795,
      "step": 691,
      "time_per_iteration": 3.330634355545044
    },
    {
      "auxiliary_loss_clip": 0.01230133,
      "auxiliary_loss_mlp": 0.01039325,
      "balance_loss_clip": 1.06582785,
      "balance_loss_mlp": 1.02703404,
      "epoch": 0.08320808032225095,
      "flos": 15079788599040.0,
      "grad_norm": 2.413250600106508,
      "language_loss": 0.82537591,
      "learning_rate": 3.970436557714932e-06,
      "loss": 0.8480705,
      "num_input_tokens_seen": 14618315,
      "step": 692,
      "time_per_iteration": 2.54731822013855
    },
    {
      "auxiliary_loss_clip": 0.01222118,
      "auxiliary_loss_mlp": 0.01039079,
      "balance_loss_clip": 1.06346726,
      "balance_loss_mlp": 1.02607346,
      "epoch": 0.08332832321289003,
      "flos": 22383085501440.0,
      "grad_norm": 8.100572751541788,
      "language_loss": 0.86280978,
      "learning_rate": 3.970302967569358e-06,
      "loss": 0.88542175,
      "num_input_tokens_seen": 14636905,
      "step": 693,
      "time_per_iteration": 3.327517509460449
    },
    {
      "auxiliary_loss_clip": 0.01247305,
      "auxiliary_loss_mlp": 0.01045359,
      "balance_loss_clip": 1.07215595,
      "balance_loss_mlp": 1.0325501,
      "epoch": 0.08344856610352913,
      "flos": 24717386655360.0,
      "grad_norm": 1.901300033673213,
      "language_loss": 0.68107098,
      "learning_rate": 3.9701690785303896e-06,
      "loss": 0.70399761,
      "num_input_tokens_seen": 14656100,
      "step": 694,
      "time_per_iteration": 2.547891139984131
    },
    {
      "auxiliary_loss_clip": 0.01250563,
      "auxiliary_loss_mlp": 0.01042239,
      "balance_loss_clip": 1.06959915,
      "balance_loss_mlp": 1.02993703,
      "epoch": 0.08356880899416821,
      "flos": 25370206387200.0,
      "grad_norm": 2.2751908071743117,
      "language_loss": 0.88450772,
      "learning_rate": 3.970034890618339e-06,
      "loss": 0.90743572,
      "num_input_tokens_seen": 14675790,
      "step": 695,
      "time_per_iteration": 3.253553867340088
    },
    {
      "auxiliary_loss_clip": 0.01229564,
      "auxiliary_loss_mlp": 0.01037178,
      "balance_loss_clip": 1.06536865,
      "balance_loss_mlp": 1.02533412,
      "epoch": 0.08368905188480731,
      "flos": 24353072962560.0,
      "grad_norm": 1.9940795528678446,
      "language_loss": 0.88021863,
      "learning_rate": 3.969900403853562e-06,
      "loss": 0.90288603,
      "num_input_tokens_seen": 14694830,
      "step": 696,
      "time_per_iteration": 2.5634326934814453
    },
    {
      "auxiliary_loss_clip": 0.01265723,
      "auxiliary_loss_mlp": 0.01052848,
      "balance_loss_clip": 1.07358658,
      "balance_loss_mlp": 1.03993201,
      "epoch": 0.08380929477544641,
      "flos": 18037319656320.0,
      "grad_norm": 1.5595384805472767,
      "language_loss": 0.77765125,
      "learning_rate": 3.96976561825646e-06,
      "loss": 0.80083692,
      "num_input_tokens_seen": 14711920,
      "step": 697,
      "time_per_iteration": 2.472980260848999
    },
    {
      "auxiliary_loss_clip": 0.01198585,
      "auxiliary_loss_mlp": 0.01038486,
      "balance_loss_clip": 1.06354189,
      "balance_loss_mlp": 1.0268271,
      "epoch": 0.08392953766608549,
      "flos": 26286287875200.0,
      "grad_norm": 2.1700018346639642,
      "language_loss": 0.87154067,
      "learning_rate": 3.969630533847479e-06,
      "loss": 0.89391136,
      "num_input_tokens_seen": 14730880,
      "step": 698,
      "time_per_iteration": 2.649151563644409
    },
    {
      "auxiliary_loss_clip": 0.01246313,
      "auxiliary_loss_mlp": 0.01039335,
      "balance_loss_clip": 1.06890726,
      "balance_loss_mlp": 1.02702081,
      "epoch": 0.08404978055672459,
      "flos": 22492146170880.0,
      "grad_norm": 2.1303424371162727,
      "language_loss": 0.84012669,
      "learning_rate": 3.969495150647113e-06,
      "loss": 0.86298317,
      "num_input_tokens_seen": 14749050,
      "step": 699,
      "time_per_iteration": 2.5055344104766846
    },
    {
      "auxiliary_loss_clip": 0.01210867,
      "auxiliary_loss_mlp": 0.01037475,
      "balance_loss_clip": 1.06637263,
      "balance_loss_mlp": 1.02541149,
      "epoch": 0.08417002344736367,
      "flos": 24826878288000.0,
      "grad_norm": 1.7501081576618442,
      "language_loss": 0.76538599,
      "learning_rate": 3.969359468675899e-06,
      "loss": 0.78786945,
      "num_input_tokens_seen": 14769180,
      "step": 700,
      "time_per_iteration": 2.6334521770477295
    },
    {
      "auxiliary_loss_clip": 0.01241755,
      "auxiliary_loss_mlp": 0.01038554,
      "balance_loss_clip": 1.06796682,
      "balance_loss_mlp": 1.02598906,
      "epoch": 0.08429026633800277,
      "flos": 16945922862720.0,
      "grad_norm": 2.099904727483784,
      "language_loss": 0.89646226,
      "learning_rate": 3.969223487954418e-06,
      "loss": 0.91926539,
      "num_input_tokens_seen": 14786640,
      "step": 701,
      "time_per_iteration": 2.493617057800293
    },
    {
      "auxiliary_loss_clip": 0.01199374,
      "auxiliary_loss_mlp": 0.01045654,
      "balance_loss_clip": 1.06746805,
      "balance_loss_mlp": 1.03342295,
      "epoch": 0.08441050922864185,
      "flos": 23841920471040.0,
      "grad_norm": 2.2777933330410383,
      "language_loss": 0.82761592,
      "learning_rate": 3.969087208503301e-06,
      "loss": 0.85006618,
      "num_input_tokens_seen": 14806720,
      "step": 702,
      "time_per_iteration": 2.6200878620147705
    },
    {
      "auxiliary_loss_clip": 0.01201294,
      "auxiliary_loss_mlp": 0.01043453,
      "balance_loss_clip": 1.06502533,
      "balance_loss_mlp": 1.03105485,
      "epoch": 0.08453075211928095,
      "flos": 25520205582720.0,
      "grad_norm": 2.426950836105903,
      "language_loss": 0.84487534,
      "learning_rate": 3.968950630343219e-06,
      "loss": 0.8673228,
      "num_input_tokens_seen": 14823705,
      "step": 703,
      "time_per_iteration": 2.6058740615844727
    },
    {
      "auxiliary_loss_clip": 0.01226597,
      "auxiliary_loss_mlp": 0.01044622,
      "balance_loss_clip": 1.06400526,
      "balance_loss_mlp": 1.03252172,
      "epoch": 0.08465099500992004,
      "flos": 19532496211200.0,
      "grad_norm": 1.9550391882105205,
      "language_loss": 0.93532425,
      "learning_rate": 3.968813753494892e-06,
      "loss": 0.95803642,
      "num_input_tokens_seen": 14841865,
      "step": 704,
      "time_per_iteration": 2.548114538192749
    },
    {
      "auxiliary_loss_clip": 0.01200686,
      "auxiliary_loss_mlp": 0.00766973,
      "balance_loss_clip": 1.05830419,
      "balance_loss_mlp": 1.00026751,
      "epoch": 0.08477123790055913,
      "flos": 29351299403520.0,
      "grad_norm": 2.5948731469874735,
      "language_loss": 0.75383914,
      "learning_rate": 3.968676577979084e-06,
      "loss": 0.77351582,
      "num_input_tokens_seen": 14861415,
      "step": 705,
      "time_per_iteration": 2.655470609664917
    },
    {
      "auxiliary_loss_clip": 0.0118913,
      "auxiliary_loss_mlp": 0.01050092,
      "balance_loss_clip": 1.05848932,
      "balance_loss_mlp": 1.03756881,
      "epoch": 0.08489148079119822,
      "flos": 18624495283200.0,
      "grad_norm": 13.204197203013443,
      "language_loss": 0.78260064,
      "learning_rate": 3.968539103816605e-06,
      "loss": 0.80499291,
      "num_input_tokens_seen": 14879215,
      "step": 706,
      "time_per_iteration": 2.5678460597991943
    },
    {
      "auxiliary_loss_clip": 0.01228512,
      "auxiliary_loss_mlp": 0.00766381,
      "balance_loss_clip": 1.06937885,
      "balance_loss_mlp": 1.00028443,
      "epoch": 0.0850117236818373,
      "flos": 23471393725440.0,
      "grad_norm": 1.8222376117592827,
      "language_loss": 0.89680481,
      "learning_rate": 3.9684013310283085e-06,
      "loss": 0.91675377,
      "num_input_tokens_seen": 14897900,
      "step": 707,
      "time_per_iteration": 2.6264874935150146
    },
    {
      "auxiliary_loss_clip": 0.01224558,
      "auxiliary_loss_mlp": 0.01050407,
      "balance_loss_clip": 1.06930017,
      "balance_loss_mlp": 1.03790176,
      "epoch": 0.0851319665724764,
      "flos": 40625058896640.0,
      "grad_norm": 1.8237705906278956,
      "language_loss": 0.63923478,
      "learning_rate": 3.9682632596350956e-06,
      "loss": 0.66198444,
      "num_input_tokens_seen": 14919065,
      "step": 708,
      "time_per_iteration": 2.7082176208496094
    },
    {
      "auxiliary_loss_clip": 0.01240995,
      "auxiliary_loss_mlp": 0.01039717,
      "balance_loss_clip": 1.06945312,
      "balance_loss_mlp": 1.02738512,
      "epoch": 0.0852522094631155,
      "flos": 15879554870400.0,
      "grad_norm": 2.556451634365381,
      "language_loss": 0.78258598,
      "learning_rate": 3.968124889657911e-06,
      "loss": 0.80539304,
      "num_input_tokens_seen": 14934165,
      "step": 709,
      "time_per_iteration": 2.4888973236083984
    },
    {
      "auxiliary_loss_clip": 0.01193257,
      "auxiliary_loss_mlp": 0.0104827,
      "balance_loss_clip": 1.06094432,
      "balance_loss_mlp": 1.03685558,
      "epoch": 0.08537245235375458,
      "flos": 14567091822720.0,
      "grad_norm": 2.1951517339270437,
      "language_loss": 0.90588379,
      "learning_rate": 3.967986221117746e-06,
      "loss": 0.92829907,
      "num_input_tokens_seen": 14950105,
      "step": 710,
      "time_per_iteration": 2.567375421524048
    },
    {
      "auxiliary_loss_clip": 0.01169918,
      "auxiliary_loss_mlp": 0.01038101,
      "balance_loss_clip": 1.05987644,
      "balance_loss_mlp": 1.02613211,
      "epoch": 0.08549269524439368,
      "flos": 26468929555200.0,
      "grad_norm": 1.762491144624555,
      "language_loss": 0.86440134,
      "learning_rate": 3.967847254035635e-06,
      "loss": 0.88648152,
      "num_input_tokens_seen": 14969490,
      "step": 711,
      "time_per_iteration": 2.763526201248169
    },
    {
      "auxiliary_loss_clip": 0.01210719,
      "auxiliary_loss_mlp": 0.01041439,
      "balance_loss_clip": 1.0624547,
      "balance_loss_mlp": 1.02914262,
      "epoch": 0.08561293813503276,
      "flos": 13590214565760.0,
      "grad_norm": 2.1984781828452333,
      "language_loss": 0.86634451,
      "learning_rate": 3.967707988432661e-06,
      "loss": 0.88886607,
      "num_input_tokens_seen": 14987195,
      "step": 712,
      "time_per_iteration": 2.5990025997161865
    },
    {
      "auxiliary_loss_clip": 0.01261284,
      "auxiliary_loss_mlp": 0.01046279,
      "balance_loss_clip": 1.06909609,
      "balance_loss_mlp": 1.03321981,
      "epoch": 0.08573318102567186,
      "flos": 26943524979840.0,
      "grad_norm": 2.3147700761463725,
      "language_loss": 0.87544751,
      "learning_rate": 3.967568424329949e-06,
      "loss": 0.89852315,
      "num_input_tokens_seen": 15007620,
      "step": 713,
      "time_per_iteration": 3.3077073097229004
    },
    {
      "auxiliary_loss_clip": 0.01133771,
      "auxiliary_loss_mlp": 0.01011697,
      "balance_loss_clip": 1.03357041,
      "balance_loss_mlp": 1.00630915,
      "epoch": 0.08585342391631094,
      "flos": 67302739319040.0,
      "grad_norm": 0.8213084999241819,
      "language_loss": 0.55536014,
      "learning_rate": 3.967428561748671e-06,
      "loss": 0.57681483,
      "num_input_tokens_seen": 15075590,
      "step": 714,
      "time_per_iteration": 3.2335643768310547
    },
    {
      "auxiliary_loss_clip": 0.01188326,
      "auxiliary_loss_mlp": 0.01046159,
      "balance_loss_clip": 1.05715561,
      "balance_loss_mlp": 1.03285527,
      "epoch": 0.08597366680695004,
      "flos": 22456594684800.0,
      "grad_norm": 1.7662692165734508,
      "language_loss": 0.87422907,
      "learning_rate": 3.967288400710045e-06,
      "loss": 0.8965739,
      "num_input_tokens_seen": 15095055,
      "step": 715,
      "time_per_iteration": 2.6219892501831055
    },
    {
      "auxiliary_loss_clip": 0.0120543,
      "auxiliary_loss_mlp": 0.01042058,
      "balance_loss_clip": 1.06550384,
      "balance_loss_mlp": 1.02917147,
      "epoch": 0.08609390969758914,
      "flos": 23550505430400.0,
      "grad_norm": 6.4115412399782326,
      "language_loss": 0.8852272,
      "learning_rate": 3.9671479412353335e-06,
      "loss": 0.90770209,
      "num_input_tokens_seen": 15113520,
      "step": 716,
      "time_per_iteration": 2.5895464420318604
    },
    {
      "auxiliary_loss_clip": 0.01245746,
      "auxiliary_loss_mlp": 0.01046487,
      "balance_loss_clip": 1.07027149,
      "balance_loss_mlp": 1.03395176,
      "epoch": 0.08621415258822822,
      "flos": 25885848078720.0,
      "grad_norm": 2.286610213923094,
      "language_loss": 0.74080592,
      "learning_rate": 3.967007183345843e-06,
      "loss": 0.76372826,
      "num_input_tokens_seen": 15133375,
      "step": 717,
      "time_per_iteration": 2.538978338241577
    },
    {
      "auxiliary_loss_clip": 0.01238085,
      "auxiliary_loss_mlp": 0.01041016,
      "balance_loss_clip": 1.06754804,
      "balance_loss_mlp": 1.0286417,
      "epoch": 0.08633439547886732,
      "flos": 13589568120960.0,
      "grad_norm": 5.229286347001644,
      "language_loss": 0.89719927,
      "learning_rate": 3.966866127062927e-06,
      "loss": 0.91999024,
      "num_input_tokens_seen": 15150500,
      "step": 718,
      "time_per_iteration": 3.2775516510009766
    },
    {
      "auxiliary_loss_clip": 0.01138275,
      "auxiliary_loss_mlp": 0.01004857,
      "balance_loss_clip": 1.03685832,
      "balance_loss_mlp": 0.99963522,
      "epoch": 0.0864546383695064,
      "flos": 57767342434560.0,
      "grad_norm": 0.8650890060221531,
      "language_loss": 0.62711769,
      "learning_rate": 3.966724772407982e-06,
      "loss": 0.64854902,
      "num_input_tokens_seen": 15208015,
      "step": 719,
      "time_per_iteration": 3.7964155673980713
    },
    {
      "auxiliary_loss_clip": 0.01203392,
      "auxiliary_loss_mlp": 0.01040317,
      "balance_loss_clip": 1.06223512,
      "balance_loss_mlp": 1.02830648,
      "epoch": 0.0865748812601455,
      "flos": 20046952753920.0,
      "grad_norm": 2.6066224651585856,
      "language_loss": 0.88814062,
      "learning_rate": 3.966583119402454e-06,
      "loss": 0.91057777,
      "num_input_tokens_seen": 15224780,
      "step": 720,
      "time_per_iteration": 2.5559685230255127
    },
    {
      "auxiliary_loss_clip": 0.0123805,
      "auxiliary_loss_mlp": 0.00766036,
      "balance_loss_clip": 1.06715417,
      "balance_loss_mlp": 1.0003736,
      "epoch": 0.08669512415078459,
      "flos": 35262446935680.0,
      "grad_norm": 1.8037432730052723,
      "language_loss": 0.82261783,
      "learning_rate": 3.9664411680678305e-06,
      "loss": 0.8426587,
      "num_input_tokens_seen": 15246535,
      "step": 721,
      "time_per_iteration": 3.4028894901275635
    },
    {
      "auxiliary_loss_clip": 0.0111008,
      "auxiliary_loss_mlp": 0.0100618,
      "balance_loss_clip": 1.03080273,
      "balance_loss_mlp": 1.00062466,
      "epoch": 0.08681536704142367,
      "flos": 65654870048640.0,
      "grad_norm": 0.846456965627966,
      "language_loss": 0.61413199,
      "learning_rate": 3.966298918425644e-06,
      "loss": 0.63529462,
      "num_input_tokens_seen": 15304025,
      "step": 722,
      "time_per_iteration": 3.0168802738189697
    },
    {
      "auxiliary_loss_clip": 0.01242945,
      "auxiliary_loss_mlp": 0.01044157,
      "balance_loss_clip": 1.06585038,
      "balance_loss_mlp": 1.03106213,
      "epoch": 0.08693560993206277,
      "flos": 34529940881280.0,
      "grad_norm": 1.7754324483920783,
      "language_loss": 0.8268522,
      "learning_rate": 3.966156370497476e-06,
      "loss": 0.84972322,
      "num_input_tokens_seen": 15327635,
      "step": 723,
      "time_per_iteration": 2.634556293487549
    },
    {
      "auxiliary_loss_clip": 0.01244077,
      "auxiliary_loss_mlp": 0.01039826,
      "balance_loss_clip": 1.0660311,
      "balance_loss_mlp": 1.02744007,
      "epoch": 0.08705585282270185,
      "flos": 23149419189120.0,
      "grad_norm": 1.7772117672542782,
      "language_loss": 0.88431096,
      "learning_rate": 3.96601352430495e-06,
      "loss": 0.90714991,
      "num_input_tokens_seen": 15347405,
      "step": 724,
      "time_per_iteration": 2.532996654510498
    },
    {
      "auxiliary_loss_clip": 0.01226274,
      "auxiliary_loss_mlp": 0.01048428,
      "balance_loss_clip": 1.06837964,
      "balance_loss_mlp": 1.03544569,
      "epoch": 0.08717609571334095,
      "flos": 29497599498240.0,
      "grad_norm": 1.7896550083133873,
      "language_loss": 0.83266312,
      "learning_rate": 3.965870379869735e-06,
      "loss": 0.85541016,
      "num_input_tokens_seen": 15369450,
      "step": 725,
      "time_per_iteration": 2.60739803314209
    },
    {
      "auxiliary_loss_clip": 0.01238696,
      "auxiliary_loss_mlp": 0.01042638,
      "balance_loss_clip": 1.06270838,
      "balance_loss_mlp": 1.03075862,
      "epoch": 0.08729633860398003,
      "flos": 20667489137280.0,
      "grad_norm": 2.141342025028866,
      "language_loss": 0.86694145,
      "learning_rate": 3.965726937213547e-06,
      "loss": 0.88975489,
      "num_input_tokens_seen": 15388085,
      "step": 726,
      "time_per_iteration": 2.5231199264526367
    },
    {
      "auxiliary_loss_clip": 0.01239395,
      "auxiliary_loss_mlp": 0.01049424,
      "balance_loss_clip": 1.06296325,
      "balance_loss_mlp": 1.03657281,
      "epoch": 0.08741658149461913,
      "flos": 18369493655040.0,
      "grad_norm": 2.0968770109349633,
      "language_loss": 0.81280845,
      "learning_rate": 3.965583196358144e-06,
      "loss": 0.83569664,
      "num_input_tokens_seen": 15407120,
      "step": 727,
      "time_per_iteration": 2.6597909927368164
    },
    {
      "auxiliary_loss_clip": 0.01260028,
      "auxiliary_loss_mlp": 0.0104298,
      "balance_loss_clip": 1.06877446,
      "balance_loss_mlp": 1.02906823,
      "epoch": 0.08753682438525823,
      "flos": 18729677283840.0,
      "grad_norm": 2.689444276680461,
      "language_loss": 0.74456441,
      "learning_rate": 3.965439157325335e-06,
      "loss": 0.76759458,
      "num_input_tokens_seen": 15424485,
      "step": 728,
      "time_per_iteration": 2.538700819015503
    },
    {
      "auxiliary_loss_clip": 0.01218934,
      "auxiliary_loss_mlp": 0.01036936,
      "balance_loss_clip": 1.05975735,
      "balance_loss_mlp": 1.02298844,
      "epoch": 0.08765706727589731,
      "flos": 27776113303680.0,
      "grad_norm": 1.9871436051786833,
      "language_loss": 0.76060796,
      "learning_rate": 3.965294820136968e-06,
      "loss": 0.78316665,
      "num_input_tokens_seen": 15446285,
      "step": 729,
      "time_per_iteration": 2.654282331466675
    },
    {
      "auxiliary_loss_clip": 0.01229199,
      "auxiliary_loss_mlp": 0.01038602,
      "balance_loss_clip": 1.06583357,
      "balance_loss_mlp": 1.0264132,
      "epoch": 0.08777731016653641,
      "flos": 24389127239040.0,
      "grad_norm": 2.2427772571083304,
      "language_loss": 0.87208581,
      "learning_rate": 3.965150184814938e-06,
      "loss": 0.89476383,
      "num_input_tokens_seen": 15465770,
      "step": 730,
      "time_per_iteration": 2.6372649669647217
    },
    {
      "auxiliary_loss_clip": 0.01215954,
      "auxiliary_loss_mlp": 0.01045195,
      "balance_loss_clip": 1.06418157,
      "balance_loss_mlp": 1.03205848,
      "epoch": 0.08789755305717549,
      "flos": 21981855605760.0,
      "grad_norm": 3.629207783504735,
      "language_loss": 0.76280951,
      "learning_rate": 3.965005251381189e-06,
      "loss": 0.78542101,
      "num_input_tokens_seen": 15483705,
      "step": 731,
      "time_per_iteration": 2.636958360671997
    },
    {
      "auxiliary_loss_clip": 0.01133541,
      "auxiliary_loss_mlp": 0.01005898,
      "balance_loss_clip": 1.03007054,
      "balance_loss_mlp": 1.00070095,
      "epoch": 0.08801779594781459,
      "flos": 58360120583040.0,
      "grad_norm": 0.9085074219945398,
      "language_loss": 0.64629811,
      "learning_rate": 3.964860019857705e-06,
      "loss": 0.66769254,
      "num_input_tokens_seen": 15548620,
      "step": 732,
      "time_per_iteration": 3.1310997009277344
    },
    {
      "auxiliary_loss_clip": 0.01260248,
      "auxiliary_loss_mlp": 0.01045512,
      "balance_loss_clip": 1.07243061,
      "balance_loss_mlp": 1.03412759,
      "epoch": 0.08813803883845367,
      "flos": 23294785530240.0,
      "grad_norm": 1.8580667711307786,
      "language_loss": 0.84247398,
      "learning_rate": 3.964714490266518e-06,
      "loss": 0.86553162,
      "num_input_tokens_seen": 15569265,
      "step": 733,
      "time_per_iteration": 2.5205700397491455
    },
    {
      "auxiliary_loss_clip": 0.0112836,
      "auxiliary_loss_mlp": 0.01007157,
      "balance_loss_clip": 1.02875125,
      "balance_loss_mlp": 1.00191176,
      "epoch": 0.08825828172909277,
      "flos": 63424924882560.0,
      "grad_norm": 0.8963673826631723,
      "language_loss": 0.6459381,
      "learning_rate": 3.964568662629706e-06,
      "loss": 0.66729331,
      "num_input_tokens_seen": 15630570,
      "step": 734,
      "time_per_iteration": 3.007103204727173
    },
    {
      "auxiliary_loss_clip": 0.01234452,
      "auxiliary_loss_mlp": 0.01042552,
      "balance_loss_clip": 1.06225801,
      "balance_loss_mlp": 1.03089297,
      "epoch": 0.08837852461973186,
      "flos": 26720986268160.0,
      "grad_norm": 3.237547215292283,
      "language_loss": 0.84352016,
      "learning_rate": 3.9644225369693895e-06,
      "loss": 0.86629015,
      "num_input_tokens_seen": 15650870,
      "step": 735,
      "time_per_iteration": 2.5715677738189697
    },
    {
      "auxiliary_loss_clip": 0.01255842,
      "auxiliary_loss_mlp": 0.01039215,
      "balance_loss_clip": 1.06903756,
      "balance_loss_mlp": 1.02711511,
      "epoch": 0.08849876751037095,
      "flos": 27265427688960.0,
      "grad_norm": 1.9027809355522933,
      "language_loss": 0.86782962,
      "learning_rate": 3.964276113307735e-06,
      "loss": 0.89078015,
      "num_input_tokens_seen": 15670835,
      "step": 736,
      "time_per_iteration": 2.5143232345581055
    },
    {
      "auxiliary_loss_clip": 0.012099,
      "auxiliary_loss_mlp": 0.01050886,
      "balance_loss_clip": 1.06421185,
      "balance_loss_mlp": 1.0377847,
      "epoch": 0.08861901040101004,
      "flos": 19828759587840.0,
      "grad_norm": 1.8659619759796535,
      "language_loss": 0.80636764,
      "learning_rate": 3.9641293916669574e-06,
      "loss": 0.8289755,
      "num_input_tokens_seen": 15689795,
      "step": 737,
      "time_per_iteration": 2.5595498085021973
    },
    {
      "auxiliary_loss_clip": 0.01205053,
      "auxiliary_loss_mlp": 0.01037356,
      "balance_loss_clip": 1.06331253,
      "balance_loss_mlp": 1.02405179,
      "epoch": 0.08873925329164913,
      "flos": 23658704173440.0,
      "grad_norm": 1.8882247729683301,
      "language_loss": 0.82776213,
      "learning_rate": 3.9639823720693115e-06,
      "loss": 0.85018617,
      "num_input_tokens_seen": 15711650,
      "step": 738,
      "time_per_iteration": 2.6314258575439453
    },
    {
      "auxiliary_loss_clip": 0.01109011,
      "auxiliary_loss_mlp": 0.01019373,
      "balance_loss_clip": 1.04003358,
      "balance_loss_mlp": 1.01491451,
      "epoch": 0.08885949618228822,
      "flos": 71831541893760.0,
      "grad_norm": 0.8482495326162162,
      "language_loss": 0.59987473,
      "learning_rate": 3.963835054537102e-06,
      "loss": 0.62115854,
      "num_input_tokens_seen": 15780615,
      "step": 739,
      "time_per_iteration": 3.193509817123413
    },
    {
      "auxiliary_loss_clip": 0.01220775,
      "auxiliary_loss_mlp": 0.01055811,
      "balance_loss_clip": 1.06141114,
      "balance_loss_mlp": 1.0437355,
      "epoch": 0.08897973907292732,
      "flos": 22346169298560.0,
      "grad_norm": 2.2139264102670073,
      "language_loss": 0.61381447,
      "learning_rate": 3.963687439092676e-06,
      "loss": 0.63658041,
      "num_input_tokens_seen": 15801300,
      "step": 740,
      "time_per_iteration": 3.3782408237457275
    },
    {
      "auxiliary_loss_clip": 0.01240986,
      "auxiliary_loss_mlp": 0.01046714,
      "balance_loss_clip": 1.06674385,
      "balance_loss_mlp": 1.03450692,
      "epoch": 0.0890999819635664,
      "flos": 21251827589760.0,
      "grad_norm": 3.9173797294777475,
      "language_loss": 0.80378187,
      "learning_rate": 3.963539525758427e-06,
      "loss": 0.8266589,
      "num_input_tokens_seen": 15820860,
      "step": 741,
      "time_per_iteration": 2.5305728912353516
    },
    {
      "auxiliary_loss_clip": 0.01226877,
      "auxiliary_loss_mlp": 0.01039917,
      "balance_loss_clip": 1.0660212,
      "balance_loss_mlp": 1.0269649,
      "epoch": 0.0892202248542055,
      "flos": 25370888745600.0,
      "grad_norm": 2.430622964395486,
      "language_loss": 0.67870855,
      "learning_rate": 3.9633913145567925e-06,
      "loss": 0.7013765,
      "num_input_tokens_seen": 15841350,
      "step": 742,
      "time_per_iteration": 2.564605951309204
    },
    {
      "auxiliary_loss_clip": 0.01225065,
      "auxiliary_loss_mlp": 0.01034582,
      "balance_loss_clip": 1.06720603,
      "balance_loss_mlp": 1.02282858,
      "epoch": 0.08934046774484458,
      "flos": 24457895827200.0,
      "grad_norm": 2.2087618900411017,
      "language_loss": 0.81364214,
      "learning_rate": 3.9632428055102575e-06,
      "loss": 0.83623862,
      "num_input_tokens_seen": 15861360,
      "step": 743,
      "time_per_iteration": 2.5781798362731934
    },
    {
      "auxiliary_loss_clip": 0.01246759,
      "auxiliary_loss_mlp": 0.01046256,
      "balance_loss_clip": 1.07117701,
      "balance_loss_mlp": 1.03242755,
      "epoch": 0.08946071063548368,
      "flos": 35772773414400.0,
      "grad_norm": 2.4649839207881596,
      "language_loss": 0.66916597,
      "learning_rate": 3.9630939986413495e-06,
      "loss": 0.69209611,
      "num_input_tokens_seen": 15883160,
      "step": 744,
      "time_per_iteration": 2.6227333545684814
    },
    {
      "auxiliary_loss_clip": 0.01197922,
      "auxiliary_loss_mlp": 0.01049724,
      "balance_loss_clip": 1.06317616,
      "balance_loss_mlp": 1.0368731,
      "epoch": 0.08958095352612276,
      "flos": 14356584167040.0,
      "grad_norm": 1.8499385600165459,
      "language_loss": 0.78343606,
      "learning_rate": 3.962944893972643e-06,
      "loss": 0.80591255,
      "num_input_tokens_seen": 15901610,
      "step": 745,
      "time_per_iteration": 3.3605949878692627
    },
    {
      "auxiliary_loss_clip": 0.01223982,
      "auxiliary_loss_mlp": 0.01044765,
      "balance_loss_clip": 1.06447363,
      "balance_loss_mlp": 1.03206921,
      "epoch": 0.08970119641676186,
      "flos": 17853277345920.0,
      "grad_norm": 3.0551993831848607,
      "language_loss": 0.91105425,
      "learning_rate": 3.962795491526756e-06,
      "loss": 0.93374175,
      "num_input_tokens_seen": 15918770,
      "step": 746,
      "time_per_iteration": 3.4027154445648193
    },
    {
      "auxiliary_loss_clip": 0.01264162,
      "auxiliary_loss_mlp": 0.01056027,
      "balance_loss_clip": 1.07289803,
      "balance_loss_mlp": 1.04190671,
      "epoch": 0.08982143930740095,
      "flos": 20811670329600.0,
      "grad_norm": 2.1750555823974267,
      "language_loss": 0.89316124,
      "learning_rate": 3.962645791326354e-06,
      "loss": 0.91636312,
      "num_input_tokens_seen": 15938025,
      "step": 747,
      "time_per_iteration": 2.490537405014038
    },
    {
      "auxiliary_loss_clip": 0.01239971,
      "auxiliary_loss_mlp": 0.01041097,
      "balance_loss_clip": 1.06860781,
      "balance_loss_mlp": 1.02948022,
      "epoch": 0.08994168219804004,
      "flos": 24097712198400.0,
      "grad_norm": 1.9740534642002414,
      "language_loss": 0.82780039,
      "learning_rate": 3.962495793394146e-06,
      "loss": 0.85061109,
      "num_input_tokens_seen": 15957215,
      "step": 748,
      "time_per_iteration": 3.3045437335968018
    },
    {
      "auxiliary_loss_clip": 0.01143532,
      "auxiliary_loss_mlp": 0.01040568,
      "balance_loss_clip": 1.03070569,
      "balance_loss_mlp": 1.03584719,
      "epoch": 0.09006192508867913,
      "flos": 57188893812480.0,
      "grad_norm": 0.7418661616341179,
      "language_loss": 0.61205,
      "learning_rate": 3.9623454977528864e-06,
      "loss": 0.63389093,
      "num_input_tokens_seen": 16015870,
      "step": 749,
      "time_per_iteration": 2.89220929145813
    },
    {
      "auxiliary_loss_clip": 0.01214954,
      "auxiliary_loss_mlp": 0.01051043,
      "balance_loss_clip": 1.06468487,
      "balance_loss_mlp": 1.03827572,
      "epoch": 0.09018216797931822,
      "flos": 20487505063680.0,
      "grad_norm": 1.638919864447389,
      "language_loss": 0.85080326,
      "learning_rate": 3.962194904425375e-06,
      "loss": 0.87346327,
      "num_input_tokens_seen": 16036500,
      "step": 750,
      "time_per_iteration": 2.6000778675079346
    },
    {
      "auxiliary_loss_clip": 0.01236507,
      "auxiliary_loss_mlp": 0.01042042,
      "balance_loss_clip": 1.06613481,
      "balance_loss_mlp": 1.029459,
      "epoch": 0.09030241086995731,
      "flos": 22638123043200.0,
      "grad_norm": 3.524315116278463,
      "language_loss": 0.6796189,
      "learning_rate": 3.9620440134344566e-06,
      "loss": 0.70240438,
      "num_input_tokens_seen": 16054655,
      "step": 751,
      "time_per_iteration": 2.5078063011169434
    },
    {
      "auxiliary_loss_clip": 0.01207956,
      "auxiliary_loss_mlp": 0.01050457,
      "balance_loss_clip": 1.0646832,
      "balance_loss_mlp": 1.03704023,
      "epoch": 0.09042265376059641,
      "flos": 21871502046720.0,
      "grad_norm": 2.90199615579018,
      "language_loss": 0.82166553,
      "learning_rate": 3.9618928248030215e-06,
      "loss": 0.84424967,
      "num_input_tokens_seen": 16074165,
      "step": 752,
      "time_per_iteration": 2.5974485874176025
    },
    {
      "auxiliary_loss_clip": 0.01239493,
      "auxiliary_loss_mlp": 0.01049187,
      "balance_loss_clip": 1.06807065,
      "balance_loss_mlp": 1.03687274,
      "epoch": 0.0905428966512355,
      "flos": 24316192673280.0,
      "grad_norm": 2.1489688556392674,
      "language_loss": 0.83093143,
      "learning_rate": 3.961741338554005e-06,
      "loss": 0.85381818,
      "num_input_tokens_seen": 16092505,
      "step": 753,
      "time_per_iteration": 2.5624661445617676
    },
    {
      "auxiliary_loss_clip": 0.01233045,
      "auxiliary_loss_mlp": 0.01055188,
      "balance_loss_clip": 1.06765342,
      "balance_loss_mlp": 1.04195607,
      "epoch": 0.09066313954187459,
      "flos": 35845061535360.0,
      "grad_norm": 2.1602943930785923,
      "language_loss": 0.75770986,
      "learning_rate": 3.9615895547103865e-06,
      "loss": 0.7805922,
      "num_input_tokens_seen": 16116150,
      "step": 754,
      "time_per_iteration": 2.742488384246826
    },
    {
      "auxiliary_loss_clip": 0.01223138,
      "auxiliary_loss_mlp": 0.01050512,
      "balance_loss_clip": 1.06320918,
      "balance_loss_mlp": 1.03761959,
      "epoch": 0.09078338243251367,
      "flos": 29168729550720.0,
      "grad_norm": 1.839582196697024,
      "language_loss": 0.77699804,
      "learning_rate": 3.961437473295193e-06,
      "loss": 0.79973447,
      "num_input_tokens_seen": 16136295,
      "step": 755,
      "time_per_iteration": 2.5963313579559326
    },
    {
      "auxiliary_loss_clip": 0.01178461,
      "auxiliary_loss_mlp": 0.01046698,
      "balance_loss_clip": 1.05443311,
      "balance_loss_mlp": 1.03419936,
      "epoch": 0.09090362532315277,
      "flos": 21907699977600.0,
      "grad_norm": 2.41548011638984,
      "language_loss": 0.72252071,
      "learning_rate": 3.961285094331495e-06,
      "loss": 0.74477232,
      "num_input_tokens_seen": 16154210,
      "step": 756,
      "time_per_iteration": 2.641754388809204
    },
    {
      "auxiliary_loss_clip": 0.01251217,
      "auxiliary_loss_mlp": 0.01041559,
      "balance_loss_clip": 1.06597853,
      "balance_loss_mlp": 1.029948,
      "epoch": 0.09102386821379185,
      "flos": 27344503480320.0,
      "grad_norm": 4.562012824283252,
      "language_loss": 0.86085641,
      "learning_rate": 3.961132417842406e-06,
      "loss": 0.88378417,
      "num_input_tokens_seen": 16173995,
      "step": 757,
      "time_per_iteration": 2.521122455596924
    },
    {
      "auxiliary_loss_clip": 0.01232284,
      "auxiliary_loss_mlp": 0.01050963,
      "balance_loss_clip": 1.06513047,
      "balance_loss_mlp": 1.03897023,
      "epoch": 0.09114411110443095,
      "flos": 20813501923200.0,
      "grad_norm": 3.489379949445828,
      "language_loss": 0.75569755,
      "learning_rate": 3.960979443851089e-06,
      "loss": 0.77853,
      "num_input_tokens_seen": 16191020,
      "step": 758,
      "time_per_iteration": 2.5540335178375244
    },
    {
      "auxiliary_loss_clip": 0.01222789,
      "auxiliary_loss_mlp": 0.01042283,
      "balance_loss_clip": 1.06476259,
      "balance_loss_mlp": 1.02890193,
      "epoch": 0.09126435399507005,
      "flos": 26145949438080.0,
      "grad_norm": 1.712151297559235,
      "language_loss": 0.7875455,
      "learning_rate": 3.96082617238075e-06,
      "loss": 0.81019622,
      "num_input_tokens_seen": 16213645,
      "step": 759,
      "time_per_iteration": 2.618605136871338
    },
    {
      "auxiliary_loss_clip": 0.0122199,
      "auxiliary_loss_mlp": 0.01038585,
      "balance_loss_clip": 1.06349874,
      "balance_loss_mlp": 1.02709365,
      "epoch": 0.09138459688570913,
      "flos": 24388911757440.0,
      "grad_norm": 2.581203286872189,
      "language_loss": 0.79955423,
      "learning_rate": 3.960672603454639e-06,
      "loss": 0.82215995,
      "num_input_tokens_seen": 16233625,
      "step": 760,
      "time_per_iteration": 2.5661139488220215
    },
    {
      "auxiliary_loss_clip": 0.01234321,
      "auxiliary_loss_mlp": 0.01047414,
      "balance_loss_clip": 1.06649208,
      "balance_loss_mlp": 1.03436661,
      "epoch": 0.09150483977634823,
      "flos": 21032664756480.0,
      "grad_norm": 2.908520015453707,
      "language_loss": 0.77360773,
      "learning_rate": 3.960518737096054e-06,
      "loss": 0.7964251,
      "num_input_tokens_seen": 16253255,
      "step": 761,
      "time_per_iteration": 2.5108797550201416
    },
    {
      "auxiliary_loss_clip": 0.01239668,
      "auxiliary_loss_mlp": 0.01039487,
      "balance_loss_clip": 1.06747317,
      "balance_loss_mlp": 1.02695203,
      "epoch": 0.09162508266698731,
      "flos": 22856998567680.0,
      "grad_norm": 2.1656183213501445,
      "language_loss": 0.72762758,
      "learning_rate": 3.960364573328334e-06,
      "loss": 0.75041914,
      "num_input_tokens_seen": 16272580,
      "step": 762,
      "time_per_iteration": 2.537421941757202
    },
    {
      "auxiliary_loss_clip": 0.01208382,
      "auxiliary_loss_mlp": 0.01040424,
      "balance_loss_clip": 1.06091428,
      "balance_loss_mlp": 1.02698278,
      "epoch": 0.0917453255576264,
      "flos": 21724411852800.0,
      "grad_norm": 1.9035893829647503,
      "language_loss": 0.8847158,
      "learning_rate": 3.9602101121748675e-06,
      "loss": 0.90720385,
      "num_input_tokens_seen": 16293075,
      "step": 763,
      "time_per_iteration": 2.5745084285736084
    },
    {
      "auxiliary_loss_clip": 0.01223701,
      "auxiliary_loss_mlp": 0.01044249,
      "balance_loss_clip": 1.06795657,
      "balance_loss_mlp": 1.03285289,
      "epoch": 0.0918655684482655,
      "flos": 14609215497600.0,
      "grad_norm": 3.0841465319861605,
      "language_loss": 0.72333789,
      "learning_rate": 3.960055353659085e-06,
      "loss": 0.7460174,
      "num_input_tokens_seen": 16310185,
      "step": 764,
      "time_per_iteration": 2.49641489982605
    },
    {
      "auxiliary_loss_clip": 0.01211061,
      "auxiliary_loss_mlp": 0.01035208,
      "balance_loss_clip": 1.06531608,
      "balance_loss_mlp": 1.02291179,
      "epoch": 0.09198581133890459,
      "flos": 23435016226560.0,
      "grad_norm": 1.8135206928884973,
      "language_loss": 0.83774364,
      "learning_rate": 3.959900297804465e-06,
      "loss": 0.86020637,
      "num_input_tokens_seen": 16330355,
      "step": 765,
      "time_per_iteration": 2.6304118633270264
    },
    {
      "auxiliary_loss_clip": 0.01210006,
      "auxiliary_loss_mlp": 0.01039267,
      "balance_loss_clip": 1.06236756,
      "balance_loss_mlp": 1.02711332,
      "epoch": 0.09210605422954368,
      "flos": 16795887753600.0,
      "grad_norm": 1.7923047136102985,
      "language_loss": 0.77408326,
      "learning_rate": 3.9597449446345276e-06,
      "loss": 0.79657596,
      "num_input_tokens_seen": 16347600,
      "step": 766,
      "time_per_iteration": 2.6017959117889404
    },
    {
      "auxiliary_loss_clip": 0.01209076,
      "auxiliary_loss_mlp": 0.01037431,
      "balance_loss_clip": 1.05900049,
      "balance_loss_mlp": 1.02579629,
      "epoch": 0.09222629712018277,
      "flos": 22674249146880.0,
      "grad_norm": 2.241517571459111,
      "language_loss": 0.8341403,
      "learning_rate": 3.95958929417284e-06,
      "loss": 0.85660541,
      "num_input_tokens_seen": 16365755,
      "step": 767,
      "time_per_iteration": 3.4614627361297607
    },
    {
      "auxiliary_loss_clip": 0.01131984,
      "auxiliary_loss_mlp": 0.0100879,
      "balance_loss_clip": 1.0303998,
      "balance_loss_mlp": 1.00435531,
      "epoch": 0.09234654001082186,
      "flos": 69976756327680.0,
      "grad_norm": 0.7498201645231182,
      "language_loss": 0.58803904,
      "learning_rate": 3.9594333464430145e-06,
      "loss": 0.60944676,
      "num_input_tokens_seen": 16435245,
      "step": 768,
      "time_per_iteration": 3.232046365737915
    },
    {
      "auxiliary_loss_clip": 0.0114631,
      "auxiliary_loss_mlp": 0.0104232,
      "balance_loss_clip": 1.05068898,
      "balance_loss_mlp": 1.03140604,
      "epoch": 0.09246678290146094,
      "flos": 20011437181440.0,
      "grad_norm": 1.877993792484274,
      "language_loss": 0.88569391,
      "learning_rate": 3.959277101468709e-06,
      "loss": 0.90758014,
      "num_input_tokens_seen": 16454795,
      "step": 769,
      "time_per_iteration": 2.844501495361328
    },
    {
      "auxiliary_loss_clip": 0.01205583,
      "auxiliary_loss_mlp": 0.01048922,
      "balance_loss_clip": 1.06103909,
      "balance_loss_mlp": 1.03714979,
      "epoch": 0.09258702579210004,
      "flos": 17747448900480.0,
      "grad_norm": 2.381511027482938,
      "language_loss": 0.78577614,
      "learning_rate": 3.959120559273624e-06,
      "loss": 0.80832124,
      "num_input_tokens_seen": 16472580,
      "step": 770,
      "time_per_iteration": 2.8863327503204346
    },
    {
      "auxiliary_loss_clip": 0.01205286,
      "auxiliary_loss_mlp": 0.01042845,
      "balance_loss_clip": 1.0624392,
      "balance_loss_mlp": 1.0311209,
      "epoch": 0.09270726868273914,
      "flos": 20886544229760.0,
      "grad_norm": 1.7712475182230991,
      "language_loss": 0.83321321,
      "learning_rate": 3.958963719881509e-06,
      "loss": 0.85569459,
      "num_input_tokens_seen": 16490670,
      "step": 771,
      "time_per_iteration": 3.3953518867492676
    },
    {
      "auxiliary_loss_clip": 0.01238179,
      "auxiliary_loss_mlp": 0.01036271,
      "balance_loss_clip": 1.06904769,
      "balance_loss_mlp": 1.02334261,
      "epoch": 0.09282751157337822,
      "flos": 17015697031680.0,
      "grad_norm": 2.4160427784363647,
      "language_loss": 0.93895793,
      "learning_rate": 3.958806583316154e-06,
      "loss": 0.96170247,
      "num_input_tokens_seen": 16508640,
      "step": 772,
      "time_per_iteration": 2.5044260025024414
    },
    {
      "auxiliary_loss_clip": 0.01253304,
      "auxiliary_loss_mlp": 0.01032895,
      "balance_loss_clip": 1.07007515,
      "balance_loss_mlp": 1.02168894,
      "epoch": 0.09294775446401732,
      "flos": 32523647748480.0,
      "grad_norm": 2.2935244338408043,
      "language_loss": 0.78857535,
      "learning_rate": 3.9586491496013985e-06,
      "loss": 0.81143737,
      "num_input_tokens_seen": 16531035,
      "step": 773,
      "time_per_iteration": 3.506129503250122
    },
    {
      "auxiliary_loss_clip": 0.01242441,
      "auxiliary_loss_mlp": 0.01047174,
      "balance_loss_clip": 1.06782627,
      "balance_loss_mlp": 1.03559327,
      "epoch": 0.0930679973546564,
      "flos": 18259750627200.0,
      "grad_norm": 2.443343993646761,
      "language_loss": 0.82959175,
      "learning_rate": 3.958491418761124e-06,
      "loss": 0.85248786,
      "num_input_tokens_seen": 16548605,
      "step": 774,
      "time_per_iteration": 2.484351873397827
    },
    {
      "auxiliary_loss_clip": 0.01220364,
      "auxiliary_loss_mlp": 0.01039879,
      "balance_loss_clip": 1.060323,
      "balance_loss_mlp": 1.02854264,
      "epoch": 0.0931882402452955,
      "flos": 21099745405440.0,
      "grad_norm": 3.511223698697841,
      "language_loss": 0.72452903,
      "learning_rate": 3.958333390819258e-06,
      "loss": 0.74713147,
      "num_input_tokens_seen": 16565535,
      "step": 775,
      "time_per_iteration": 3.320323944091797
    },
    {
      "auxiliary_loss_clip": 0.01252348,
      "auxiliary_loss_mlp": 0.01040878,
      "balance_loss_clip": 1.06885231,
      "balance_loss_mlp": 1.02976751,
      "epoch": 0.0933084831359346,
      "flos": 24207275658240.0,
      "grad_norm": 2.162574745285735,
      "language_loss": 0.80319798,
      "learning_rate": 3.9581750657997754e-06,
      "loss": 0.82613021,
      "num_input_tokens_seen": 16584900,
      "step": 776,
      "time_per_iteration": 2.5163486003875732
    },
    {
      "auxiliary_loss_clip": 0.01218066,
      "auxiliary_loss_mlp": 0.0103836,
      "balance_loss_clip": 1.06113338,
      "balance_loss_mlp": 1.02718973,
      "epoch": 0.09342872602657368,
      "flos": 25480272637440.0,
      "grad_norm": 1.8029756803353811,
      "language_loss": 0.89400685,
      "learning_rate": 3.95801644372669e-06,
      "loss": 0.91657108,
      "num_input_tokens_seen": 16604805,
      "step": 777,
      "time_per_iteration": 2.582204818725586
    },
    {
      "auxiliary_loss_clip": 0.01227318,
      "auxiliary_loss_mlp": 0.01041934,
      "balance_loss_clip": 1.06274056,
      "balance_loss_mlp": 1.03037691,
      "epoch": 0.09354896891721277,
      "flos": 23149060053120.0,
      "grad_norm": 2.7455108891514133,
      "language_loss": 0.84638035,
      "learning_rate": 3.957857524624068e-06,
      "loss": 0.86907291,
      "num_input_tokens_seen": 16623685,
      "step": 778,
      "time_per_iteration": 2.5764808654785156
    },
    {
      "auxiliary_loss_clip": 0.01220554,
      "auxiliary_loss_mlp": 0.01041214,
      "balance_loss_clip": 1.06451082,
      "balance_loss_mlp": 1.02994251,
      "epoch": 0.09366921180785186,
      "flos": 24279563779200.0,
      "grad_norm": 1.7713504253024963,
      "language_loss": 0.89744759,
      "learning_rate": 3.957698308516016e-06,
      "loss": 0.92006528,
      "num_input_tokens_seen": 16644985,
      "step": 779,
      "time_per_iteration": 2.5860352516174316
    },
    {
      "auxiliary_loss_clip": 0.01232785,
      "auxiliary_loss_mlp": 0.00765174,
      "balance_loss_clip": 1.06731415,
      "balance_loss_mlp": 1.00008166,
      "epoch": 0.09378945469849095,
      "flos": 18730036419840.0,
      "grad_norm": 1.9528626408057206,
      "language_loss": 0.82340336,
      "learning_rate": 3.957538795426688e-06,
      "loss": 0.84338295,
      "num_input_tokens_seen": 16662410,
      "step": 780,
      "time_per_iteration": 2.4987144470214844
    },
    {
      "auxiliary_loss_clip": 0.01222187,
      "auxiliary_loss_mlp": 0.01045749,
      "balance_loss_clip": 1.06392038,
      "balance_loss_mlp": 1.0328387,
      "epoch": 0.09390969758913004,
      "flos": 23218834222080.0,
      "grad_norm": 2.775819738756101,
      "language_loss": 0.7731142,
      "learning_rate": 3.9573789853802804e-06,
      "loss": 0.79579353,
      "num_input_tokens_seen": 16680885,
      "step": 781,
      "time_per_iteration": 2.5398433208465576
    },
    {
      "auxiliary_loss_clip": 0.01222441,
      "auxiliary_loss_mlp": 0.00764928,
      "balance_loss_clip": 1.0673008,
      "balance_loss_mlp": 1.00009966,
      "epoch": 0.09402994047976913,
      "flos": 19646728439040.0,
      "grad_norm": 2.2231868911922863,
      "language_loss": 0.75059783,
      "learning_rate": 3.957218878401037e-06,
      "loss": 0.77047157,
      "num_input_tokens_seen": 16699375,
      "step": 782,
      "time_per_iteration": 2.52759051322937
    },
    {
      "auxiliary_loss_clip": 0.01253558,
      "auxiliary_loss_mlp": 0.01044325,
      "balance_loss_clip": 1.06937087,
      "balance_loss_mlp": 1.03211188,
      "epoch": 0.09415018337040823,
      "flos": 29420463041280.0,
      "grad_norm": 2.1228779026288866,
      "language_loss": 0.89221221,
      "learning_rate": 3.957058474513246e-06,
      "loss": 0.91519105,
      "num_input_tokens_seen": 16719230,
      "step": 783,
      "time_per_iteration": 2.536959409713745
    },
    {
      "auxiliary_loss_clip": 0.0123381,
      "auxiliary_loss_mlp": 0.01047226,
      "balance_loss_clip": 1.06738698,
      "balance_loss_mlp": 1.03640795,
      "epoch": 0.09427042626104731,
      "flos": 24572092141440.0,
      "grad_norm": 2.098397684064178,
      "language_loss": 0.78701937,
      "learning_rate": 3.956897773741241e-06,
      "loss": 0.80982971,
      "num_input_tokens_seen": 16738220,
      "step": 784,
      "time_per_iteration": 2.520097494125366
    },
    {
      "auxiliary_loss_clip": 0.01211045,
      "auxiliary_loss_mlp": 0.01045078,
      "balance_loss_clip": 1.06269038,
      "balance_loss_mlp": 1.03306139,
      "epoch": 0.09439066915168641,
      "flos": 26359581576960.0,
      "grad_norm": 1.803975054384304,
      "language_loss": 0.71477062,
      "learning_rate": 3.956736776109398e-06,
      "loss": 0.73733187,
      "num_input_tokens_seen": 16759395,
      "step": 785,
      "time_per_iteration": 2.5932247638702393
    },
    {
      "auxiliary_loss_clip": 0.01227363,
      "auxiliary_loss_mlp": 0.00765931,
      "balance_loss_clip": 1.06248522,
      "balance_loss_mlp": 1.00014472,
      "epoch": 0.09451091204232549,
      "flos": 19427278296960.0,
      "grad_norm": 1.96090050861924,
      "language_loss": 0.83667952,
      "learning_rate": 3.956575481642143e-06,
      "loss": 0.85661244,
      "num_input_tokens_seen": 16778285,
      "step": 786,
      "time_per_iteration": 2.526232957839966
    },
    {
      "auxiliary_loss_clip": 0.01180531,
      "auxiliary_loss_mlp": 0.01035462,
      "balance_loss_clip": 1.05511522,
      "balance_loss_mlp": 1.02349329,
      "epoch": 0.09463115493296459,
      "flos": 25368051571200.0,
      "grad_norm": 2.9720776375749427,
      "language_loss": 0.74837685,
      "learning_rate": 3.956413890363943e-06,
      "loss": 0.77053678,
      "num_input_tokens_seen": 16795265,
      "step": 787,
      "time_per_iteration": 2.658855676651001
    },
    {
      "auxiliary_loss_clip": 0.01234635,
      "auxiliary_loss_mlp": 0.01039224,
      "balance_loss_clip": 1.06727445,
      "balance_loss_mlp": 1.02813196,
      "epoch": 0.09475139782360369,
      "flos": 10123254869760.0,
      "grad_norm": 1.9638130848844293,
      "language_loss": 0.81917673,
      "learning_rate": 3.956252002299312e-06,
      "loss": 0.84191531,
      "num_input_tokens_seen": 16811165,
      "step": 788,
      "time_per_iteration": 2.5442371368408203
    },
    {
      "auxiliary_loss_clip": 0.01251322,
      "auxiliary_loss_mlp": 0.01034021,
      "balance_loss_clip": 1.06852293,
      "balance_loss_mlp": 1.02258253,
      "epoch": 0.09487164071424277,
      "flos": 17231088936960.0,
      "grad_norm": 1.8412229805445395,
      "language_loss": 0.90676838,
      "learning_rate": 3.956089817472807e-06,
      "loss": 0.92962176,
      "num_input_tokens_seen": 16828470,
      "step": 789,
      "time_per_iteration": 2.519622802734375
    },
    {
      "auxiliary_loss_clip": 0.01220214,
      "auxiliary_loss_mlp": 0.01038964,
      "balance_loss_clip": 1.06710017,
      "balance_loss_mlp": 1.02788401,
      "epoch": 0.09499188360488187,
      "flos": 30849564528000.0,
      "grad_norm": 2.086778008739669,
      "language_loss": 0.85708684,
      "learning_rate": 3.955927335909032e-06,
      "loss": 0.87967861,
      "num_input_tokens_seen": 16851680,
      "step": 790,
      "time_per_iteration": 2.6684296131134033
    },
    {
      "auxiliary_loss_clip": 0.01188987,
      "auxiliary_loss_mlp": 0.01040374,
      "balance_loss_clip": 1.06550288,
      "balance_loss_mlp": 1.02929318,
      "epoch": 0.09511212649552095,
      "flos": 29351694453120.0,
      "grad_norm": 2.037735741394394,
      "language_loss": 0.75897175,
      "learning_rate": 3.955764557632634e-06,
      "loss": 0.78126538,
      "num_input_tokens_seen": 16871490,
      "step": 791,
      "time_per_iteration": 2.6928322315216064
    },
    {
      "auxiliary_loss_clip": 0.01215442,
      "auxiliary_loss_mlp": 0.01038112,
      "balance_loss_clip": 1.06282735,
      "balance_loss_mlp": 1.02656031,
      "epoch": 0.09523236938616005,
      "flos": 10378687461120.0,
      "grad_norm": 2.083291490242172,
      "language_loss": 0.94643569,
      "learning_rate": 3.955601482668309e-06,
      "loss": 0.96897125,
      "num_input_tokens_seen": 16889350,
      "step": 792,
      "time_per_iteration": 2.5200576782226562
    },
    {
      "auxiliary_loss_clip": 0.01183547,
      "auxiliary_loss_mlp": 0.01040419,
      "balance_loss_clip": 1.05602777,
      "balance_loss_mlp": 1.02699637,
      "epoch": 0.09535261227679913,
      "flos": 19061815368960.0,
      "grad_norm": 1.7384333115757473,
      "language_loss": 0.88378692,
      "learning_rate": 3.955438111040794e-06,
      "loss": 0.9060266,
      "num_input_tokens_seen": 16907625,
      "step": 793,
      "time_per_iteration": 2.6130173206329346
    },
    {
      "auxiliary_loss_clip": 0.01183201,
      "auxiliary_loss_mlp": 0.0104513,
      "balance_loss_clip": 1.05778408,
      "balance_loss_mlp": 1.034199,
      "epoch": 0.09547285516743823,
      "flos": 20922993555840.0,
      "grad_norm": 2.007447161959936,
      "language_loss": 0.79972219,
      "learning_rate": 3.955274442774873e-06,
      "loss": 0.82200551,
      "num_input_tokens_seen": 16926205,
      "step": 794,
      "time_per_iteration": 3.4259731769561768
    },
    {
      "auxiliary_loss_clip": 0.01234252,
      "auxiliary_loss_mlp": 0.01044026,
      "balance_loss_clip": 1.06554341,
      "balance_loss_mlp": 1.03181863,
      "epoch": 0.09559309805807732,
      "flos": 30154405639680.0,
      "grad_norm": 2.8176491728850506,
      "language_loss": 0.70629764,
      "learning_rate": 3.9551104778953725e-06,
      "loss": 0.72908038,
      "num_input_tokens_seen": 16946500,
      "step": 795,
      "time_per_iteration": 2.583831310272217
    },
    {
      "auxiliary_loss_clip": 0.01203554,
      "auxiliary_loss_mlp": 0.010329,
      "balance_loss_clip": 1.06006777,
      "balance_loss_mlp": 1.02169394,
      "epoch": 0.0957133409487164,
      "flos": 21066743784960.0,
      "grad_norm": 1.8080487606794478,
      "language_loss": 0.85488039,
      "learning_rate": 3.954946216427167e-06,
      "loss": 0.87724489,
      "num_input_tokens_seen": 16966960,
      "step": 796,
      "time_per_iteration": 2.593061923980713
    },
    {
      "auxiliary_loss_clip": 0.01102803,
      "auxiliary_loss_mlp": 0.01008399,
      "balance_loss_clip": 1.0285213,
      "balance_loss_mlp": 1.0036782,
      "epoch": 0.0958335838393555,
      "flos": 71297979315840.0,
      "grad_norm": 0.9598719269203762,
      "language_loss": 0.61594796,
      "learning_rate": 3.954781658395176e-06,
      "loss": 0.63705993,
      "num_input_tokens_seen": 17023215,
      "step": 797,
      "time_per_iteration": 3.097337245941162
    },
    {
      "auxiliary_loss_clip": 0.01224686,
      "auxiliary_loss_mlp": 0.01038883,
      "balance_loss_clip": 1.06469059,
      "balance_loss_mlp": 1.02677083,
      "epoch": 0.09595382672999458,
      "flos": 21872974504320.0,
      "grad_norm": 3.0385188852364826,
      "language_loss": 0.92234981,
      "learning_rate": 3.95461680382436e-06,
      "loss": 0.94498551,
      "num_input_tokens_seen": 17042140,
      "step": 798,
      "time_per_iteration": 3.3518388271331787
    },
    {
      "auxiliary_loss_clip": 0.01240478,
      "auxiliary_loss_mlp": 0.01042074,
      "balance_loss_clip": 1.06897712,
      "balance_loss_mlp": 1.02977157,
      "epoch": 0.09607406962063368,
      "flos": 18695562341760.0,
      "grad_norm": 3.224595268756587,
      "language_loss": 0.86197603,
      "learning_rate": 3.9544516527397295e-06,
      "loss": 0.88480151,
      "num_input_tokens_seen": 17058490,
      "step": 799,
      "time_per_iteration": 2.55964994430542
    },
    {
      "auxiliary_loss_clip": 0.01204065,
      "auxiliary_loss_mlp": 0.01035895,
      "balance_loss_clip": 1.06159639,
      "balance_loss_mlp": 1.02425396,
      "epoch": 0.09619431251127276,
      "flos": 22568456615040.0,
      "grad_norm": 2.21245382892145,
      "language_loss": 0.80572426,
      "learning_rate": 3.954286205166338e-06,
      "loss": 0.82812387,
      "num_input_tokens_seen": 17079655,
      "step": 800,
      "time_per_iteration": 3.525792360305786
    },
    {
      "auxiliary_loss_clip": 0.012428,
      "auxiliary_loss_mlp": 0.01043224,
      "balance_loss_clip": 1.07241571,
      "balance_loss_mlp": 1.03049839,
      "epoch": 0.09631455540191186,
      "flos": 14246230608000.0,
      "grad_norm": 8.926001993125734,
      "language_loss": 0.83849692,
      "learning_rate": 3.954120461129282e-06,
      "loss": 0.86135709,
      "num_input_tokens_seen": 17097065,
      "step": 801,
      "time_per_iteration": 3.286698818206787
    },
    {
      "auxiliary_loss_clip": 0.01256027,
      "auxiliary_loss_mlp": 0.010466,
      "balance_loss_clip": 1.07260573,
      "balance_loss_mlp": 1.03525114,
      "epoch": 0.09643479829255096,
      "flos": 20740387789440.0,
      "grad_norm": 2.315680098338758,
      "language_loss": 0.84151202,
      "learning_rate": 3.953954420653706e-06,
      "loss": 0.86453825,
      "num_input_tokens_seen": 17114090,
      "step": 802,
      "time_per_iteration": 2.5248186588287354
    },
    {
      "auxiliary_loss_clip": 0.01232975,
      "auxiliary_loss_mlp": 0.01043224,
      "balance_loss_clip": 1.06682491,
      "balance_loss_mlp": 1.03179741,
      "epoch": 0.09655504118319004,
      "flos": 24420476833920.0,
      "grad_norm": 2.0181071838890965,
      "language_loss": 0.88420212,
      "learning_rate": 3.953788083764798e-06,
      "loss": 0.90696412,
      "num_input_tokens_seen": 17133325,
      "step": 803,
      "time_per_iteration": 2.5352556705474854
    },
    {
      "auxiliary_loss_clip": 0.01189739,
      "auxiliary_loss_mlp": 0.01048035,
      "balance_loss_clip": 1.06242228,
      "balance_loss_mlp": 1.0364002,
      "epoch": 0.09667528407382914,
      "flos": 18441961344000.0,
      "grad_norm": 2.496995321964076,
      "language_loss": 0.92321628,
      "learning_rate": 3.953621450487792e-06,
      "loss": 0.94559401,
      "num_input_tokens_seen": 17151945,
      "step": 804,
      "time_per_iteration": 2.595557928085327
    },
    {
      "auxiliary_loss_clip": 0.01142567,
      "auxiliary_loss_mlp": 0.0101372,
      "balance_loss_clip": 1.03245211,
      "balance_loss_mlp": 1.00902319,
      "epoch": 0.09679552696446822,
      "flos": 70816455544320.0,
      "grad_norm": 0.8332030199490637,
      "language_loss": 0.61185551,
      "learning_rate": 3.953454520847964e-06,
      "loss": 0.63341832,
      "num_input_tokens_seen": 17216790,
      "step": 805,
      "time_per_iteration": 3.22357439994812
    },
    {
      "auxiliary_loss_clip": 0.0121738,
      "auxiliary_loss_mlp": 0.01045988,
      "balance_loss_clip": 1.06415582,
      "balance_loss_mlp": 1.03172421,
      "epoch": 0.09691576985510732,
      "flos": 21945514020480.0,
      "grad_norm": 1.9788814838925464,
      "language_loss": 0.73775649,
      "learning_rate": 3.9532872948706395e-06,
      "loss": 0.76039016,
      "num_input_tokens_seen": 17236285,
      "step": 806,
      "time_per_iteration": 2.7168655395507812
    },
    {
      "auxiliary_loss_clip": 0.01220711,
      "auxiliary_loss_mlp": 0.01047229,
      "balance_loss_clip": 1.06429696,
      "balance_loss_mlp": 1.03431857,
      "epoch": 0.09703601274574641,
      "flos": 17965211103360.0,
      "grad_norm": 2.846603981391811,
      "language_loss": 0.83077765,
      "learning_rate": 3.9531197725811845e-06,
      "loss": 0.85345703,
      "num_input_tokens_seen": 17251670,
      "step": 807,
      "time_per_iteration": 2.551278591156006
    },
    {
      "auxiliary_loss_clip": 0.01252512,
      "auxiliary_loss_mlp": 0.01048074,
      "balance_loss_clip": 1.07249331,
      "balance_loss_mlp": 1.03608131,
      "epoch": 0.0971562556363855,
      "flos": 22162162901760.0,
      "grad_norm": 2.3128161223949384,
      "language_loss": 0.88030207,
      "learning_rate": 3.952951954005013e-06,
      "loss": 0.90330791,
      "num_input_tokens_seen": 17271355,
      "step": 808,
      "time_per_iteration": 2.4848601818084717
    },
    {
      "auxiliary_loss_clip": 0.01216005,
      "auxiliary_loss_mlp": 0.01037847,
      "balance_loss_clip": 1.06033742,
      "balance_loss_mlp": 1.0267663,
      "epoch": 0.0972764985270246,
      "flos": 25848716394240.0,
      "grad_norm": 1.8299890768925609,
      "language_loss": 0.84859526,
      "learning_rate": 3.952783839167584e-06,
      "loss": 0.8711338,
      "num_input_tokens_seen": 17291400,
      "step": 809,
      "time_per_iteration": 2.5771279335021973
    },
    {
      "auxiliary_loss_clip": 0.01233059,
      "auxiliary_loss_mlp": 0.01050485,
      "balance_loss_clip": 1.06492054,
      "balance_loss_mlp": 1.03826046,
      "epoch": 0.09739674141766368,
      "flos": 20339373375360.0,
      "grad_norm": 2.4756381520636106,
      "language_loss": 0.74346733,
      "learning_rate": 3.952615428094398e-06,
      "loss": 0.76630276,
      "num_input_tokens_seen": 17310920,
      "step": 810,
      "time_per_iteration": 2.5418622493743896
    },
    {
      "auxiliary_loss_clip": 0.01179405,
      "auxiliary_loss_mlp": 0.0104426,
      "balance_loss_clip": 1.05624688,
      "balance_loss_mlp": 1.03253555,
      "epoch": 0.09751698430830277,
      "flos": 15743059188480.0,
      "grad_norm": 1.8843149909741272,
      "language_loss": 0.73117381,
      "learning_rate": 3.952446720811004e-06,
      "loss": 0.75341046,
      "num_input_tokens_seen": 17329245,
      "step": 811,
      "time_per_iteration": 2.5892035961151123
    },
    {
      "auxiliary_loss_clip": 0.01098297,
      "auxiliary_loss_mlp": 0.01012904,
      "balance_loss_clip": 1.02645946,
      "balance_loss_mlp": 1.00827909,
      "epoch": 0.09763722719894186,
      "flos": 63716806800000.0,
      "grad_norm": 0.9533060309205846,
      "language_loss": 0.63613534,
      "learning_rate": 3.952277717342995e-06,
      "loss": 0.65724736,
      "num_input_tokens_seen": 17395680,
      "step": 812,
      "time_per_iteration": 3.256331205368042
    },
    {
      "auxiliary_loss_clip": 0.01225391,
      "auxiliary_loss_mlp": 0.01047311,
      "balance_loss_clip": 1.06686258,
      "balance_loss_mlp": 1.03509188,
      "epoch": 0.09775747008958095,
      "flos": 22090916275200.0,
      "grad_norm": 3.7191780324000128,
      "language_loss": 0.85725152,
      "learning_rate": 3.952108417716009e-06,
      "loss": 0.87997854,
      "num_input_tokens_seen": 17415135,
      "step": 813,
      "time_per_iteration": 2.5758559703826904
    },
    {
      "auxiliary_loss_clip": 0.01238119,
      "auxiliary_loss_mlp": 0.01037425,
      "balance_loss_clip": 1.06992388,
      "balance_loss_mlp": 1.02531981,
      "epoch": 0.09787771298022005,
      "flos": 21286050272640.0,
      "grad_norm": 2.002374886920885,
      "language_loss": 0.85238481,
      "learning_rate": 3.951938821955727e-06,
      "loss": 0.87514025,
      "num_input_tokens_seen": 17434535,
      "step": 814,
      "time_per_iteration": 2.526736259460449
    },
    {
      "auxiliary_loss_clip": 0.01220191,
      "auxiliary_loss_mlp": 0.01054191,
      "balance_loss_clip": 1.06682432,
      "balance_loss_mlp": 1.04103017,
      "epoch": 0.09799795587085913,
      "flos": 22054574689920.0,
      "grad_norm": 1.761950191157259,
      "language_loss": 0.76614404,
      "learning_rate": 3.9517689300878786e-06,
      "loss": 0.78888786,
      "num_input_tokens_seen": 17454270,
      "step": 815,
      "time_per_iteration": 2.563603162765503
    },
    {
      "auxiliary_loss_clip": 0.0124907,
      "auxiliary_loss_mlp": 0.0104743,
      "balance_loss_clip": 1.06668615,
      "balance_loss_mlp": 1.035074,
      "epoch": 0.09811819876149823,
      "flos": 22163743100160.0,
      "grad_norm": 1.6154304050276231,
      "language_loss": 0.78369677,
      "learning_rate": 3.951598742138236e-06,
      "loss": 0.80666184,
      "num_input_tokens_seen": 17472995,
      "step": 816,
      "time_per_iteration": 2.5028696060180664
    },
    {
      "auxiliary_loss_clip": 0.01222801,
      "auxiliary_loss_mlp": 0.01044022,
      "balance_loss_clip": 1.0601058,
      "balance_loss_mlp": 1.03222632,
      "epoch": 0.09823844165213731,
      "flos": 22231111057920.0,
      "grad_norm": 2.087037870464484,
      "language_loss": 0.79446566,
      "learning_rate": 3.951428258132615e-06,
      "loss": 0.81713378,
      "num_input_tokens_seen": 17491115,
      "step": 817,
      "time_per_iteration": 2.542043685913086
    },
    {
      "auxiliary_loss_clip": 0.01221113,
      "auxiliary_loss_mlp": 0.00766135,
      "balance_loss_clip": 1.06701481,
      "balance_loss_mlp": 1.000049,
      "epoch": 0.09835868454277641,
      "flos": 22487728798080.0,
      "grad_norm": 1.8101882484421883,
      "language_loss": 0.84654403,
      "learning_rate": 3.951257478096879e-06,
      "loss": 0.86641657,
      "num_input_tokens_seen": 17509480,
      "step": 818,
      "time_per_iteration": 2.5610127449035645
    },
    {
      "auxiliary_loss_clip": 0.01225898,
      "auxiliary_loss_mlp": 0.00767,
      "balance_loss_clip": 1.06874037,
      "balance_loss_mlp": 1.00004125,
      "epoch": 0.0984789274334155,
      "flos": 16362554077440.0,
      "grad_norm": 11.664055505286736,
      "language_loss": 0.68325204,
      "learning_rate": 3.951086402056936e-06,
      "loss": 0.70318103,
      "num_input_tokens_seen": 17524080,
      "step": 819,
      "time_per_iteration": 2.517002582550049
    },
    {
      "auxiliary_loss_clip": 0.01151403,
      "auxiliary_loss_mlp": 0.01040122,
      "balance_loss_clip": 1.05917859,
      "balance_loss_mlp": 1.02807021,
      "epoch": 0.09859917032405459,
      "flos": 24243545416320.0,
      "grad_norm": 1.6781942512696926,
      "language_loss": 0.83340466,
      "learning_rate": 3.950915030038735e-06,
      "loss": 0.85531998,
      "num_input_tokens_seen": 17543875,
      "step": 820,
      "time_per_iteration": 2.673495292663574
    },
    {
      "auxiliary_loss_clip": 0.01231708,
      "auxiliary_loss_mlp": 0.0104036,
      "balance_loss_clip": 1.06697381,
      "balance_loss_mlp": 1.02834392,
      "epoch": 0.09871941321469369,
      "flos": 17420195064960.0,
      "grad_norm": 2.1538299088470545,
      "language_loss": 0.83764315,
      "learning_rate": 3.9507433620682765e-06,
      "loss": 0.86036384,
      "num_input_tokens_seen": 17560810,
      "step": 821,
      "time_per_iteration": 3.302797794342041
    },
    {
      "auxiliary_loss_clip": 0.01204123,
      "auxiliary_loss_mlp": 0.01044282,
      "balance_loss_clip": 1.06205118,
      "balance_loss_mlp": 1.03183103,
      "epoch": 0.09883965610533277,
      "flos": 28477341590400.0,
      "grad_norm": 1.8392728428405152,
      "language_loss": 0.88368249,
      "learning_rate": 3.9505713981716e-06,
      "loss": 0.90616655,
      "num_input_tokens_seen": 17583640,
      "step": 822,
      "time_per_iteration": 2.639491319656372
    },
    {
      "auxiliary_loss_clip": 0.01217438,
      "auxiliary_loss_mlp": 0.01040398,
      "balance_loss_clip": 1.06719327,
      "balance_loss_mlp": 1.02942514,
      "epoch": 0.09895989899597187,
      "flos": 23693932437120.0,
      "grad_norm": 1.7954747837844185,
      "language_loss": 0.8082515,
      "learning_rate": 3.950399138374795e-06,
      "loss": 0.83082992,
      "num_input_tokens_seen": 17602720,
      "step": 823,
      "time_per_iteration": 2.5586977005004883
    },
    {
      "auxiliary_loss_clip": 0.01235169,
      "auxiliary_loss_mlp": 0.0104385,
      "balance_loss_clip": 1.06695068,
      "balance_loss_mlp": 1.03111267,
      "epoch": 0.09908014188661095,
      "flos": 24679608526080.0,
      "grad_norm": 1.5718540344801786,
      "language_loss": 0.74346745,
      "learning_rate": 3.95022658270399e-06,
      "loss": 0.76625764,
      "num_input_tokens_seen": 17623085,
      "step": 824,
      "time_per_iteration": 2.5423552989959717
    },
    {
      "auxiliary_loss_clip": 0.01218675,
      "auxiliary_loss_mlp": 0.01043008,
      "balance_loss_clip": 1.06728852,
      "balance_loss_mlp": 1.03121233,
      "epoch": 0.09920038477725004,
      "flos": 14064307200000.0,
      "grad_norm": 1.8990821428257043,
      "language_loss": 0.78294086,
      "learning_rate": 3.9500537311853635e-06,
      "loss": 0.80555767,
      "num_input_tokens_seen": 17641040,
      "step": 825,
      "time_per_iteration": 3.282687187194824
    },
    {
      "auxiliary_loss_clip": 0.0123229,
      "auxiliary_loss_mlp": 0.01039592,
      "balance_loss_clip": 1.06240439,
      "balance_loss_mlp": 1.02661586,
      "epoch": 0.09932062766788914,
      "flos": 13407070095360.0,
      "grad_norm": 2.3387405094626996,
      "language_loss": 0.83182621,
      "learning_rate": 3.949880583845136e-06,
      "loss": 0.854545,
      "num_input_tokens_seen": 17659115,
      "step": 826,
      "time_per_iteration": 2.4942092895507812
    },
    {
      "auxiliary_loss_clip": 0.01218665,
      "auxiliary_loss_mlp": 0.01036333,
      "balance_loss_clip": 1.06479549,
      "balance_loss_mlp": 1.02437007,
      "epoch": 0.09944087055852822,
      "flos": 19500751566720.0,
      "grad_norm": 1.7635335002744208,
      "language_loss": 0.8168053,
      "learning_rate": 3.949707140709575e-06,
      "loss": 0.83935529,
      "num_input_tokens_seen": 17678845,
      "step": 827,
      "time_per_iteration": 3.41426157951355
    },
    {
      "auxiliary_loss_clip": 0.0123503,
      "auxiliary_loss_mlp": 0.01041133,
      "balance_loss_clip": 1.06363916,
      "balance_loss_mlp": 1.02837777,
      "epoch": 0.09956111344916732,
      "flos": 17749100926080.0,
      "grad_norm": 2.188741142924056,
      "language_loss": 0.83308917,
      "learning_rate": 3.949533401804991e-06,
      "loss": 0.85585082,
      "num_input_tokens_seen": 17695750,
      "step": 828,
      "time_per_iteration": 3.250134229660034
    },
    {
      "auxiliary_loss_clip": 0.01233639,
      "auxiliary_loss_mlp": 0.00766858,
      "balance_loss_clip": 1.06704879,
      "balance_loss_mlp": 0.99998927,
      "epoch": 0.0996813563398064,
      "flos": 17967581400960.0,
      "grad_norm": 1.9560624122765464,
      "language_loss": 0.90871429,
      "learning_rate": 3.949359367157739e-06,
      "loss": 0.92871928,
      "num_input_tokens_seen": 17714445,
      "step": 829,
      "time_per_iteration": 2.4974539279937744
    },
    {
      "auxiliary_loss_clip": 0.01239137,
      "auxiliary_loss_mlp": 0.01043372,
      "balance_loss_clip": 1.06791723,
      "balance_loss_mlp": 1.03074765,
      "epoch": 0.0998015992304455,
      "flos": 17457039440640.0,
      "grad_norm": 1.9942286437401784,
      "language_loss": 0.75674176,
      "learning_rate": 3.949185036794222e-06,
      "loss": 0.77956688,
      "num_input_tokens_seen": 17732455,
      "step": 830,
      "time_per_iteration": 2.5000433921813965
    },
    {
      "auxiliary_loss_clip": 0.01249127,
      "auxiliary_loss_mlp": 0.01044695,
      "balance_loss_clip": 1.06883395,
      "balance_loss_mlp": 1.03331614,
      "epoch": 0.0999218421210846,
      "flos": 25888757080320.0,
      "grad_norm": 1.6033214206592308,
      "language_loss": 0.78983754,
      "learning_rate": 3.949010410740884e-06,
      "loss": 0.81277573,
      "num_input_tokens_seen": 17755280,
      "step": 831,
      "time_per_iteration": 2.525277853012085
    },
    {
      "auxiliary_loss_clip": 0.01208779,
      "auxiliary_loss_mlp": 0.0076628,
      "balance_loss_clip": 1.06050897,
      "balance_loss_mlp": 1.00001538,
      "epoch": 0.10004208501172368,
      "flos": 21215916967680.0,
      "grad_norm": 1.62522100689321,
      "language_loss": 0.86340111,
      "learning_rate": 3.948835489024216e-06,
      "loss": 0.88315165,
      "num_input_tokens_seen": 17775015,
      "step": 832,
      "time_per_iteration": 2.541314125061035
    },
    {
      "auxiliary_loss_clip": 0.01236704,
      "auxiliary_loss_mlp": 0.01045288,
      "balance_loss_clip": 1.06627107,
      "balance_loss_mlp": 1.03300381,
      "epoch": 0.10016232790236278,
      "flos": 17348409734400.0,
      "grad_norm": 3.1618027524712726,
      "language_loss": 0.90499282,
      "learning_rate": 3.948660271670755e-06,
      "loss": 0.9278127,
      "num_input_tokens_seen": 17792165,
      "step": 833,
      "time_per_iteration": 2.4987680912017822
    },
    {
      "auxiliary_loss_clip": 0.01215297,
      "auxiliary_loss_mlp": 0.01044734,
      "balance_loss_clip": 1.06385088,
      "balance_loss_mlp": 1.03330183,
      "epoch": 0.10028257079300186,
      "flos": 25666541591040.0,
      "grad_norm": 2.5402864867469956,
      "language_loss": 0.84196162,
      "learning_rate": 3.948484758707079e-06,
      "loss": 0.86456198,
      "num_input_tokens_seen": 17811765,
      "step": 834,
      "time_per_iteration": 2.575428009033203
    },
    {
      "auxiliary_loss_clip": 0.01193654,
      "auxiliary_loss_mlp": 0.01041632,
      "balance_loss_clip": 1.05822778,
      "balance_loss_mlp": 1.02848887,
      "epoch": 0.10040281368364096,
      "flos": 25156035544320.0,
      "grad_norm": 2.205290033562829,
      "language_loss": 0.83342886,
      "learning_rate": 3.948308950159815e-06,
      "loss": 0.85578179,
      "num_input_tokens_seen": 17830445,
      "step": 835,
      "time_per_iteration": 2.605041980743408
    },
    {
      "auxiliary_loss_clip": 0.01198919,
      "auxiliary_loss_mlp": 0.01047501,
      "balance_loss_clip": 1.05968094,
      "balance_loss_mlp": 1.03349376,
      "epoch": 0.10052305657428004,
      "flos": 17603303621760.0,
      "grad_norm": 3.343470919223055,
      "language_loss": 0.75939095,
      "learning_rate": 3.9481328460556326e-06,
      "loss": 0.78185511,
      "num_input_tokens_seen": 17847665,
      "step": 836,
      "time_per_iteration": 2.577462911605835
    },
    {
      "auxiliary_loss_clip": 0.01209342,
      "auxiliary_loss_mlp": 0.01038248,
      "balance_loss_clip": 1.06079316,
      "balance_loss_mlp": 1.0261606,
      "epoch": 0.10064329946491914,
      "flos": 18660154510080.0,
      "grad_norm": 2.119490816381886,
      "language_loss": 0.89630353,
      "learning_rate": 3.9479564464212455e-06,
      "loss": 0.91877943,
      "num_input_tokens_seen": 17866825,
      "step": 837,
      "time_per_iteration": 2.529484987258911
    },
    {
      "auxiliary_loss_clip": 0.01255556,
      "auxiliary_loss_mlp": 0.01040527,
      "balance_loss_clip": 1.06921005,
      "balance_loss_mlp": 1.02846348,
      "epoch": 0.10076354235555823,
      "flos": 17199056983680.0,
      "grad_norm": 2.4916729210920843,
      "language_loss": 0.76543212,
      "learning_rate": 3.947779751283414e-06,
      "loss": 0.7883929,
      "num_input_tokens_seen": 17883995,
      "step": 838,
      "time_per_iteration": 2.4439988136291504
    },
    {
      "auxiliary_loss_clip": 0.01237269,
      "auxiliary_loss_mlp": 0.00767009,
      "balance_loss_clip": 1.07233191,
      "balance_loss_mlp": 0.99994469,
      "epoch": 0.10088378524619732,
      "flos": 22962252395520.0,
      "grad_norm": 1.7667242844680333,
      "language_loss": 0.76098895,
      "learning_rate": 3.947602760668944e-06,
      "loss": 0.78103173,
      "num_input_tokens_seen": 17903785,
      "step": 839,
      "time_per_iteration": 2.494786262512207
    },
    {
      "auxiliary_loss_clip": 0.012346,
      "auxiliary_loss_mlp": 0.01044215,
      "balance_loss_clip": 1.06970143,
      "balance_loss_mlp": 1.03175759,
      "epoch": 0.10100402813683641,
      "flos": 37885828746240.0,
      "grad_norm": 2.005843276947239,
      "language_loss": 0.71448809,
      "learning_rate": 3.947425474604684e-06,
      "loss": 0.73727626,
      "num_input_tokens_seen": 17927720,
      "step": 840,
      "time_per_iteration": 2.651487112045288
    },
    {
      "auxiliary_loss_clip": 0.01217405,
      "auxiliary_loss_mlp": 0.01048681,
      "balance_loss_clip": 1.06426644,
      "balance_loss_mlp": 1.03670037,
      "epoch": 0.1011242710274755,
      "flos": 21543458112000.0,
      "grad_norm": 2.5419988743685646,
      "language_loss": 0.92191446,
      "learning_rate": 3.947247893117528e-06,
      "loss": 0.94457537,
      "num_input_tokens_seen": 17946225,
      "step": 841,
      "time_per_iteration": 2.5118865966796875
    },
    {
      "auxiliary_loss_clip": 0.01230428,
      "auxiliary_loss_mlp": 0.01046582,
      "balance_loss_clip": 1.06437933,
      "balance_loss_mlp": 1.03419077,
      "epoch": 0.10124451391811459,
      "flos": 13621456419840.0,
      "grad_norm": 3.355128122581476,
      "language_loss": 0.69190425,
      "learning_rate": 3.947070016234413e-06,
      "loss": 0.71467435,
      "num_input_tokens_seen": 17962015,
      "step": 842,
      "time_per_iteration": 2.452122688293457
    },
    {
      "auxiliary_loss_clip": 0.01229044,
      "auxiliary_loss_mlp": 0.01042643,
      "balance_loss_clip": 1.06816494,
      "balance_loss_mlp": 1.02964902,
      "epoch": 0.10136475680875369,
      "flos": 16649228522880.0,
      "grad_norm": 2.8394286288835353,
      "language_loss": 0.74950635,
      "learning_rate": 3.946891843982326e-06,
      "loss": 0.77222323,
      "num_input_tokens_seen": 17979680,
      "step": 843,
      "time_per_iteration": 2.517972707748413
    },
    {
      "auxiliary_loss_clip": 0.0123458,
      "auxiliary_loss_mlp": 0.01040493,
      "balance_loss_clip": 1.06779242,
      "balance_loss_mlp": 1.02793479,
      "epoch": 0.10148499969939277,
      "flos": 19461034103040.0,
      "grad_norm": 2.3815770918193384,
      "language_loss": 0.74619412,
      "learning_rate": 3.9467133763882935e-06,
      "loss": 0.76894486,
      "num_input_tokens_seen": 17998145,
      "step": 844,
      "time_per_iteration": 2.463911533355713
    },
    {
      "auxiliary_loss_clip": 0.01223991,
      "auxiliary_loss_mlp": 0.01045653,
      "balance_loss_clip": 1.06532001,
      "balance_loss_mlp": 1.03283191,
      "epoch": 0.10160524259003187,
      "flos": 21104988791040.0,
      "grad_norm": 2.137850562545883,
      "language_loss": 0.8633101,
      "learning_rate": 3.9465346134793905e-06,
      "loss": 0.88600647,
      "num_input_tokens_seen": 18017955,
      "step": 845,
      "time_per_iteration": 2.5580990314483643
    },
    {
      "auxiliary_loss_clip": 0.01206699,
      "auxiliary_loss_mlp": 0.0104082,
      "balance_loss_clip": 1.0662179,
      "balance_loss_mlp": 1.02924466,
      "epoch": 0.10172548548067095,
      "flos": 17712687513600.0,
      "grad_norm": 1.9391952462209916,
      "language_loss": 0.79561269,
      "learning_rate": 3.9463555552827335e-06,
      "loss": 0.81808794,
      "num_input_tokens_seen": 18035125,
      "step": 846,
      "time_per_iteration": 2.539644718170166
    },
    {
      "auxiliary_loss_clip": 0.012232,
      "auxiliary_loss_mlp": 0.01048094,
      "balance_loss_clip": 1.06467104,
      "balance_loss_mlp": 1.03598249,
      "epoch": 0.10184572837131005,
      "flos": 21104845136640.0,
      "grad_norm": 2.3799335297335693,
      "language_loss": 0.86475122,
      "learning_rate": 3.946176201825487e-06,
      "loss": 0.88746417,
      "num_input_tokens_seen": 18053160,
      "step": 847,
      "time_per_iteration": 3.2794134616851807
    },
    {
      "auxiliary_loss_clip": 0.01220933,
      "auxiliary_loss_mlp": 0.01048109,
      "balance_loss_clip": 1.06788039,
      "balance_loss_mlp": 1.03592002,
      "epoch": 0.10196597126194913,
      "flos": 26067591918720.0,
      "grad_norm": 1.8757728826935736,
      "language_loss": 0.83539331,
      "learning_rate": 3.9459965531348575e-06,
      "loss": 0.85808378,
      "num_input_tokens_seen": 18072815,
      "step": 848,
      "time_per_iteration": 2.5506062507629395
    },
    {
      "auxiliary_loss_clip": 0.01220936,
      "auxiliary_loss_mlp": 0.00766942,
      "balance_loss_clip": 1.06723249,
      "balance_loss_mlp": 1.00012016,
      "epoch": 0.10208621415258823,
      "flos": 29314634595840.0,
      "grad_norm": 2.1663592208987756,
      "language_loss": 0.85718745,
      "learning_rate": 3.945816609238098e-06,
      "loss": 0.87706625,
      "num_input_tokens_seen": 18092225,
      "step": 849,
      "time_per_iteration": 2.592404842376709
    },
    {
      "auxiliary_loss_clip": 0.01180991,
      "auxiliary_loss_mlp": 0.0104352,
      "balance_loss_clip": 1.06252837,
      "balance_loss_mlp": 1.0308001,
      "epoch": 0.10220645704322733,
      "flos": 23805794367360.0,
      "grad_norm": 2.5953664706914634,
      "language_loss": 0.85040855,
      "learning_rate": 3.945636370162507e-06,
      "loss": 0.8726536,
      "num_input_tokens_seen": 18112335,
      "step": 850,
      "time_per_iteration": 2.6394407749176025
    },
    {
      "auxiliary_loss_clip": 0.01231576,
      "auxiliary_loss_mlp": 0.01047507,
      "balance_loss_clip": 1.0672524,
      "balance_loss_mlp": 1.03615785,
      "epoch": 0.10232669993386641,
      "flos": 23218546913280.0,
      "grad_norm": 2.0973420090993704,
      "language_loss": 0.79131997,
      "learning_rate": 3.945455835935425e-06,
      "loss": 0.81411076,
      "num_input_tokens_seen": 18131520,
      "step": 851,
      "time_per_iteration": 2.672485589981079
    },
    {
      "auxiliary_loss_clip": 0.01220252,
      "auxiliary_loss_mlp": 0.01046661,
      "balance_loss_clip": 1.06529629,
      "balance_loss_mlp": 1.03484726,
      "epoch": 0.1024469428245055,
      "flos": 22922929981440.0,
      "grad_norm": 2.166325375407431,
      "language_loss": 0.75056547,
      "learning_rate": 3.94527500658424e-06,
      "loss": 0.77323461,
      "num_input_tokens_seen": 18149185,
      "step": 852,
      "time_per_iteration": 3.268260955810547
    },
    {
      "auxiliary_loss_clip": 0.01188679,
      "auxiliary_loss_mlp": 0.01038561,
      "balance_loss_clip": 1.06291997,
      "balance_loss_mlp": 1.02733111,
      "epoch": 0.10256718571514459,
      "flos": 31359495957120.0,
      "grad_norm": 1.9073348344713514,
      "language_loss": 0.8100422,
      "learning_rate": 3.945093882136382e-06,
      "loss": 0.83231461,
      "num_input_tokens_seen": 18172960,
      "step": 853,
      "time_per_iteration": 3.671428918838501
    },
    {
      "auxiliary_loss_clip": 0.01216055,
      "auxiliary_loss_mlp": 0.00765419,
      "balance_loss_clip": 1.06689453,
      "balance_loss_mlp": 1.00006795,
      "epoch": 0.10268742860578368,
      "flos": 23474877344640.0,
      "grad_norm": 2.0369046488082856,
      "language_loss": 0.84509921,
      "learning_rate": 3.944912462619329e-06,
      "loss": 0.86491388,
      "num_input_tokens_seen": 18191925,
      "step": 854,
      "time_per_iteration": 2.6228551864624023
    },
    {
      "auxiliary_loss_clip": 0.01221666,
      "auxiliary_loss_mlp": 0.01049189,
      "balance_loss_clip": 1.06585908,
      "balance_loss_mlp": 1.03633189,
      "epoch": 0.10280767149642277,
      "flos": 25520313323520.0,
      "grad_norm": 3.404673092178469,
      "language_loss": 0.80233961,
      "learning_rate": 3.9447307480606025e-06,
      "loss": 0.82504815,
      "num_input_tokens_seen": 18212010,
      "step": 855,
      "time_per_iteration": 3.3405556678771973
    },
    {
      "auxiliary_loss_clip": 0.01214746,
      "auxiliary_loss_mlp": 0.01043223,
      "balance_loss_clip": 1.06535614,
      "balance_loss_mlp": 1.03015125,
      "epoch": 0.10292791438706186,
      "flos": 17347691462400.0,
      "grad_norm": 3.003890064888439,
      "language_loss": 0.89972758,
      "learning_rate": 3.944548738487767e-06,
      "loss": 0.92230725,
      "num_input_tokens_seen": 18229525,
      "step": 856,
      "time_per_iteration": 2.5274417400360107
    },
    {
      "auxiliary_loss_clip": 0.01257635,
      "auxiliary_loss_mlp": 0.01044011,
      "balance_loss_clip": 1.07388747,
      "balance_loss_mlp": 1.03259122,
      "epoch": 0.10304815727770096,
      "flos": 27052693390080.0,
      "grad_norm": 2.100832403580187,
      "language_loss": 0.90825057,
      "learning_rate": 3.944366433928434e-06,
      "loss": 0.93126702,
      "num_input_tokens_seen": 18249505,
      "step": 857,
      "time_per_iteration": 2.5308327674865723
    },
    {
      "auxiliary_loss_clip": 0.01213004,
      "auxiliary_loss_mlp": 0.01047817,
      "balance_loss_clip": 1.06232917,
      "balance_loss_mlp": 1.03597975,
      "epoch": 0.10316840016834004,
      "flos": 22782591544320.0,
      "grad_norm": 2.237102036843472,
      "language_loss": 0.83802682,
      "learning_rate": 3.9441838344102594e-06,
      "loss": 0.86063504,
      "num_input_tokens_seen": 18269230,
      "step": 858,
      "time_per_iteration": 2.627354860305786
    },
    {
      "auxiliary_loss_clip": 0.01226715,
      "auxiliary_loss_mlp": 0.01042823,
      "balance_loss_clip": 1.06885815,
      "balance_loss_mlp": 1.03116405,
      "epoch": 0.10328864305897914,
      "flos": 20704584908160.0,
      "grad_norm": 3.303488850851589,
      "language_loss": 0.67042291,
      "learning_rate": 3.944000939960943e-06,
      "loss": 0.69311827,
      "num_input_tokens_seen": 18287955,
      "step": 859,
      "time_per_iteration": 2.5827934741973877
    },
    {
      "auxiliary_loss_clip": 0.01237978,
      "auxiliary_loss_mlp": 0.01039218,
      "balance_loss_clip": 1.06616306,
      "balance_loss_mlp": 1.02867961,
      "epoch": 0.10340888594961822,
      "flos": 28478814048000.0,
      "grad_norm": 6.370256847344052,
      "language_loss": 0.80057919,
      "learning_rate": 3.943817750608229e-06,
      "loss": 0.8233512,
      "num_input_tokens_seen": 18310505,
      "step": 860,
      "time_per_iteration": 2.584465265274048
    },
    {
      "auxiliary_loss_clip": 0.01241111,
      "auxiliary_loss_mlp": 0.0104005,
      "balance_loss_clip": 1.07246232,
      "balance_loss_mlp": 1.0286541,
      "epoch": 0.10352912884025732,
      "flos": 13370333460480.0,
      "grad_norm": 2.175762921988747,
      "language_loss": 0.81847197,
      "learning_rate": 3.943634266379908e-06,
      "loss": 0.84128356,
      "num_input_tokens_seen": 18327400,
      "step": 861,
      "time_per_iteration": 2.47428822517395
    },
    {
      "auxiliary_loss_clip": 0.01237188,
      "auxiliary_loss_mlp": 0.01040193,
      "balance_loss_clip": 1.06687546,
      "balance_loss_mlp": 1.02873671,
      "epoch": 0.10364937173089642,
      "flos": 25558558329600.0,
      "grad_norm": 1.8234020500404189,
      "language_loss": 0.84948528,
      "learning_rate": 3.943450487303815e-06,
      "loss": 0.87225908,
      "num_input_tokens_seen": 18347895,
      "step": 862,
      "time_per_iteration": 2.5316033363342285
    },
    {
      "auxiliary_loss_clip": 0.01232499,
      "auxiliary_loss_mlp": 0.01039847,
      "balance_loss_clip": 1.06812179,
      "balance_loss_mlp": 1.02818191,
      "epoch": 0.1037696146215355,
      "flos": 21215486004480.0,
      "grad_norm": 1.7833454724004991,
      "language_loss": 0.85031354,
      "learning_rate": 3.943266413407827e-06,
      "loss": 0.87303698,
      "num_input_tokens_seen": 18367170,
      "step": 863,
      "time_per_iteration": 2.4865143299102783
    },
    {
      "auxiliary_loss_clip": 0.01237765,
      "auxiliary_loss_mlp": 0.01043358,
      "balance_loss_clip": 1.07003617,
      "balance_loss_mlp": 1.03169918,
      "epoch": 0.1038898575121746,
      "flos": 25807382818560.0,
      "grad_norm": 1.8825425416363033,
      "language_loss": 0.84739596,
      "learning_rate": 3.94308204471987e-06,
      "loss": 0.87020719,
      "num_input_tokens_seen": 18386185,
      "step": 864,
      "time_per_iteration": 2.539384365081787
    },
    {
      "auxiliary_loss_clip": 0.01205002,
      "auxiliary_loss_mlp": 0.01035991,
      "balance_loss_clip": 1.06376529,
      "balance_loss_mlp": 1.02414763,
      "epoch": 0.10401010040281368,
      "flos": 19062425900160.0,
      "grad_norm": 3.319257796739146,
      "language_loss": 0.7492885,
      "learning_rate": 3.942897381267912e-06,
      "loss": 0.77169847,
      "num_input_tokens_seen": 18402550,
      "step": 865,
      "time_per_iteration": 2.544003486633301
    },
    {
      "auxiliary_loss_clip": 0.01243042,
      "auxiliary_loss_mlp": 0.01034302,
      "balance_loss_clip": 1.07225144,
      "balance_loss_mlp": 1.02283382,
      "epoch": 0.10413034329345278,
      "flos": 16355119962240.0,
      "grad_norm": 2.7712562649215355,
      "language_loss": 0.65953261,
      "learning_rate": 3.942712423079965e-06,
      "loss": 0.68230605,
      "num_input_tokens_seen": 18418940,
      "step": 866,
      "time_per_iteration": 2.4790189266204834
    },
    {
      "auxiliary_loss_clip": 0.01184209,
      "auxiliary_loss_mlp": 0.01035459,
      "balance_loss_clip": 1.0545032,
      "balance_loss_mlp": 1.02528405,
      "epoch": 0.10425058618409186,
      "flos": 17236511890560.0,
      "grad_norm": 2.526350744957532,
      "language_loss": 0.90156156,
      "learning_rate": 3.942527170184088e-06,
      "loss": 0.92375827,
      "num_input_tokens_seen": 18435560,
      "step": 867,
      "time_per_iteration": 2.5399932861328125
    },
    {
      "auxiliary_loss_clip": 0.01253517,
      "auxiliary_loss_mlp": 0.01041219,
      "balance_loss_clip": 1.07183504,
      "balance_loss_mlp": 1.0294168,
      "epoch": 0.10437082907473096,
      "flos": 17967365919360.0,
      "grad_norm": 2.264853869802826,
      "language_loss": 0.77554595,
      "learning_rate": 3.942341622608385e-06,
      "loss": 0.79849333,
      "num_input_tokens_seen": 18452590,
      "step": 868,
      "time_per_iteration": 2.451336145401001
    },
    {
      "auxiliary_loss_clip": 0.01222189,
      "auxiliary_loss_mlp": 0.01043724,
      "balance_loss_clip": 1.0691812,
      "balance_loss_mlp": 1.03211904,
      "epoch": 0.10449107196537005,
      "flos": 36283315374720.0,
      "grad_norm": 2.001293168578348,
      "language_loss": 0.78016436,
      "learning_rate": 3.942155780381001e-06,
      "loss": 0.80282348,
      "num_input_tokens_seen": 18476325,
      "step": 869,
      "time_per_iteration": 2.6595215797424316
    },
    {
      "auxiliary_loss_clip": 0.01220711,
      "auxiliary_loss_mlp": 0.0104597,
      "balance_loss_clip": 1.06374049,
      "balance_loss_mlp": 1.03341126,
      "epoch": 0.10461131485600914,
      "flos": 23802095266560.0,
      "grad_norm": 1.9106085158732284,
      "language_loss": 0.75781316,
      "learning_rate": 3.94196964353013e-06,
      "loss": 0.78047997,
      "num_input_tokens_seen": 18495775,
      "step": 870,
      "time_per_iteration": 2.5559327602386475
    },
    {
      "auxiliary_loss_clip": 0.01216956,
      "auxiliary_loss_mlp": 0.00765485,
      "balance_loss_clip": 1.06374252,
      "balance_loss_mlp": 1.00012648,
      "epoch": 0.10473155774664823,
      "flos": 18405476104320.0,
      "grad_norm": 1.9737760200382524,
      "language_loss": 0.80670154,
      "learning_rate": 3.941783212084008e-06,
      "loss": 0.82652593,
      "num_input_tokens_seen": 18513530,
      "step": 871,
      "time_per_iteration": 2.5382020473480225
    },
    {
      "auxiliary_loss_clip": 0.01205355,
      "auxiliary_loss_mlp": 0.01044931,
      "balance_loss_clip": 1.06511593,
      "balance_loss_mlp": 1.03271246,
      "epoch": 0.10485180063728732,
      "flos": 25592637358080.0,
      "grad_norm": 2.4557297627007997,
      "language_loss": 0.79317546,
      "learning_rate": 3.941596486070916e-06,
      "loss": 0.81567836,
      "num_input_tokens_seen": 18531575,
      "step": 872,
      "time_per_iteration": 2.588190793991089
    },
    {
      "auxiliary_loss_clip": 0.01186879,
      "auxiliary_loss_mlp": 0.01036622,
      "balance_loss_clip": 1.06414151,
      "balance_loss_mlp": 1.02384257,
      "epoch": 0.10497204352792641,
      "flos": 27088747666560.0,
      "grad_norm": 2.1412634343711154,
      "language_loss": 0.58533347,
      "learning_rate": 3.941409465519182e-06,
      "loss": 0.60756838,
      "num_input_tokens_seen": 18552100,
      "step": 873,
      "time_per_iteration": 2.6472792625427246
    },
    {
      "auxiliary_loss_clip": 0.0122552,
      "auxiliary_loss_mlp": 0.01044806,
      "balance_loss_clip": 1.06407785,
      "balance_loss_mlp": 1.03165078,
      "epoch": 0.10509228641856551,
      "flos": 32858479353600.0,
      "grad_norm": 1.6447512575985188,
      "language_loss": 0.85198975,
      "learning_rate": 3.941222150457176e-06,
      "loss": 0.87469304,
      "num_input_tokens_seen": 18575355,
      "step": 874,
      "time_per_iteration": 3.423719882965088
    },
    {
      "auxiliary_loss_clip": 0.01240177,
      "auxiliary_loss_mlp": 0.01038778,
      "balance_loss_clip": 1.0676465,
      "balance_loss_mlp": 1.02667212,
      "epoch": 0.10521252930920459,
      "flos": 14319165173760.0,
      "grad_norm": 2.8463727202923286,
      "language_loss": 0.71149546,
      "learning_rate": 3.941034540913311e-06,
      "loss": 0.734285,
      "num_input_tokens_seen": 18592885,
      "step": 875,
      "time_per_iteration": 2.475145101547241
    },
    {
      "auxiliary_loss_clip": 0.01236973,
      "auxiliary_loss_mlp": 0.00766624,
      "balance_loss_clip": 1.07008505,
      "balance_loss_mlp": 1.00016022,
      "epoch": 0.10533277219984369,
      "flos": 21687028773120.0,
      "grad_norm": 1.5962355853127843,
      "language_loss": 0.82411563,
      "learning_rate": 3.940846636916051e-06,
      "loss": 0.84415162,
      "num_input_tokens_seen": 18612920,
      "step": 876,
      "time_per_iteration": 2.5428812503814697
    },
    {
      "auxiliary_loss_clip": 0.01220273,
      "auxiliary_loss_mlp": 0.01048191,
      "balance_loss_clip": 1.07053804,
      "balance_loss_mlp": 1.03508949,
      "epoch": 0.10545301509048277,
      "flos": 22269787027200.0,
      "grad_norm": 2.0829580387345055,
      "language_loss": 0.86610603,
      "learning_rate": 3.940658438493899e-06,
      "loss": 0.88879067,
      "num_input_tokens_seen": 18630765,
      "step": 877,
      "time_per_iteration": 2.545813798904419
    },
    {
      "auxiliary_loss_clip": 0.01254124,
      "auxiliary_loss_mlp": 0.01042168,
      "balance_loss_clip": 1.06617868,
      "balance_loss_mlp": 1.0295136,
      "epoch": 0.10557325798112187,
      "flos": 22199725549440.0,
      "grad_norm": 2.1651349502749775,
      "language_loss": 0.76163447,
      "learning_rate": 3.940469945675405e-06,
      "loss": 0.7845974,
      "num_input_tokens_seen": 18649150,
      "step": 878,
      "time_per_iteration": 3.246293067932129
    },
    {
      "auxiliary_loss_clip": 0.01166553,
      "auxiliary_loss_mlp": 0.01045867,
      "balance_loss_clip": 1.05816996,
      "balance_loss_mlp": 1.03444648,
      "epoch": 0.10569350087176095,
      "flos": 25775889569280.0,
      "grad_norm": 2.023790246003973,
      "language_loss": 0.91725224,
      "learning_rate": 3.940281158489163e-06,
      "loss": 0.93937647,
      "num_input_tokens_seen": 18668380,
      "step": 879,
      "time_per_iteration": 2.630768060684204
    },
    {
      "auxiliary_loss_clip": 0.01168387,
      "auxiliary_loss_mlp": 0.01044457,
      "balance_loss_clip": 1.0555619,
      "balance_loss_mlp": 1.03275084,
      "epoch": 0.10581374376240005,
      "flos": 17311385790720.0,
      "grad_norm": 7.253074076075061,
      "language_loss": 0.82890636,
      "learning_rate": 3.940092076963812e-06,
      "loss": 0.85103476,
      "num_input_tokens_seen": 18685875,
      "step": 880,
      "time_per_iteration": 3.491748809814453
    },
    {
      "auxiliary_loss_clip": 0.01216041,
      "auxiliary_loss_mlp": 0.01046981,
      "balance_loss_clip": 1.06327569,
      "balance_loss_mlp": 1.03451133,
      "epoch": 0.10593398665303914,
      "flos": 34349454017280.0,
      "grad_norm": 2.201171697155286,
      "language_loss": 0.78926682,
      "learning_rate": 3.9399027011280355e-06,
      "loss": 0.81189704,
      "num_input_tokens_seen": 18707970,
      "step": 881,
      "time_per_iteration": 2.6801040172576904
    },
    {
      "auxiliary_loss_clip": 0.0121879,
      "auxiliary_loss_mlp": 0.01040829,
      "balance_loss_clip": 1.06817174,
      "balance_loss_mlp": 1.02867508,
      "epoch": 0.10605422954367823,
      "flos": 23257977068160.0,
      "grad_norm": 2.1336259516275486,
      "language_loss": 0.77490532,
      "learning_rate": 3.939713031010561e-06,
      "loss": 0.7975015,
      "num_input_tokens_seen": 18726335,
      "step": 882,
      "time_per_iteration": 3.284637451171875
    },
    {
      "auxiliary_loss_clip": 0.0120143,
      "auxiliary_loss_mlp": 0.01042383,
      "balance_loss_clip": 1.06556129,
      "balance_loss_mlp": 1.02910328,
      "epoch": 0.10617447243431732,
      "flos": 22820118278400.0,
      "grad_norm": 2.3591702407021558,
      "language_loss": 0.77737093,
      "learning_rate": 3.939523066640163e-06,
      "loss": 0.7998091,
      "num_input_tokens_seen": 18745230,
      "step": 883,
      "time_per_iteration": 2.578439474105835
    },
    {
      "auxiliary_loss_clip": 0.01237568,
      "auxiliary_loss_mlp": 0.01041504,
      "balance_loss_clip": 1.06838441,
      "balance_loss_mlp": 1.02965486,
      "epoch": 0.10629471532495641,
      "flos": 24386577373440.0,
      "grad_norm": 1.8746076137111203,
      "language_loss": 0.81338352,
      "learning_rate": 3.939332808045657e-06,
      "loss": 0.83617419,
      "num_input_tokens_seen": 18764880,
      "step": 884,
      "time_per_iteration": 2.5560302734375
    },
    {
      "auxiliary_loss_clip": 0.0120491,
      "auxiliary_loss_mlp": 0.01043612,
      "balance_loss_clip": 1.0650779,
      "balance_loss_mlp": 1.03197157,
      "epoch": 0.1064149582155955,
      "flos": 21105491581440.0,
      "grad_norm": 1.6841616910548782,
      "language_loss": 0.84690976,
      "learning_rate": 3.939142255255906e-06,
      "loss": 0.86939502,
      "num_input_tokens_seen": 18785765,
      "step": 885,
      "time_per_iteration": 2.6163365840911865
    },
    {
      "auxiliary_loss_clip": 0.01236461,
      "auxiliary_loss_mlp": 0.01035274,
      "balance_loss_clip": 1.06873155,
      "balance_loss_mlp": 1.02271545,
      "epoch": 0.1065352011062346,
      "flos": 20702035042560.0,
      "grad_norm": 2.3705089487107935,
      "language_loss": 0.86455911,
      "learning_rate": 3.938951408299817e-06,
      "loss": 0.88727641,
      "num_input_tokens_seen": 18804605,
      "step": 886,
      "time_per_iteration": 2.5161798000335693
    },
    {
      "auxiliary_loss_clip": 0.01107957,
      "auxiliary_loss_mlp": 0.01022806,
      "balance_loss_clip": 1.05832243,
      "balance_loss_mlp": 1.01918221,
      "epoch": 0.10665544399687368,
      "flos": 62659632689280.0,
      "grad_norm": 0.8062078229451766,
      "language_loss": 0.5441612,
      "learning_rate": 3.938760267206342e-06,
      "loss": 0.56546885,
      "num_input_tokens_seen": 18866425,
      "step": 887,
      "time_per_iteration": 3.104126214981079
    },
    {
      "auxiliary_loss_clip": 0.01250889,
      "auxiliary_loss_mlp": 0.01039364,
      "balance_loss_clip": 1.0702492,
      "balance_loss_mlp": 1.02737117,
      "epoch": 0.10677568688751278,
      "flos": 26140382830080.0,
      "grad_norm": 2.2213090147609504,
      "language_loss": 0.79058969,
      "learning_rate": 3.938568832004475e-06,
      "loss": 0.81349224,
      "num_input_tokens_seen": 18885130,
      "step": 888,
      "time_per_iteration": 2.5349457263946533
    },
    {
      "auxiliary_loss_clip": 0.0121022,
      "auxiliary_loss_mlp": 0.01051649,
      "balance_loss_clip": 1.06313348,
      "balance_loss_mlp": 1.03900731,
      "epoch": 0.10689592977815186,
      "flos": 12786533712000.0,
      "grad_norm": 2.250505992968594,
      "language_loss": 0.75351048,
      "learning_rate": 3.938377102723257e-06,
      "loss": 0.77612913,
      "num_input_tokens_seen": 18902265,
      "step": 889,
      "time_per_iteration": 2.5150461196899414
    },
    {
      "auxiliary_loss_clip": 0.01171473,
      "auxiliary_loss_mlp": 0.01051904,
      "balance_loss_clip": 1.05809522,
      "balance_loss_mlp": 1.03889275,
      "epoch": 0.10701617266879096,
      "flos": 22126683242880.0,
      "grad_norm": 2.0272943104641965,
      "language_loss": 0.8326872,
      "learning_rate": 3.938185079391774e-06,
      "loss": 0.85492098,
      "num_input_tokens_seen": 18919310,
      "step": 890,
      "time_per_iteration": 2.6090900897979736
    },
    {
      "auxiliary_loss_clip": 0.01250107,
      "auxiliary_loss_mlp": 0.01035693,
      "balance_loss_clip": 1.06854534,
      "balance_loss_mlp": 1.02387333,
      "epoch": 0.10713641555943004,
      "flos": 19745625559680.0,
      "grad_norm": 2.4502619689593828,
      "language_loss": 1.05944252,
      "learning_rate": 3.937992762039157e-06,
      "loss": 1.08230042,
      "num_input_tokens_seen": 18932635,
      "step": 891,
      "time_per_iteration": 2.4560630321502686
    },
    {
      "auxiliary_loss_clip": 0.01232828,
      "auxiliary_loss_mlp": 0.01049007,
      "balance_loss_clip": 1.06847763,
      "balance_loss_mlp": 1.03724122,
      "epoch": 0.10725665845006914,
      "flos": 23952992302080.0,
      "grad_norm": 1.5648784372640385,
      "language_loss": 0.8026191,
      "learning_rate": 3.937800150694577e-06,
      "loss": 0.82543743,
      "num_input_tokens_seen": 18953810,
      "step": 892,
      "time_per_iteration": 2.628404378890991
    },
    {
      "auxiliary_loss_clip": 0.01188786,
      "auxiliary_loss_mlp": 0.01041942,
      "balance_loss_clip": 1.06502533,
      "balance_loss_mlp": 1.02937186,
      "epoch": 0.10737690134070824,
      "flos": 18551704371840.0,
      "grad_norm": 2.125893412374179,
      "language_loss": 0.75950187,
      "learning_rate": 3.937607245387255e-06,
      "loss": 0.78180915,
      "num_input_tokens_seen": 18973175,
      "step": 893,
      "time_per_iteration": 2.6575801372528076
    },
    {
      "auxiliary_loss_clip": 0.01226172,
      "auxiliary_loss_mlp": 0.01046969,
      "balance_loss_clip": 1.06583548,
      "balance_loss_mlp": 1.03556085,
      "epoch": 0.10749714423134732,
      "flos": 22707609903360.0,
      "grad_norm": 1.8954759059579405,
      "language_loss": 0.72281992,
      "learning_rate": 3.937414046146455e-06,
      "loss": 0.74555135,
      "num_input_tokens_seen": 18991130,
      "step": 894,
      "time_per_iteration": 2.5614848136901855
    },
    {
      "auxiliary_loss_clip": 0.01252668,
      "auxiliary_loss_mlp": 0.01051168,
      "balance_loss_clip": 1.07163644,
      "balance_loss_mlp": 1.03790641,
      "epoch": 0.10761738712198642,
      "flos": 21106066199040.0,
      "grad_norm": 2.4041544553142336,
      "language_loss": 0.75544989,
      "learning_rate": 3.9372205530014845e-06,
      "loss": 0.77848828,
      "num_input_tokens_seen": 19009610,
      "step": 895,
      "time_per_iteration": 2.501729965209961
    },
    {
      "auxiliary_loss_clip": 0.0124967,
      "auxiliary_loss_mlp": 0.01057419,
      "balance_loss_clip": 1.06933403,
      "balance_loss_mlp": 1.04577255,
      "epoch": 0.1077376300126255,
      "flos": 23766723348480.0,
      "grad_norm": 3.3340744936338242,
      "language_loss": 0.7168293,
      "learning_rate": 3.937026765981696e-06,
      "loss": 0.73990023,
      "num_input_tokens_seen": 19029680,
      "step": 896,
      "time_per_iteration": 2.5032219886779785
    },
    {
      "auxiliary_loss_clip": 0.01209457,
      "auxiliary_loss_mlp": 0.01048998,
      "balance_loss_clip": 1.06693029,
      "balance_loss_mlp": 1.03682637,
      "epoch": 0.1078578729032646,
      "flos": 20919581763840.0,
      "grad_norm": 1.944272017540537,
      "language_loss": 0.79521108,
      "learning_rate": 3.936832685116488e-06,
      "loss": 0.81779563,
      "num_input_tokens_seen": 19047775,
      "step": 897,
      "time_per_iteration": 2.5997555255889893
    },
    {
      "auxiliary_loss_clip": 0.01250602,
      "auxiliary_loss_mlp": 0.01050089,
      "balance_loss_clip": 1.0701344,
      "balance_loss_mlp": 1.03820968,
      "epoch": 0.10797811579390369,
      "flos": 14829886702080.0,
      "grad_norm": 2.2601872502838085,
      "language_loss": 0.90124583,
      "learning_rate": 3.936638310435301e-06,
      "loss": 0.92425281,
      "num_input_tokens_seen": 19065640,
      "step": 898,
      "time_per_iteration": 2.476712226867676
    },
    {
      "auxiliary_loss_clip": 0.01238517,
      "auxiliary_loss_mlp": 0.01041953,
      "balance_loss_clip": 1.06957984,
      "balance_loss_mlp": 1.02962112,
      "epoch": 0.10809835868454278,
      "flos": 19536985411200.0,
      "grad_norm": 1.8516766769738098,
      "language_loss": 0.81500447,
      "learning_rate": 3.936443641967623e-06,
      "loss": 0.83780915,
      "num_input_tokens_seen": 19084470,
      "step": 899,
      "time_per_iteration": 2.5221753120422363
    },
    {
      "auxiliary_loss_clip": 0.01220105,
      "auxiliary_loss_mlp": 0.01046999,
      "balance_loss_clip": 1.06801021,
      "balance_loss_mlp": 1.03463709,
      "epoch": 0.10821860157518187,
      "flos": 18442320480000.0,
      "grad_norm": 3.7749681671373114,
      "language_loss": 0.83078492,
      "learning_rate": 3.936248679742983e-06,
      "loss": 0.8534559,
      "num_input_tokens_seen": 19102965,
      "step": 900,
      "time_per_iteration": 3.4695215225219727
    },
    {
      "auxiliary_loss_clip": 0.01095446,
      "auxiliary_loss_mlp": 0.01046967,
      "balance_loss_clip": 1.03037715,
      "balance_loss_mlp": 1.0428896,
      "epoch": 0.10833884446582095,
      "flos": 49359468447360.0,
      "grad_norm": 1.059869429396583,
      "language_loss": 0.70163828,
      "learning_rate": 3.936053423790959e-06,
      "loss": 0.7230624,
      "num_input_tokens_seen": 19151285,
      "step": 901,
      "time_per_iteration": 2.9060444831848145
    },
    {
      "auxiliary_loss_clip": 0.01250976,
      "auxiliary_loss_mlp": 0.01053356,
      "balance_loss_clip": 1.07237124,
      "balance_loss_mlp": 1.04193604,
      "epoch": 0.10845908735646005,
      "flos": 20411912891520.0,
      "grad_norm": 1.9478997185895697,
      "language_loss": 0.77381372,
      "learning_rate": 3.935857874141168e-06,
      "loss": 0.79685712,
      "num_input_tokens_seen": 19170120,
      "step": 902,
      "time_per_iteration": 2.510826587677002
    },
    {
      "auxiliary_loss_clip": 0.01211105,
      "auxiliary_loss_mlp": 0.01037111,
      "balance_loss_clip": 1.06463456,
      "balance_loss_mlp": 1.02439713,
      "epoch": 0.10857933024709913,
      "flos": 14027750133120.0,
      "grad_norm": 2.0829838065274995,
      "language_loss": 0.83508182,
      "learning_rate": 3.935662030823279e-06,
      "loss": 0.85756397,
      "num_input_tokens_seen": 19186305,
      "step": 903,
      "time_per_iteration": 2.5632071495056152
    },
    {
      "auxiliary_loss_clip": 0.01232649,
      "auxiliary_loss_mlp": 0.01047161,
      "balance_loss_clip": 1.06551313,
      "balance_loss_mlp": 1.035604,
      "epoch": 0.10869957313773823,
      "flos": 13369004657280.0,
      "grad_norm": 2.153745957717676,
      "language_loss": 0.7304405,
      "learning_rate": 3.935465893866998e-06,
      "loss": 0.75323868,
      "num_input_tokens_seen": 19204530,
      "step": 904,
      "time_per_iteration": 2.532444953918457
    },
    {
      "auxiliary_loss_clip": 0.01221785,
      "auxiliary_loss_mlp": 0.01041092,
      "balance_loss_clip": 1.06830382,
      "balance_loss_mlp": 1.02881348,
      "epoch": 0.10881981602837733,
      "flos": 25807095509760.0,
      "grad_norm": 2.131759386209718,
      "language_loss": 0.80578297,
      "learning_rate": 3.935269463302079e-06,
      "loss": 0.8284117,
      "num_input_tokens_seen": 19222735,
      "step": 905,
      "time_per_iteration": 3.355724334716797
    },
    {
      "auxiliary_loss_clip": 0.0123828,
      "auxiliary_loss_mlp": 0.01044739,
      "balance_loss_clip": 1.06912005,
      "balance_loss_mlp": 1.03218603,
      "epoch": 0.10894005891901641,
      "flos": 20777555387520.0,
      "grad_norm": 1.9055713645301335,
      "language_loss": 0.76770175,
      "learning_rate": 3.935072739158322e-06,
      "loss": 0.79053187,
      "num_input_tokens_seen": 19242445,
      "step": 906,
      "time_per_iteration": 2.534024477005005
    },
    {
      "auxiliary_loss_clip": 0.01221895,
      "auxiliary_loss_mlp": 0.01046419,
      "balance_loss_clip": 1.0669359,
      "balance_loss_mlp": 1.03391957,
      "epoch": 0.10906030180965551,
      "flos": 26649883296000.0,
      "grad_norm": 1.666361891444486,
      "language_loss": 0.7970683,
      "learning_rate": 3.934875721465569e-06,
      "loss": 0.81975144,
      "num_input_tokens_seen": 19262865,
      "step": 907,
      "time_per_iteration": 3.5007591247558594
    },
    {
      "auxiliary_loss_clip": 0.0121226,
      "auxiliary_loss_mlp": 0.01039761,
      "balance_loss_clip": 1.06006896,
      "balance_loss_mlp": 1.0268625,
      "epoch": 0.10918054470029459,
      "flos": 36534402420480.0,
      "grad_norm": 2.2352780812368684,
      "language_loss": 0.72035009,
      "learning_rate": 3.9346784102537076e-06,
      "loss": 0.74287027,
      "num_input_tokens_seen": 19285000,
      "step": 908,
      "time_per_iteration": 3.4692440032958984
    },
    {
      "auxiliary_loss_clip": 0.01247544,
      "auxiliary_loss_mlp": 0.01036063,
      "balance_loss_clip": 1.06801498,
      "balance_loss_mlp": 1.02451789,
      "epoch": 0.10930078759093369,
      "flos": 21762549118080.0,
      "grad_norm": 2.152585757675597,
      "language_loss": 0.78707099,
      "learning_rate": 3.934480805552669e-06,
      "loss": 0.80990702,
      "num_input_tokens_seen": 19306010,
      "step": 909,
      "time_per_iteration": 2.4838972091674805
    },
    {
      "auxiliary_loss_clip": 0.01247914,
      "auxiliary_loss_mlp": 0.00766377,
      "balance_loss_clip": 1.06895864,
      "balance_loss_mlp": 1.00022078,
      "epoch": 0.10942103048157277,
      "flos": 22601781457920.0,
      "grad_norm": 2.239238319449174,
      "language_loss": 0.87944019,
      "learning_rate": 3.93428290739243e-06,
      "loss": 0.8995831,
      "num_input_tokens_seen": 19325380,
      "step": 910,
      "time_per_iteration": 2.55458927154541
    },
    {
      "auxiliary_loss_clip": 0.0121816,
      "auxiliary_loss_mlp": 0.01043011,
      "balance_loss_clip": 1.06523955,
      "balance_loss_mlp": 1.03058898,
      "epoch": 0.10954127337221187,
      "flos": 15045781397760.0,
      "grad_norm": 2.6231745476004176,
      "language_loss": 0.79772121,
      "learning_rate": 3.9340847158030125e-06,
      "loss": 0.82033288,
      "num_input_tokens_seen": 19338960,
      "step": 911,
      "time_per_iteration": 2.498889684677124
    },
    {
      "auxiliary_loss_clip": 0.01232069,
      "auxiliary_loss_mlp": 0.01049271,
      "balance_loss_clip": 1.06386662,
      "balance_loss_mlp": 1.03742158,
      "epoch": 0.10966151626285096,
      "flos": 21650974496640.0,
      "grad_norm": 1.792408265944816,
      "language_loss": 0.75346315,
      "learning_rate": 3.9338862308144814e-06,
      "loss": 0.77627653,
      "num_input_tokens_seen": 19357780,
      "step": 912,
      "time_per_iteration": 2.5177719593048096
    },
    {
      "auxiliary_loss_clip": 0.01246908,
      "auxiliary_loss_mlp": 0.01044201,
      "balance_loss_clip": 1.06724405,
      "balance_loss_mlp": 1.03180289,
      "epoch": 0.10978175915349005,
      "flos": 20121359777280.0,
      "grad_norm": 2.077854498788477,
      "language_loss": 0.84350652,
      "learning_rate": 3.933687452456946e-06,
      "loss": 0.86641765,
      "num_input_tokens_seen": 19377680,
      "step": 913,
      "time_per_iteration": 2.493380069732666
    },
    {
      "auxiliary_loss_clip": 0.01198801,
      "auxiliary_loss_mlp": 0.01042249,
      "balance_loss_clip": 1.05911553,
      "balance_loss_mlp": 1.02928543,
      "epoch": 0.10990200204412914,
      "flos": 20412667077120.0,
      "grad_norm": 5.516580276220256,
      "language_loss": 0.86608702,
      "learning_rate": 3.933488380760562e-06,
      "loss": 0.88849753,
      "num_input_tokens_seen": 19397040,
      "step": 914,
      "time_per_iteration": 2.588554859161377
    },
    {
      "auxiliary_loss_clip": 0.01246548,
      "auxiliary_loss_mlp": 0.00767264,
      "balance_loss_clip": 1.06695592,
      "balance_loss_mlp": 1.00021315,
      "epoch": 0.11002224493476823,
      "flos": 17530117660800.0,
      "grad_norm": 1.925260600445307,
      "language_loss": 0.87039983,
      "learning_rate": 3.9332890157555286e-06,
      "loss": 0.89053786,
      "num_input_tokens_seen": 19413975,
      "step": 915,
      "time_per_iteration": 2.531214952468872
    },
    {
      "auxiliary_loss_clip": 0.01222129,
      "auxiliary_loss_mlp": 0.0104416,
      "balance_loss_clip": 1.06563783,
      "balance_loss_mlp": 1.03213167,
      "epoch": 0.11014248782540732,
      "flos": 12203093099520.0,
      "grad_norm": 1.7937093139015061,
      "language_loss": 0.76440042,
      "learning_rate": 3.933089357472088e-06,
      "loss": 0.78706336,
      "num_input_tokens_seen": 19432005,
      "step": 916,
      "time_per_iteration": 2.5258312225341797
    },
    {
      "auxiliary_loss_clip": 0.01244837,
      "auxiliary_loss_mlp": 0.01040597,
      "balance_loss_clip": 1.06782269,
      "balance_loss_mlp": 1.02932549,
      "epoch": 0.11026273071604642,
      "flos": 22382977760640.0,
      "grad_norm": 2.740617401560644,
      "language_loss": 0.85856855,
      "learning_rate": 3.932889405940529e-06,
      "loss": 0.88142288,
      "num_input_tokens_seen": 19450100,
      "step": 917,
      "time_per_iteration": 2.4750916957855225
    },
    {
      "auxiliary_loss_clip": 0.01219534,
      "auxiliary_loss_mlp": 0.01045129,
      "balance_loss_clip": 1.0698576,
      "balance_loss_mlp": 1.03369641,
      "epoch": 0.1103829736066855,
      "flos": 19829046896640.0,
      "grad_norm": 2.430533758463773,
      "language_loss": 0.79831803,
      "learning_rate": 3.932689161191184e-06,
      "loss": 0.82096469,
      "num_input_tokens_seen": 19467805,
      "step": 918,
      "time_per_iteration": 2.524777889251709
    },
    {
      "auxiliary_loss_clip": 0.01229415,
      "auxiliary_loss_mlp": 0.01046226,
      "balance_loss_clip": 1.06445098,
      "balance_loss_mlp": 1.03371453,
      "epoch": 0.1105032164973246,
      "flos": 22669616292480.0,
      "grad_norm": 2.0472912761321327,
      "language_loss": 0.88172996,
      "learning_rate": 3.93248862325443e-06,
      "loss": 0.90448642,
      "num_input_tokens_seen": 19486710,
      "step": 919,
      "time_per_iteration": 2.501171112060547
    },
    {
      "auxiliary_loss_clip": 0.01133369,
      "auxiliary_loss_mlp": 0.01026136,
      "balance_loss_clip": 1.03769565,
      "balance_loss_mlp": 1.02213085,
      "epoch": 0.11062345938796368,
      "flos": 66483507876480.0,
      "grad_norm": 0.9466168740475717,
      "language_loss": 0.64495075,
      "learning_rate": 3.932287792160688e-06,
      "loss": 0.66654587,
      "num_input_tokens_seen": 19545170,
      "step": 920,
      "time_per_iteration": 2.9975528717041016
    },
    {
      "auxiliary_loss_clip": 0.01234142,
      "auxiliary_loss_mlp": 0.01043238,
      "balance_loss_clip": 1.06399441,
      "balance_loss_mlp": 1.02985084,
      "epoch": 0.11074370227860278,
      "flos": 21907771804800.0,
      "grad_norm": 2.3951888163336976,
      "language_loss": 0.80806577,
      "learning_rate": 3.932086667940424e-06,
      "loss": 0.83083957,
      "num_input_tokens_seen": 19561875,
      "step": 921,
      "time_per_iteration": 2.5301268100738525
    },
    {
      "auxiliary_loss_clip": 0.01230691,
      "auxiliary_loss_mlp": 0.00766549,
      "balance_loss_clip": 1.06736255,
      "balance_loss_mlp": 1.0002768,
      "epoch": 0.11086394516924186,
      "flos": 28658115763200.0,
      "grad_norm": 1.9402681870680323,
      "language_loss": 0.81970757,
      "learning_rate": 3.93188525062415e-06,
      "loss": 0.83967996,
      "num_input_tokens_seen": 19582340,
      "step": 922,
      "time_per_iteration": 2.5770580768585205
    },
    {
      "auxiliary_loss_clip": 0.01231217,
      "auxiliary_loss_mlp": 0.01052854,
      "balance_loss_clip": 1.06536674,
      "balance_loss_mlp": 1.04053402,
      "epoch": 0.11098418805988096,
      "flos": 24535247765760.0,
      "grad_norm": 1.997487286234216,
      "language_loss": 0.86130506,
      "learning_rate": 3.931683540242418e-06,
      "loss": 0.8841458,
      "num_input_tokens_seen": 19603405,
      "step": 923,
      "time_per_iteration": 2.62070369720459
    },
    {
      "auxiliary_loss_clip": 0.01225107,
      "auxiliary_loss_mlp": 0.01042776,
      "balance_loss_clip": 1.06392956,
      "balance_loss_mlp": 1.03000283,
      "epoch": 0.11110443095052006,
      "flos": 22960384888320.0,
      "grad_norm": 2.956341397776146,
      "language_loss": 0.9099561,
      "learning_rate": 3.9314815368258295e-06,
      "loss": 0.93263489,
      "num_input_tokens_seen": 19619885,
      "step": 924,
      "time_per_iteration": 2.5184521675109863
    },
    {
      "auxiliary_loss_clip": 0.01235169,
      "auxiliary_loss_mlp": 0.01036184,
      "balance_loss_clip": 1.07062948,
      "balance_loss_mlp": 1.02430487,
      "epoch": 0.11122467384115914,
      "flos": 18950025265920.0,
      "grad_norm": 1.7315162262529846,
      "language_loss": 0.78997731,
      "learning_rate": 3.9312792404050275e-06,
      "loss": 0.81269085,
      "num_input_tokens_seen": 19637940,
      "step": 925,
      "time_per_iteration": 2.5577445030212402
    },
    {
      "auxiliary_loss_clip": 0.01245269,
      "auxiliary_loss_mlp": 0.01042558,
      "balance_loss_clip": 1.0685674,
      "balance_loss_mlp": 1.03186452,
      "epoch": 0.11134491673179824,
      "flos": 25082957324160.0,
      "grad_norm": 2.9284512006801258,
      "language_loss": 0.77137518,
      "learning_rate": 3.9310766510107e-06,
      "loss": 0.79425347,
      "num_input_tokens_seen": 19657115,
      "step": 926,
      "time_per_iteration": 2.555898427963257
    },
    {
      "auxiliary_loss_clip": 0.01199394,
      "auxiliary_loss_mlp": 0.01046524,
      "balance_loss_clip": 1.05817652,
      "balance_loss_mlp": 1.03373289,
      "epoch": 0.11146515962243732,
      "flos": 24499121662080.0,
      "grad_norm": 1.7924464628446843,
      "language_loss": 0.92370033,
      "learning_rate": 3.9308737686735806e-06,
      "loss": 0.94615948,
      "num_input_tokens_seen": 19677075,
      "step": 927,
      "time_per_iteration": 2.6491332054138184
    },
    {
      "auxiliary_loss_clip": 0.0124858,
      "auxiliary_loss_mlp": 0.01048801,
      "balance_loss_clip": 1.0684545,
      "balance_loss_mlp": 1.03753579,
      "epoch": 0.11158540251307641,
      "flos": 22343763087360.0,
      "grad_norm": 2.0353876311859302,
      "language_loss": 0.82979387,
      "learning_rate": 3.9306705934244455e-06,
      "loss": 0.85276765,
      "num_input_tokens_seen": 19697155,
      "step": 928,
      "time_per_iteration": 3.3209948539733887
    },
    {
      "auxiliary_loss_clip": 0.01205064,
      "auxiliary_loss_mlp": 0.01036856,
      "balance_loss_clip": 1.06205869,
      "balance_loss_mlp": 1.02527523,
      "epoch": 0.11170564540371551,
      "flos": 19902304684800.0,
      "grad_norm": 1.7112590411728725,
      "language_loss": 0.88181931,
      "learning_rate": 3.930467125294116e-06,
      "loss": 0.90423858,
      "num_input_tokens_seen": 19716705,
      "step": 929,
      "time_per_iteration": 2.533008337020874
    },
    {
      "auxiliary_loss_clip": 0.01068866,
      "auxiliary_loss_mlp": 0.01006282,
      "balance_loss_clip": 1.0246582,
      "balance_loss_mlp": 1.00311112,
      "epoch": 0.1118258882943546,
      "flos": 64586239499520.0,
      "grad_norm": 0.9239433802578965,
      "language_loss": 0.6047616,
      "learning_rate": 3.930263364313458e-06,
      "loss": 0.62551308,
      "num_input_tokens_seen": 19767275,
      "step": 930,
      "time_per_iteration": 3.112867832183838
    },
    {
      "auxiliary_loss_clip": 0.0119898,
      "auxiliary_loss_mlp": 0.01052218,
      "balance_loss_clip": 1.06153238,
      "balance_loss_mlp": 1.03958821,
      "epoch": 0.11194613118499369,
      "flos": 17201965985280.0,
      "grad_norm": 2.0271495757930964,
      "language_loss": 0.82767642,
      "learning_rate": 3.930059310513384e-06,
      "loss": 0.85018837,
      "num_input_tokens_seen": 19786315,
      "step": 931,
      "time_per_iteration": 3.5619053840637207
    },
    {
      "auxiliary_loss_clip": 0.01183773,
      "auxiliary_loss_mlp": 0.00766056,
      "balance_loss_clip": 1.05905557,
      "balance_loss_mlp": 1.0001719,
      "epoch": 0.11206637407563277,
      "flos": 31863465728640.0,
      "grad_norm": 1.8536881824331894,
      "language_loss": 0.839719,
      "learning_rate": 3.929854963924846e-06,
      "loss": 0.85921729,
      "num_input_tokens_seen": 19806580,
      "step": 932,
      "time_per_iteration": 2.6600515842437744
    },
    {
      "auxiliary_loss_clip": 0.01201518,
      "auxiliary_loss_mlp": 0.01037229,
      "balance_loss_clip": 1.06075644,
      "balance_loss_mlp": 1.02571964,
      "epoch": 0.11218661696627187,
      "flos": 21945621761280.0,
      "grad_norm": 1.9242009350584048,
      "language_loss": 0.77293712,
      "learning_rate": 3.929650324578845e-06,
      "loss": 0.79532456,
      "num_input_tokens_seen": 19826045,
      "step": 933,
      "time_per_iteration": 2.5921401977539062
    },
    {
      "auxiliary_loss_clip": 0.01220124,
      "auxiliary_loss_mlp": 0.01043782,
      "balance_loss_clip": 1.06360102,
      "balance_loss_mlp": 1.03066325,
      "epoch": 0.11230685985691095,
      "flos": 25878198481920.0,
      "grad_norm": 2.682825909780638,
      "language_loss": 0.81794131,
      "learning_rate": 3.929445392506423e-06,
      "loss": 0.84058028,
      "num_input_tokens_seen": 19843985,
      "step": 934,
      "time_per_iteration": 3.467923641204834
    },
    {
      "auxiliary_loss_clip": 0.01231093,
      "auxiliary_loss_mlp": 0.01046508,
      "balance_loss_clip": 1.06905484,
      "balance_loss_mlp": 1.03536761,
      "epoch": 0.11242710274755005,
      "flos": 22231506107520.0,
      "grad_norm": 1.9855036795577263,
      "language_loss": 0.7621538,
      "learning_rate": 3.92924016773867e-06,
      "loss": 0.78492975,
      "num_input_tokens_seen": 19860480,
      "step": 935,
      "time_per_iteration": 3.287496328353882
    },
    {
      "auxiliary_loss_clip": 0.01216609,
      "auxiliary_loss_mlp": 0.00765709,
      "balance_loss_clip": 1.06162333,
      "balance_loss_mlp": 1.00018072,
      "epoch": 0.11254734563818915,
      "flos": 17712184723200.0,
      "grad_norm": 2.498503811494862,
      "language_loss": 0.73961407,
      "learning_rate": 3.9290346503067175e-06,
      "loss": 0.75943726,
      "num_input_tokens_seen": 19877145,
      "step": 936,
      "time_per_iteration": 2.5347349643707275
    },
    {
      "auxiliary_loss_clip": 0.0123198,
      "auxiliary_loss_mlp": 0.0104414,
      "balance_loss_clip": 1.0633018,
      "balance_loss_mlp": 1.03251135,
      "epoch": 0.11266758852882823,
      "flos": 54930397334400.0,
      "grad_norm": 1.879736759513515,
      "language_loss": 0.78577709,
      "learning_rate": 3.9288288402417415e-06,
      "loss": 0.80853832,
      "num_input_tokens_seen": 19903405,
      "step": 937,
      "time_per_iteration": 2.8518640995025635
    },
    {
      "auxiliary_loss_clip": 0.01235733,
      "auxiliary_loss_mlp": 0.01041264,
      "balance_loss_clip": 1.06848824,
      "balance_loss_mlp": 1.02865195,
      "epoch": 0.11278783141946733,
      "flos": 18878132194560.0,
      "grad_norm": 2.4776593630824952,
      "language_loss": 0.7051841,
      "learning_rate": 3.928622737574964e-06,
      "loss": 0.72795415,
      "num_input_tokens_seen": 19918740,
      "step": 938,
      "time_per_iteration": 2.540126085281372
    },
    {
      "auxiliary_loss_clip": 0.01213743,
      "auxiliary_loss_mlp": 0.01045296,
      "balance_loss_clip": 1.06124425,
      "balance_loss_mlp": 1.03332162,
      "epoch": 0.11290807431010641,
      "flos": 26469252777600.0,
      "grad_norm": 2.185617239783044,
      "language_loss": 0.90913594,
      "learning_rate": 3.928416342337652e-06,
      "loss": 0.93172634,
      "num_input_tokens_seen": 19938475,
      "step": 939,
      "time_per_iteration": 2.5731821060180664
    },
    {
      "auxiliary_loss_clip": 0.01217515,
      "auxiliary_loss_mlp": 0.01042492,
      "balance_loss_clip": 1.06504345,
      "balance_loss_mlp": 1.03062451,
      "epoch": 0.1130283172007455,
      "flos": 22710590732160.0,
      "grad_norm": 1.709423476168821,
      "language_loss": 0.82656705,
      "learning_rate": 3.928209654561113e-06,
      "loss": 0.84916717,
      "num_input_tokens_seen": 19959310,
      "step": 940,
      "time_per_iteration": 2.55037522315979
    },
    {
      "auxiliary_loss_clip": 0.01208729,
      "auxiliary_loss_mlp": 0.01042085,
      "balance_loss_clip": 1.06423962,
      "balance_loss_mlp": 1.0308553,
      "epoch": 0.1131485600913846,
      "flos": 23219911630080.0,
      "grad_norm": 2.2333480538657104,
      "language_loss": 0.81722045,
      "learning_rate": 3.928002674276703e-06,
      "loss": 0.83972859,
      "num_input_tokens_seen": 19978700,
      "step": 941,
      "time_per_iteration": 2.5602898597717285
    },
    {
      "auxiliary_loss_clip": 0.01165758,
      "auxiliary_loss_mlp": 0.01038979,
      "balance_loss_clip": 1.0541929,
      "balance_loss_mlp": 1.0266645,
      "epoch": 0.11326880298202369,
      "flos": 14064271286400.0,
      "grad_norm": 2.308935222181706,
      "language_loss": 0.75160742,
      "learning_rate": 3.92779540151582e-06,
      "loss": 0.77365482,
      "num_input_tokens_seen": 19995785,
      "step": 942,
      "time_per_iteration": 2.5651979446411133
    },
    {
      "auxiliary_loss_clip": 0.01213113,
      "auxiliary_loss_mlp": 0.01033353,
      "balance_loss_clip": 1.06258774,
      "balance_loss_mlp": 1.02242732,
      "epoch": 0.11338904587266278,
      "flos": 16325386479360.0,
      "grad_norm": 3.5205517173683547,
      "language_loss": 0.8551293,
      "learning_rate": 3.927587836309907e-06,
      "loss": 0.87759399,
      "num_input_tokens_seen": 20013615,
      "step": 943,
      "time_per_iteration": 2.5361077785491943
    },
    {
      "auxiliary_loss_clip": 0.01210224,
      "auxiliary_loss_mlp": 0.0104377,
      "balance_loss_clip": 1.06141591,
      "balance_loss_mlp": 1.03237987,
      "epoch": 0.11350928876330187,
      "flos": 24426258923520.0,
      "grad_norm": 2.2304199453775584,
      "language_loss": 0.78512174,
      "learning_rate": 3.927379978690452e-06,
      "loss": 0.80766165,
      "num_input_tokens_seen": 20032880,
      "step": 944,
      "time_per_iteration": 2.5932183265686035
    },
    {
      "auxiliary_loss_clip": 0.01185526,
      "auxiliary_loss_mlp": 0.01046516,
      "balance_loss_clip": 1.05306458,
      "balance_loss_mlp": 1.03479791,
      "epoch": 0.11362953165394096,
      "flos": 24497074586880.0,
      "grad_norm": 4.6681776614956965,
      "language_loss": 0.87398177,
      "learning_rate": 3.927171828688987e-06,
      "loss": 0.89630222,
      "num_input_tokens_seen": 20052405,
      "step": 945,
      "time_per_iteration": 2.6163206100463867
    },
    {
      "auxiliary_loss_clip": 0.01249499,
      "auxiliary_loss_mlp": 0.01037155,
      "balance_loss_clip": 1.07157063,
      "balance_loss_mlp": 1.02580023,
      "epoch": 0.11374977454458005,
      "flos": 24060831909120.0,
      "grad_norm": 4.964377312919228,
      "language_loss": 0.82049829,
      "learning_rate": 3.926963386337088e-06,
      "loss": 0.84336483,
      "num_input_tokens_seen": 20070635,
      "step": 946,
      "time_per_iteration": 2.511021614074707
    },
    {
      "auxiliary_loss_clip": 0.01251038,
      "auxiliary_loss_mlp": 0.01040158,
      "balance_loss_clip": 1.06869376,
      "balance_loss_mlp": 1.026896,
      "epoch": 0.11387001743521914,
      "flos": 39457638967680.0,
      "grad_norm": 2.477697303273926,
      "language_loss": 0.70211494,
      "learning_rate": 3.926754651666375e-06,
      "loss": 0.72502685,
      "num_input_tokens_seen": 20091195,
      "step": 947,
      "time_per_iteration": 2.6101086139678955
    },
    {
      "auxiliary_loss_clip": 0.01201567,
      "auxiliary_loss_mlp": 0.01045759,
      "balance_loss_clip": 1.06499004,
      "balance_loss_mlp": 1.03382611,
      "epoch": 0.11399026032585824,
      "flos": 25082454533760.0,
      "grad_norm": 2.5759321953506773,
      "language_loss": 0.78511375,
      "learning_rate": 3.926545624708513e-06,
      "loss": 0.80758703,
      "num_input_tokens_seen": 20110435,
      "step": 948,
      "time_per_iteration": 2.662794351577759
    },
    {
      "auxiliary_loss_clip": 0.01196644,
      "auxiliary_loss_mlp": 0.01048158,
      "balance_loss_clip": 1.06100798,
      "balance_loss_mlp": 1.0364933,
      "epoch": 0.11411050321649732,
      "flos": 17961835224960.0,
      "grad_norm": 2.2317643758693615,
      "language_loss": 0.8537426,
      "learning_rate": 3.926336305495213e-06,
      "loss": 0.87619054,
      "num_input_tokens_seen": 20128995,
      "step": 949,
      "time_per_iteration": 2.5853984355926514
    },
    {
      "auxiliary_loss_clip": 0.01187161,
      "auxiliary_loss_mlp": 0.01044806,
      "balance_loss_clip": 1.06098485,
      "balance_loss_mlp": 1.03126359,
      "epoch": 0.11423074610713642,
      "flos": 22455409536000.0,
      "grad_norm": 2.057630330366888,
      "language_loss": 0.88978767,
      "learning_rate": 3.926126694058226e-06,
      "loss": 0.91210735,
      "num_input_tokens_seen": 20148145,
      "step": 950,
      "time_per_iteration": 2.6092495918273926
    },
    {
      "auxiliary_loss_clip": 0.01183678,
      "auxiliary_loss_mlp": 0.01044353,
      "balance_loss_clip": 1.06493485,
      "balance_loss_mlp": 1.03386319,
      "epoch": 0.1143509889977755,
      "flos": 19717687756800.0,
      "grad_norm": 1.4908712306820662,
      "language_loss": 0.82143223,
      "learning_rate": 3.92591679042935e-06,
      "loss": 0.84371257,
      "num_input_tokens_seen": 20168035,
      "step": 951,
      "time_per_iteration": 2.6263771057128906
    },
    {
      "auxiliary_loss_clip": 0.012316,
      "auxiliary_loss_mlp": 0.01043318,
      "balance_loss_clip": 1.06818664,
      "balance_loss_mlp": 1.03068161,
      "epoch": 0.1144712318884146,
      "flos": 19822869757440.0,
      "grad_norm": 1.9576754472941424,
      "language_loss": 0.82164466,
      "learning_rate": 3.92570659464043e-06,
      "loss": 0.84439385,
      "num_input_tokens_seen": 20186095,
      "step": 952,
      "time_per_iteration": 2.5014209747314453
    },
    {
      "auxiliary_loss_clip": 0.01228435,
      "auxiliary_loss_mlp": 0.00766193,
      "balance_loss_clip": 1.06834412,
      "balance_loss_mlp": 1.00021887,
      "epoch": 0.1145914747790537,
      "flos": 14939198766720.0,
      "grad_norm": 2.068305216321503,
      "language_loss": 0.79911947,
      "learning_rate": 3.925496106723349e-06,
      "loss": 0.81906575,
      "num_input_tokens_seen": 20203535,
      "step": 953,
      "time_per_iteration": 2.4958319664001465
    },
    {
      "auxiliary_loss_clip": 0.01234318,
      "auxiliary_loss_mlp": 0.01040557,
      "balance_loss_clip": 1.06794047,
      "balance_loss_mlp": 1.02936339,
      "epoch": 0.11471171766969278,
      "flos": 19865029345920.0,
      "grad_norm": 2.1686564435671607,
      "language_loss": 0.83851522,
      "learning_rate": 3.9252853267100405e-06,
      "loss": 0.86126399,
      "num_input_tokens_seen": 20222780,
      "step": 954,
      "time_per_iteration": 2.5247936248779297
    },
    {
      "auxiliary_loss_clip": 0.01189794,
      "auxiliary_loss_mlp": 0.01044997,
      "balance_loss_clip": 1.06037807,
      "balance_loss_mlp": 1.03305256,
      "epoch": 0.11483196056033187,
      "flos": 22526476594560.0,
      "grad_norm": 4.389400191734664,
      "language_loss": 0.83695138,
      "learning_rate": 3.9250742546324786e-06,
      "loss": 0.8592993,
      "num_input_tokens_seen": 20243015,
      "step": 955,
      "time_per_iteration": 3.384770631790161
    },
    {
      "auxiliary_loss_clip": 0.0121383,
      "auxiliary_loss_mlp": 0.01043415,
      "balance_loss_clip": 1.0634445,
      "balance_loss_mlp": 1.03280544,
      "epoch": 0.11495220345097096,
      "flos": 28220292887040.0,
      "grad_norm": 5.98804414515056,
      "language_loss": 0.87021178,
      "learning_rate": 3.924862890522683e-06,
      "loss": 0.89278424,
      "num_input_tokens_seen": 20263025,
      "step": 956,
      "time_per_iteration": 2.6026947498321533
    },
    {
      "auxiliary_loss_clip": 0.0123029,
      "auxiliary_loss_mlp": 0.01041209,
      "balance_loss_clip": 1.06451845,
      "balance_loss_mlp": 1.02943146,
      "epoch": 0.11507244634161005,
      "flos": 17492267704320.0,
      "grad_norm": 2.104482533949446,
      "language_loss": 0.85904467,
      "learning_rate": 3.9246512344127174e-06,
      "loss": 0.88175964,
      "num_input_tokens_seen": 20280685,
      "step": 957,
      "time_per_iteration": 2.5276901721954346
    },
    {
      "auxiliary_loss_clip": 0.01149633,
      "auxiliary_loss_mlp": 0.01039141,
      "balance_loss_clip": 1.05571187,
      "balance_loss_mlp": 1.02764297,
      "epoch": 0.11519268923224914,
      "flos": 22564937082240.0,
      "grad_norm": 3.0797442543689293,
      "language_loss": 0.81831956,
      "learning_rate": 3.9244392863346895e-06,
      "loss": 0.84020722,
      "num_input_tokens_seen": 20300090,
      "step": 958,
      "time_per_iteration": 3.444572925567627
    },
    {
      "auxiliary_loss_clip": 0.01220195,
      "auxiliary_loss_mlp": 0.01046949,
      "balance_loss_clip": 1.06974697,
      "balance_loss_mlp": 1.03428292,
      "epoch": 0.11531293212288823,
      "flos": 16982839065600.0,
      "grad_norm": 1.9642476485491807,
      "language_loss": 0.92335945,
      "learning_rate": 3.9242270463207524e-06,
      "loss": 0.94603091,
      "num_input_tokens_seen": 20318480,
      "step": 959,
      "time_per_iteration": 2.530734062194824
    },
    {
      "auxiliary_loss_clip": 0.01170837,
      "auxiliary_loss_mlp": 0.01041607,
      "balance_loss_clip": 1.05797625,
      "balance_loss_mlp": 1.0292151,
      "epoch": 0.11543317501352733,
      "flos": 12422004537600.0,
      "grad_norm": 5.7963241261524105,
      "language_loss": 0.85763395,
      "learning_rate": 3.924014514403102e-06,
      "loss": 0.87975836,
      "num_input_tokens_seen": 20334635,
      "step": 960,
      "time_per_iteration": 2.582185745239258
    },
    {
      "auxiliary_loss_clip": 0.01173004,
      "auxiliary_loss_mlp": 0.01047938,
      "balance_loss_clip": 1.05811596,
      "balance_loss_mlp": 1.03498578,
      "epoch": 0.11555341790416641,
      "flos": 19821648695040.0,
      "grad_norm": 2.0216681726805295,
      "language_loss": 0.91281712,
      "learning_rate": 3.92380169061398e-06,
      "loss": 0.93502653,
      "num_input_tokens_seen": 20352415,
      "step": 961,
      "time_per_iteration": 3.510355234146118
    },
    {
      "auxiliary_loss_clip": 0.01190988,
      "auxiliary_loss_mlp": 0.00766267,
      "balance_loss_clip": 1.05787051,
      "balance_loss_mlp": 1.00016916,
      "epoch": 0.11567366079480551,
      "flos": 25738865625600.0,
      "grad_norm": 4.168283357649946,
      "language_loss": 0.83740705,
      "learning_rate": 3.9235885749856705e-06,
      "loss": 0.85697961,
      "num_input_tokens_seen": 20371095,
      "step": 962,
      "time_per_iteration": 3.366799831390381
    },
    {
      "auxiliary_loss_clip": 0.01221592,
      "auxiliary_loss_mlp": 0.0104391,
      "balance_loss_clip": 1.07287169,
      "balance_loss_mlp": 1.03172088,
      "epoch": 0.1157939036854446,
      "flos": 18223301301120.0,
      "grad_norm": 2.3843259083233894,
      "language_loss": 0.82769305,
      "learning_rate": 3.9233751675505035e-06,
      "loss": 0.85034811,
      "num_input_tokens_seen": 20389805,
      "step": 963,
      "time_per_iteration": 2.538644790649414
    },
    {
      "auxiliary_loss_clip": 0.01211749,
      "auxiliary_loss_mlp": 0.01041751,
      "balance_loss_clip": 1.0667789,
      "balance_loss_mlp": 1.02849495,
      "epoch": 0.11591414657608369,
      "flos": 23073755189760.0,
      "grad_norm": 1.8655173095163404,
      "language_loss": 0.84628808,
      "learning_rate": 3.923161468340853e-06,
      "loss": 0.86882311,
      "num_input_tokens_seen": 20409640,
      "step": 964,
      "time_per_iteration": 2.5701324939727783
    },
    {
      "auxiliary_loss_clip": 0.01170812,
      "auxiliary_loss_mlp": 0.01041327,
      "balance_loss_clip": 1.05627096,
      "balance_loss_mlp": 1.02952552,
      "epoch": 0.11603438946672277,
      "flos": 19461716461440.0,
      "grad_norm": 1.6981682620451493,
      "language_loss": 0.81574696,
      "learning_rate": 3.9229474773891374e-06,
      "loss": 0.83786833,
      "num_input_tokens_seen": 20428180,
      "step": 965,
      "time_per_iteration": 2.60032320022583
    },
    {
      "auxiliary_loss_clip": 0.01205725,
      "auxiliary_loss_mlp": 0.01050564,
      "balance_loss_clip": 1.05884755,
      "balance_loss_mlp": 1.03767192,
      "epoch": 0.11615463235736187,
      "flos": 26831986272000.0,
      "grad_norm": 3.0569375799848006,
      "language_loss": 0.83467007,
      "learning_rate": 3.922733194727818e-06,
      "loss": 0.85723293,
      "num_input_tokens_seen": 20447975,
      "step": 966,
      "time_per_iteration": 2.639197826385498
    },
    {
      "auxiliary_loss_clip": 0.01236906,
      "auxiliary_loss_mlp": 0.01042939,
      "balance_loss_clip": 1.06900549,
      "balance_loss_mlp": 1.03072011,
      "epoch": 0.11627487524800097,
      "flos": 18580324533120.0,
      "grad_norm": 2.3384779580136685,
      "language_loss": 0.8749249,
      "learning_rate": 3.922518620389402e-06,
      "loss": 0.89772332,
      "num_input_tokens_seen": 20464840,
      "step": 967,
      "time_per_iteration": 2.4678986072540283
    },
    {
      "auxiliary_loss_clip": 0.01123668,
      "auxiliary_loss_mlp": 0.01039579,
      "balance_loss_clip": 1.05272841,
      "balance_loss_mlp": 1.02700853,
      "epoch": 0.11639511813864005,
      "flos": 18150474476160.0,
      "grad_norm": 3.5046784383285283,
      "language_loss": 0.89498007,
      "learning_rate": 3.922303754406439e-06,
      "loss": 0.91661251,
      "num_input_tokens_seen": 20482680,
      "step": 968,
      "time_per_iteration": 2.648559331893921
    },
    {
      "auxiliary_loss_clip": 0.01181333,
      "auxiliary_loss_mlp": 0.01049982,
      "balance_loss_clip": 1.05794382,
      "balance_loss_mlp": 1.03638005,
      "epoch": 0.11651536102927915,
      "flos": 20922023888640.0,
      "grad_norm": 2.4029588315319392,
      "language_loss": 0.79096127,
      "learning_rate": 3.922088596811526e-06,
      "loss": 0.81327444,
      "num_input_tokens_seen": 20501810,
      "step": 969,
      "time_per_iteration": 2.5997016429901123
    },
    {
      "auxiliary_loss_clip": 0.01216686,
      "auxiliary_loss_mlp": 0.01041144,
      "balance_loss_clip": 1.06253982,
      "balance_loss_mlp": 1.02989638,
      "epoch": 0.11663560391991823,
      "flos": 16508602776960.0,
      "grad_norm": 2.2351264901267016,
      "language_loss": 0.86859179,
      "learning_rate": 3.9218731476373e-06,
      "loss": 0.89117014,
      "num_input_tokens_seen": 20517995,
      "step": 970,
      "time_per_iteration": 2.4892120361328125
    },
    {
      "auxiliary_loss_clip": 0.01239136,
      "auxiliary_loss_mlp": 0.01048607,
      "balance_loss_clip": 1.07091939,
      "balance_loss_mlp": 1.03542817,
      "epoch": 0.11675584681055733,
      "flos": 19865029345920.0,
      "grad_norm": 2.091045713161438,
      "language_loss": 0.84533858,
      "learning_rate": 3.9216574069164455e-06,
      "loss": 0.86821598,
      "num_input_tokens_seen": 20536970,
      "step": 971,
      "time_per_iteration": 2.4939916133880615
    },
    {
      "auxiliary_loss_clip": 0.01242973,
      "auxiliary_loss_mlp": 0.01041652,
      "balance_loss_clip": 1.06727505,
      "balance_loss_mlp": 1.03064299,
      "epoch": 0.11687608970119642,
      "flos": 21944364785280.0,
      "grad_norm": 1.5287384535308486,
      "language_loss": 0.79943645,
      "learning_rate": 3.921441374681691e-06,
      "loss": 0.82228267,
      "num_input_tokens_seen": 20557030,
      "step": 972,
      "time_per_iteration": 2.519932508468628
    },
    {
      "auxiliary_loss_clip": 0.0121033,
      "auxiliary_loss_mlp": 0.01038171,
      "balance_loss_clip": 1.06396246,
      "balance_loss_mlp": 1.0262444,
      "epoch": 0.1169963325918355,
      "flos": 24061155131520.0,
      "grad_norm": 2.1838902758807968,
      "language_loss": 0.65210819,
      "learning_rate": 3.921225050965808e-06,
      "loss": 0.67459321,
      "num_input_tokens_seen": 20576915,
      "step": 973,
      "time_per_iteration": 2.5642313957214355
    },
    {
      "auxiliary_loss_clip": 0.01196285,
      "auxiliary_loss_mlp": 0.01040274,
      "balance_loss_clip": 1.06056392,
      "balance_loss_mlp": 1.02817452,
      "epoch": 0.1171165754824746,
      "flos": 23368151059200.0,
      "grad_norm": 2.168247555973414,
      "language_loss": 0.74746329,
      "learning_rate": 3.921008435801612e-06,
      "loss": 0.76982886,
      "num_input_tokens_seen": 20596000,
      "step": 974,
      "time_per_iteration": 2.5717360973358154
    },
    {
      "auxiliary_loss_clip": 0.01217838,
      "auxiliary_loss_mlp": 0.01040911,
      "balance_loss_clip": 1.06437182,
      "balance_loss_mlp": 1.02840626,
      "epoch": 0.11723681837311369,
      "flos": 18552243075840.0,
      "grad_norm": 3.907495793045152,
      "language_loss": 0.75140554,
      "learning_rate": 3.920791529221963e-06,
      "loss": 0.77399302,
      "num_input_tokens_seen": 20614675,
      "step": 975,
      "time_per_iteration": 2.504629135131836
    },
    {
      "auxiliary_loss_clip": 0.01218052,
      "auxiliary_loss_mlp": 0.00766203,
      "balance_loss_clip": 1.06507802,
      "balance_loss_mlp": 1.00019956,
      "epoch": 0.11735706126375278,
      "flos": 23550541344000.0,
      "grad_norm": 2.22358341578932,
      "language_loss": 0.7644068,
      "learning_rate": 3.920574331259768e-06,
      "loss": 0.78424937,
      "num_input_tokens_seen": 20635875,
      "step": 976,
      "time_per_iteration": 2.600533962249756
    },
    {
      "auxiliary_loss_clip": 0.01203751,
      "auxiliary_loss_mlp": 0.01039368,
      "balance_loss_clip": 1.0616771,
      "balance_loss_mlp": 1.02849603,
      "epoch": 0.11747730415439187,
      "flos": 22381541216640.0,
      "grad_norm": 2.478980233029111,
      "language_loss": 0.79192472,
      "learning_rate": 3.9203568419479716e-06,
      "loss": 0.81435591,
      "num_input_tokens_seen": 20656430,
      "step": 977,
      "time_per_iteration": 2.5490169525146484
    },
    {
      "auxiliary_loss_clip": 0.01213339,
      "auxiliary_loss_mlp": 0.01031249,
      "balance_loss_clip": 1.06437027,
      "balance_loss_mlp": 1.02024031,
      "epoch": 0.11759754704503096,
      "flos": 22200731130240.0,
      "grad_norm": 1.95727999590609,
      "language_loss": 0.75444448,
      "learning_rate": 3.92013906131957e-06,
      "loss": 0.77689034,
      "num_input_tokens_seen": 20675360,
      "step": 978,
      "time_per_iteration": 2.5403544902801514
    },
    {
      "auxiliary_loss_clip": 0.01197648,
      "auxiliary_loss_mlp": 0.01051672,
      "balance_loss_clip": 1.06457043,
      "balance_loss_mlp": 1.04108584,
      "epoch": 0.11771778993567006,
      "flos": 22309755886080.0,
      "grad_norm": 1.6120511554181212,
      "language_loss": 0.8221814,
      "learning_rate": 3.9199209894076e-06,
      "loss": 0.84467459,
      "num_input_tokens_seen": 20695675,
      "step": 979,
      "time_per_iteration": 2.5954396724700928
    },
    {
      "auxiliary_loss_clip": 0.01246823,
      "auxiliary_loss_mlp": 0.01036814,
      "balance_loss_clip": 1.06715775,
      "balance_loss_mlp": 1.02373064,
      "epoch": 0.11783803282630914,
      "flos": 21288169175040.0,
      "grad_norm": 1.808185574917874,
      "language_loss": 0.89461297,
      "learning_rate": 3.919702626245142e-06,
      "loss": 0.91744941,
      "num_input_tokens_seen": 20715330,
      "step": 980,
      "time_per_iteration": 2.491637706756592
    },
    {
      "auxiliary_loss_clip": 0.01199504,
      "auxiliary_loss_mlp": 0.0103896,
      "balance_loss_clip": 1.06062829,
      "balance_loss_mlp": 1.02705121,
      "epoch": 0.11795827571694824,
      "flos": 25371535190400.0,
      "grad_norm": 2.1781573558250193,
      "language_loss": 0.6653322,
      "learning_rate": 3.919483971865322e-06,
      "loss": 0.68771684,
      "num_input_tokens_seen": 20735325,
      "step": 981,
      "time_per_iteration": 3.409590482711792
    },
    {
      "auxiliary_loss_clip": 0.01212901,
      "auxiliary_loss_mlp": 0.0103667,
      "balance_loss_clip": 1.06766129,
      "balance_loss_mlp": 1.02573252,
      "epoch": 0.11807851860758732,
      "flos": 23622218933760.0,
      "grad_norm": 2.1999078750208447,
      "language_loss": 0.88020611,
      "learning_rate": 3.91926502630131e-06,
      "loss": 0.9027018,
      "num_input_tokens_seen": 20755940,
      "step": 982,
      "time_per_iteration": 2.5646488666534424
    },
    {
      "auxiliary_loss_clip": 0.01234963,
      "auxiliary_loss_mlp": 0.01041378,
      "balance_loss_clip": 1.07178938,
      "balance_loss_mlp": 1.0300653,
      "epoch": 0.11819876149822642,
      "flos": 24972496024320.0,
      "grad_norm": 1.8677443841249768,
      "language_loss": 0.72238576,
      "learning_rate": 3.91904578958632e-06,
      "loss": 0.74514914,
      "num_input_tokens_seen": 20775355,
      "step": 983,
      "time_per_iteration": 2.537376642227173
    },
    {
      "auxiliary_loss_clip": 0.01248471,
      "auxiliary_loss_mlp": 0.01043772,
      "balance_loss_clip": 1.06952524,
      "balance_loss_mlp": 1.03206599,
      "epoch": 0.11831900438886551,
      "flos": 23003226835200.0,
      "grad_norm": 4.291269702588566,
      "language_loss": 0.83957946,
      "learning_rate": 3.918826261753608e-06,
      "loss": 0.86250186,
      "num_input_tokens_seen": 20794935,
      "step": 984,
      "time_per_iteration": 2.5083096027374268
    },
    {
      "auxiliary_loss_clip": 0.01212761,
      "auxiliary_loss_mlp": 0.01033618,
      "balance_loss_clip": 1.06468964,
      "balance_loss_mlp": 1.0236696,
      "epoch": 0.1184392472795046,
      "flos": 27965147604480.0,
      "grad_norm": 2.5894371408203622,
      "language_loss": 0.70987403,
      "learning_rate": 3.918606442836478e-06,
      "loss": 0.73233777,
      "num_input_tokens_seen": 20817155,
      "step": 985,
      "time_per_iteration": 3.4155235290527344
    },
    {
      "auxiliary_loss_clip": 0.01228825,
      "auxiliary_loss_mlp": 0.01037365,
      "balance_loss_clip": 1.07058084,
      "balance_loss_mlp": 1.02657628,
      "epoch": 0.1185594901701437,
      "flos": 19898497843200.0,
      "grad_norm": 1.841555434564848,
      "language_loss": 0.77150846,
      "learning_rate": 3.918386332868277e-06,
      "loss": 0.79417026,
      "num_input_tokens_seen": 20835125,
      "step": 986,
      "time_per_iteration": 2.5039713382720947
    },
    {
      "auxiliary_loss_clip": 0.01218375,
      "auxiliary_loss_mlp": 0.01045865,
      "balance_loss_clip": 1.06409431,
      "balance_loss_mlp": 1.03445721,
      "epoch": 0.11867973306078278,
      "flos": 18912354877440.0,
      "grad_norm": 1.845581159073504,
      "language_loss": 0.94383878,
      "learning_rate": 3.918165931882394e-06,
      "loss": 0.96648121,
      "num_input_tokens_seen": 20853525,
      "step": 987,
      "time_per_iteration": 2.5112380981445312
    },
    {
      "auxiliary_loss_clip": 0.0115319,
      "auxiliary_loss_mlp": 0.01038497,
      "balance_loss_clip": 1.05210078,
      "balance_loss_mlp": 1.02598,
      "epoch": 0.11879997595142187,
      "flos": 16982803152000.0,
      "grad_norm": 3.3857860573184153,
      "language_loss": 0.75145078,
      "learning_rate": 3.917945239912264e-06,
      "loss": 0.7733677,
      "num_input_tokens_seen": 20871000,
      "step": 988,
      "time_per_iteration": 4.2755584716796875
    },
    {
      "auxiliary_loss_clip": 0.01178952,
      "auxiliary_loss_mlp": 0.0104041,
      "balance_loss_clip": 1.05881488,
      "balance_loss_mlp": 1.03004456,
      "epoch": 0.11892021884206096,
      "flos": 17530369056000.0,
      "grad_norm": 1.9657056685116383,
      "language_loss": 0.75387728,
      "learning_rate": 3.917724256991367e-06,
      "loss": 0.77607089,
      "num_input_tokens_seen": 20889745,
      "step": 989,
      "time_per_iteration": 2.686502456665039
    },
    {
      "auxiliary_loss_clip": 0.01203132,
      "auxiliary_loss_mlp": 0.010496,
      "balance_loss_clip": 1.06265628,
      "balance_loss_mlp": 1.03812623,
      "epoch": 0.11904046173270005,
      "flos": 30955895763840.0,
      "grad_norm": 2.1633720294946435,
      "language_loss": 0.81432229,
      "learning_rate": 3.9175029831532245e-06,
      "loss": 0.83684963,
      "num_input_tokens_seen": 20909260,
      "step": 990,
      "time_per_iteration": 2.6568806171417236
    },
    {
      "auxiliary_loss_clip": 0.01202964,
      "auxiliary_loss_mlp": 0.01034942,
      "balance_loss_clip": 1.06832671,
      "balance_loss_mlp": 1.02449894,
      "epoch": 0.11916070462333915,
      "flos": 20157234485760.0,
      "grad_norm": 2.0790307712942884,
      "language_loss": 0.88281453,
      "learning_rate": 3.917281418431404e-06,
      "loss": 0.90519357,
      "num_input_tokens_seen": 20928305,
      "step": 991,
      "time_per_iteration": 2.6638309955596924
    },
    {
      "auxiliary_loss_clip": 0.01212428,
      "auxiliary_loss_mlp": 0.01042167,
      "balance_loss_clip": 1.06696117,
      "balance_loss_mlp": 1.0302701,
      "epoch": 0.11928094751397823,
      "flos": 23551115961600.0,
      "grad_norm": 1.9549558493065298,
      "language_loss": 0.76683331,
      "learning_rate": 3.917059562859516e-06,
      "loss": 0.78937924,
      "num_input_tokens_seen": 20947630,
      "step": 992,
      "time_per_iteration": 2.564865827560425
    },
    {
      "auxiliary_loss_clip": 0.01204012,
      "auxiliary_loss_mlp": 0.01048542,
      "balance_loss_clip": 1.0655551,
      "balance_loss_mlp": 1.03557777,
      "epoch": 0.11940119040461733,
      "flos": 23908426502400.0,
      "grad_norm": 2.5006763860702588,
      "language_loss": 0.88598096,
      "learning_rate": 3.916837416471218e-06,
      "loss": 0.90850651,
      "num_input_tokens_seen": 20964250,
      "step": 993,
      "time_per_iteration": 2.5598461627960205
    },
    {
      "auxiliary_loss_clip": 0.01221992,
      "auxiliary_loss_mlp": 0.0104173,
      "balance_loss_clip": 1.06321526,
      "balance_loss_mlp": 1.0302918,
      "epoch": 0.11952143329525641,
      "flos": 13844533835520.0,
      "grad_norm": 3.045862317496398,
      "language_loss": 0.71790588,
      "learning_rate": 3.916614979300207e-06,
      "loss": 0.74054313,
      "num_input_tokens_seen": 20979095,
      "step": 994,
      "time_per_iteration": 2.4886584281921387
    },
    {
      "auxiliary_loss_clip": 0.01170908,
      "auxiliary_loss_mlp": 0.01042568,
      "balance_loss_clip": 1.05977333,
      "balance_loss_mlp": 1.03173828,
      "epoch": 0.11964167618589551,
      "flos": 27015525792000.0,
      "grad_norm": 1.554600703162546,
      "language_loss": 0.78779697,
      "learning_rate": 3.9163922513802274e-06,
      "loss": 0.80993176,
      "num_input_tokens_seen": 21001430,
      "step": 995,
      "time_per_iteration": 2.7433764934539795
    },
    {
      "auxiliary_loss_clip": 0.01247442,
      "auxiliary_loss_mlp": 0.01039953,
      "balance_loss_clip": 1.06869578,
      "balance_loss_mlp": 1.02862787,
      "epoch": 0.1197619190765346,
      "flos": 12567622273920.0,
      "grad_norm": 2.7222687295562857,
      "language_loss": 0.82726228,
      "learning_rate": 3.916169232745067e-06,
      "loss": 0.85013628,
      "num_input_tokens_seen": 21019105,
      "step": 996,
      "time_per_iteration": 2.5554778575897217
    },
    {
      "auxiliary_loss_clip": 0.01201251,
      "auxiliary_loss_mlp": 0.01045157,
      "balance_loss_clip": 1.0626092,
      "balance_loss_mlp": 1.033355,
      "epoch": 0.11988216196717369,
      "flos": 16909437623040.0,
      "grad_norm": 2.586800204020113,
      "language_loss": 0.91545796,
      "learning_rate": 3.915945923428559e-06,
      "loss": 0.93792206,
      "num_input_tokens_seen": 21035630,
      "step": 997,
      "time_per_iteration": 2.5254056453704834
    },
    {
      "auxiliary_loss_clip": 0.01223692,
      "auxiliary_loss_mlp": 0.01038237,
      "balance_loss_clip": 1.06402433,
      "balance_loss_mlp": 1.02670395,
      "epoch": 0.12000240485781279,
      "flos": 16216577205120.0,
      "grad_norm": 2.3763455338426884,
      "language_loss": 0.82817006,
      "learning_rate": 3.915722323464577e-06,
      "loss": 0.85078937,
      "num_input_tokens_seen": 21054235,
      "step": 998,
      "time_per_iteration": 2.485231876373291
    },
    {
      "auxiliary_loss_clip": 0.01229292,
      "auxiliary_loss_mlp": 0.01042071,
      "balance_loss_clip": 1.06680942,
      "balance_loss_mlp": 1.03034723,
      "epoch": 0.12012264774845187,
      "flos": 49344887525760.0,
      "grad_norm": 2.3394692009951137,
      "language_loss": 0.70101988,
      "learning_rate": 3.91549843288704e-06,
      "loss": 0.72373348,
      "num_input_tokens_seen": 21077915,
      "step": 999,
      "time_per_iteration": 2.760650396347046
    },
    {
      "auxiliary_loss_clip": 0.01193444,
      "auxiliary_loss_mlp": 0.00765927,
      "balance_loss_clip": 1.05798531,
      "balance_loss_mlp": 1.00029325,
      "epoch": 0.12024289063909097,
      "flos": 26979435601920.0,
      "grad_norm": 1.9741332994676182,
      "language_loss": 0.79050684,
      "learning_rate": 3.915274251729916e-06,
      "loss": 0.81010056,
      "num_input_tokens_seen": 21099205,
      "step": 1000,
      "time_per_iteration": 2.697967290878296
    },
    {
      "auxiliary_loss_clip": 0.01199432,
      "auxiliary_loss_mlp": 0.01035334,
      "balance_loss_clip": 1.0646801,
      "balance_loss_mlp": 1.02346087,
      "epoch": 0.12036313352973005,
      "flos": 19537308633600.0,
      "grad_norm": 1.9642043382316006,
      "language_loss": 0.89977753,
      "learning_rate": 3.91504978002721e-06,
      "loss": 0.92212522,
      "num_input_tokens_seen": 21118260,
      "step": 1001,
      "time_per_iteration": 2.580869197845459
    },
    {
      "auxiliary_loss_clip": 0.01215672,
      "auxiliary_loss_mlp": 0.00765759,
      "balance_loss_clip": 1.06333685,
      "balance_loss_mlp": 1.00017905,
      "epoch": 0.12048337642036915,
      "flos": 17268256535040.0,
      "grad_norm": 1.956488779013999,
      "language_loss": 0.76307422,
      "learning_rate": 3.914825017812974e-06,
      "loss": 0.78288853,
      "num_input_tokens_seen": 21134910,
      "step": 1002,
      "time_per_iteration": 2.6101584434509277
    },
    {
      "auxiliary_loss_clip": 0.0121423,
      "auxiliary_loss_mlp": 0.01042335,
      "balance_loss_clip": 1.06611443,
      "balance_loss_mlp": 1.03052139,
      "epoch": 0.12060361931100824,
      "flos": 22856962654080.0,
      "grad_norm": 2.6868168685223743,
      "language_loss": 0.72553027,
      "learning_rate": 3.9145999651213065e-06,
      "loss": 0.74809587,
      "num_input_tokens_seen": 21154150,
      "step": 1003,
      "time_per_iteration": 2.6212010383605957
    },
    {
      "auxiliary_loss_clip": 0.01230815,
      "auxiliary_loss_mlp": 0.01043561,
      "balance_loss_clip": 1.06716847,
      "balance_loss_mlp": 1.03109193,
      "epoch": 0.12072386220164733,
      "flos": 16726795943040.0,
      "grad_norm": 3.28910021969422,
      "language_loss": 0.88465434,
      "learning_rate": 3.9143746219863465e-06,
      "loss": 0.9073981,
      "num_input_tokens_seen": 21171255,
      "step": 1004,
      "time_per_iteration": 2.501833915710449
    },
    {
      "auxiliary_loss_clip": 0.01119583,
      "auxiliary_loss_mlp": 0.01007689,
      "balance_loss_clip": 1.03047168,
      "balance_loss_mlp": 1.00439918,
      "epoch": 0.12084410509228642,
      "flos": 55144176105600.0,
      "grad_norm": 0.9718233589977261,
      "language_loss": 0.64785594,
      "learning_rate": 3.914148988442278e-06,
      "loss": 0.66912866,
      "num_input_tokens_seen": 21227045,
      "step": 1005,
      "time_per_iteration": 3.104682683944702
    },
    {
      "auxiliary_loss_clip": 0.01200345,
      "auxiliary_loss_mlp": 0.01036453,
      "balance_loss_clip": 1.06267571,
      "balance_loss_mlp": 1.02454448,
      "epoch": 0.1209643479829255,
      "flos": 26760236855040.0,
      "grad_norm": 2.7257032149404576,
      "language_loss": 0.94802999,
      "learning_rate": 3.91392306452333e-06,
      "loss": 0.97039801,
      "num_input_tokens_seen": 21244120,
      "step": 1006,
      "time_per_iteration": 2.603006601333618
    },
    {
      "auxiliary_loss_clip": 0.0124971,
      "auxiliary_loss_mlp": 0.0103622,
      "balance_loss_clip": 1.07035995,
      "balance_loss_mlp": 1.02459073,
      "epoch": 0.1210845908735646,
      "flos": 11035026725760.0,
      "grad_norm": 4.355028201786306,
      "language_loss": 0.66476691,
      "learning_rate": 3.913696850263774e-06,
      "loss": 0.68762624,
      "num_input_tokens_seen": 21258485,
      "step": 1007,
      "time_per_iteration": 2.5299501419067383
    },
    {
      "auxiliary_loss_clip": 0.01228557,
      "auxiliary_loss_mlp": 0.01037599,
      "balance_loss_clip": 1.06662476,
      "balance_loss_mlp": 1.02625632,
      "epoch": 0.1212048337642037,
      "flos": 20484631975680.0,
      "grad_norm": 2.345442799468082,
      "language_loss": 0.79027092,
      "learning_rate": 3.913470345697929e-06,
      "loss": 0.81293243,
      "num_input_tokens_seen": 21277115,
      "step": 1008,
      "time_per_iteration": 3.3289883136749268
    },
    {
      "auxiliary_loss_clip": 0.01184583,
      "auxiliary_loss_mlp": 0.01040448,
      "balance_loss_clip": 1.06186759,
      "balance_loss_mlp": 1.02905202,
      "epoch": 0.12132507665484278,
      "flos": 22346061557760.0,
      "grad_norm": 2.14806818769836,
      "language_loss": 0.85448384,
      "learning_rate": 3.913243550860153e-06,
      "loss": 0.87673414,
      "num_input_tokens_seen": 21294880,
      "step": 1009,
      "time_per_iteration": 2.643724203109741
    },
    {
      "auxiliary_loss_clip": 0.01235156,
      "auxiliary_loss_mlp": 0.01041644,
      "balance_loss_clip": 1.07236242,
      "balance_loss_mlp": 1.02952623,
      "epoch": 0.12144531954548188,
      "flos": 29314957818240.0,
      "grad_norm": 4.906364335392105,
      "language_loss": 0.76191723,
      "learning_rate": 3.913016465784852e-06,
      "loss": 0.78468519,
      "num_input_tokens_seen": 21315555,
      "step": 1010,
      "time_per_iteration": 2.5796854496002197
    },
    {
      "auxiliary_loss_clip": 0.01181669,
      "auxiliary_loss_mlp": 0.01040886,
      "balance_loss_clip": 1.05798078,
      "balance_loss_mlp": 1.02863169,
      "epoch": 0.12156556243612096,
      "flos": 20485242506880.0,
      "grad_norm": 2.396291420679566,
      "language_loss": 0.72145033,
      "learning_rate": 3.912789090506474e-06,
      "loss": 0.74367583,
      "num_input_tokens_seen": 21334815,
      "step": 1011,
      "time_per_iteration": 2.6079001426696777
    },
    {
      "auxiliary_loss_clip": 0.01204323,
      "auxiliary_loss_mlp": 0.01046784,
      "balance_loss_clip": 1.0612576,
      "balance_loss_mlp": 1.03437412,
      "epoch": 0.12168580532676006,
      "flos": 16472009796480.0,
      "grad_norm": 2.8260098849450337,
      "language_loss": 0.71941292,
      "learning_rate": 3.9125614250595114e-06,
      "loss": 0.74192393,
      "num_input_tokens_seen": 21351025,
      "step": 1012,
      "time_per_iteration": 3.292996406555176
    },
    {
      "auxiliary_loss_clip": 0.01228391,
      "auxiliary_loss_mlp": 0.01038409,
      "balance_loss_clip": 1.06464124,
      "balance_loss_mlp": 1.0263629,
      "epoch": 0.12180604821739914,
      "flos": 15341290588800.0,
      "grad_norm": 4.1629457312837275,
      "language_loss": 0.8891508,
      "learning_rate": 3.912333469478502e-06,
      "loss": 0.91181874,
      "num_input_tokens_seen": 21368990,
      "step": 1013,
      "time_per_iteration": 2.4810686111450195
    },
    {
      "auxiliary_loss_clip": 0.01211681,
      "auxiliary_loss_mlp": 0.01036762,
      "balance_loss_clip": 1.06220818,
      "balance_loss_mlp": 1.02556276,
      "epoch": 0.12192629110803824,
      "flos": 19318038059520.0,
      "grad_norm": 2.192660204255425,
      "language_loss": 0.7811994,
      "learning_rate": 3.912105223798025e-06,
      "loss": 0.80368388,
      "num_input_tokens_seen": 21388410,
      "step": 1014,
      "time_per_iteration": 4.213102102279663
    },
    {
      "auxiliary_loss_clip": 0.01104352,
      "auxiliary_loss_mlp": 0.01006984,
      "balance_loss_clip": 1.02628875,
      "balance_loss_mlp": 1.00400352,
      "epoch": 0.12204653399867733,
      "flos": 47725354085760.0,
      "grad_norm": 1.004827507167101,
      "language_loss": 0.67704082,
      "learning_rate": 3.9118766880527065e-06,
      "loss": 0.69815409,
      "num_input_tokens_seen": 21442845,
      "step": 1015,
      "time_per_iteration": 3.033341884613037
    },
    {
      "auxiliary_loss_clip": 0.01172378,
      "auxiliary_loss_mlp": 0.01033919,
      "balance_loss_clip": 1.05895948,
      "balance_loss_mlp": 1.02293396,
      "epoch": 0.12216677688931642,
      "flos": 18221936584320.0,
      "grad_norm": 1.7625180339866533,
      "language_loss": 0.73916751,
      "learning_rate": 3.9116478622772145e-06,
      "loss": 0.76123047,
      "num_input_tokens_seen": 21461420,
      "step": 1016,
      "time_per_iteration": 2.603590965270996
    },
    {
      "auxiliary_loss_clip": 0.01228523,
      "auxiliary_loss_mlp": 0.01047251,
      "balance_loss_clip": 1.06761563,
      "balance_loss_mlp": 1.03543186,
      "epoch": 0.12228701977995551,
      "flos": 27525636789120.0,
      "grad_norm": 1.641851045288504,
      "language_loss": 0.87946522,
      "learning_rate": 3.911418746506261e-06,
      "loss": 0.90222299,
      "num_input_tokens_seen": 21481550,
      "step": 1017,
      "time_per_iteration": 2.5945804119110107
    },
    {
      "auxiliary_loss_clip": 0.01236243,
      "auxiliary_loss_mlp": 0.0104591,
      "balance_loss_clip": 1.07309771,
      "balance_loss_mlp": 1.03404868,
      "epoch": 0.1224072626705946,
      "flos": 21798136517760.0,
      "grad_norm": 1.7461045162125741,
      "language_loss": 0.78283679,
      "learning_rate": 3.911189340774604e-06,
      "loss": 0.80565834,
      "num_input_tokens_seen": 21501680,
      "step": 1018,
      "time_per_iteration": 2.583975315093994
    },
    {
      "auxiliary_loss_clip": 0.01222008,
      "auxiliary_loss_mlp": 0.01040371,
      "balance_loss_clip": 1.06582403,
      "balance_loss_mlp": 1.0284853,
      "epoch": 0.1225275055612337,
      "flos": 20703758895360.0,
      "grad_norm": 1.8798613184521535,
      "language_loss": 0.79470217,
      "learning_rate": 3.910959645117043e-06,
      "loss": 0.81732595,
      "num_input_tokens_seen": 21521015,
      "step": 1019,
      "time_per_iteration": 2.5504541397094727
    },
    {
      "auxiliary_loss_clip": 0.01108783,
      "auxiliary_loss_mlp": 0.00755753,
      "balance_loss_clip": 1.02608109,
      "balance_loss_mlp": 0.99982893,
      "epoch": 0.12264774845187278,
      "flos": 57745294462080.0,
      "grad_norm": 0.8250716008429229,
      "language_loss": 0.5673728,
      "learning_rate": 3.910729659568423e-06,
      "loss": 0.58601809,
      "num_input_tokens_seen": 21578200,
      "step": 1020,
      "time_per_iteration": 3.0903711318969727
    },
    {
      "auxiliary_loss_clip": 0.01214141,
      "auxiliary_loss_mlp": 0.01040553,
      "balance_loss_clip": 1.06681371,
      "balance_loss_mlp": 1.02984226,
      "epoch": 0.12276799134251187,
      "flos": 26396282298240.0,
      "grad_norm": 1.8921659514973732,
      "language_loss": 0.82302976,
      "learning_rate": 3.9104993841636344e-06,
      "loss": 0.8455767,
      "num_input_tokens_seen": 21598770,
      "step": 1021,
      "time_per_iteration": 2.5958285331726074
    },
    {
      "auxiliary_loss_clip": 0.01213043,
      "auxiliary_loss_mlp": 0.00765046,
      "balance_loss_clip": 1.06837499,
      "balance_loss_mlp": 1.00014257,
      "epoch": 0.12288823423315097,
      "flos": 21064193919360.0,
      "grad_norm": 1.838728974577964,
      "language_loss": 0.80708408,
      "learning_rate": 3.910268818937608e-06,
      "loss": 0.82686502,
      "num_input_tokens_seen": 21616925,
      "step": 1022,
      "time_per_iteration": 2.6083619594573975
    },
    {
      "auxiliary_loss_clip": 0.01182559,
      "auxiliary_loss_mlp": 0.01040164,
      "balance_loss_clip": 1.06323099,
      "balance_loss_mlp": 1.02887464,
      "epoch": 0.12300847712379005,
      "flos": 12312441077760.0,
      "grad_norm": 2.79147987123607,
      "language_loss": 0.87200677,
      "learning_rate": 3.9100379639253196e-06,
      "loss": 0.89423394,
      "num_input_tokens_seen": 21633645,
      "step": 1023,
      "time_per_iteration": 2.563525676727295
    },
    {
      "auxiliary_loss_clip": 0.01209931,
      "auxiliary_loss_mlp": 0.01038633,
      "balance_loss_clip": 1.05983126,
      "balance_loss_mlp": 1.02677727,
      "epoch": 0.12312872001442915,
      "flos": 16762239688320.0,
      "grad_norm": 7.128808907151381,
      "language_loss": 0.86316663,
      "learning_rate": 3.909806819161791e-06,
      "loss": 0.8856523,
      "num_input_tokens_seen": 21649120,
      "step": 1024,
      "time_per_iteration": 2.5015146732330322
    },
    {
      "auxiliary_loss_clip": 0.01200876,
      "auxiliary_loss_mlp": 0.01035231,
      "balance_loss_clip": 1.06189966,
      "balance_loss_mlp": 1.02354836,
      "epoch": 0.12324896290506823,
      "flos": 18404937400320.0,
      "grad_norm": 3.367361678904254,
      "language_loss": 0.85992765,
      "learning_rate": 3.909575384682086e-06,
      "loss": 0.88228869,
      "num_input_tokens_seen": 21668000,
      "step": 1025,
      "time_per_iteration": 2.5635976791381836
    },
    {
      "auxiliary_loss_clip": 0.01230146,
      "auxiliary_loss_mlp": 0.01053171,
      "balance_loss_clip": 1.06511676,
      "balance_loss_mlp": 1.04119062,
      "epoch": 0.12336920579570733,
      "flos": 18915407533440.0,
      "grad_norm": 2.107270800292623,
      "language_loss": 0.69035459,
      "learning_rate": 3.9093436605213144e-06,
      "loss": 0.71318769,
      "num_input_tokens_seen": 21688500,
      "step": 1026,
      "time_per_iteration": 2.5208303928375244
    },
    {
      "auxiliary_loss_clip": 0.01213136,
      "auxiliary_loss_mlp": 0.01045114,
      "balance_loss_clip": 1.06448293,
      "balance_loss_mlp": 1.03399217,
      "epoch": 0.12348944868634643,
      "flos": 23878369797120.0,
      "grad_norm": 1.931677531205367,
      "language_loss": 0.79118574,
      "learning_rate": 3.909111646714627e-06,
      "loss": 0.81376833,
      "num_input_tokens_seen": 21709345,
      "step": 1027,
      "time_per_iteration": 2.563126802444458
    },
    {
      "auxiliary_loss_clip": 0.01239718,
      "auxiliary_loss_mlp": 0.01032906,
      "balance_loss_clip": 1.06591201,
      "balance_loss_mlp": 1.02241564,
      "epoch": 0.12360969157698551,
      "flos": 19026084314880.0,
      "grad_norm": 3.2425410989092485,
      "language_loss": 0.72293043,
      "learning_rate": 3.9088793432972206e-06,
      "loss": 0.74565667,
      "num_input_tokens_seen": 21728165,
      "step": 1028,
      "time_per_iteration": 2.495054006576538
    },
    {
      "auxiliary_loss_clip": 0.0118091,
      "auxiliary_loss_mlp": 0.01041615,
      "balance_loss_clip": 1.06134009,
      "balance_loss_mlp": 1.03015304,
      "epoch": 0.1237299344676246,
      "flos": 13224607983360.0,
      "grad_norm": 2.1911629470058616,
      "language_loss": 0.81957233,
      "learning_rate": 3.908646750304336e-06,
      "loss": 0.84179765,
      "num_input_tokens_seen": 21745850,
      "step": 1029,
      "time_per_iteration": 2.695385217666626
    },
    {
      "auxiliary_loss_clip": 0.01215696,
      "auxiliary_loss_mlp": 0.01038473,
      "balance_loss_clip": 1.06603193,
      "balance_loss_mlp": 1.02723098,
      "epoch": 0.12385017735826369,
      "flos": 20485673470080.0,
      "grad_norm": 1.736101877835852,
      "language_loss": 0.87471676,
      "learning_rate": 3.908413867771257e-06,
      "loss": 0.89725852,
      "num_input_tokens_seen": 21764760,
      "step": 1030,
      "time_per_iteration": 2.531994342803955
    },
    {
      "auxiliary_loss_clip": 0.01227987,
      "auxiliary_loss_mlp": 0.01044605,
      "balance_loss_clip": 1.06733656,
      "balance_loss_mlp": 1.03248167,
      "epoch": 0.12397042024890279,
      "flos": 17347835116800.0,
      "grad_norm": 1.783422361228728,
      "language_loss": 0.80791593,
      "learning_rate": 3.908180695733311e-06,
      "loss": 0.83064187,
      "num_input_tokens_seen": 21784250,
      "step": 1031,
      "time_per_iteration": 2.479642152786255
    },
    {
      "auxiliary_loss_clip": 0.01159212,
      "auxiliary_loss_mlp": 0.01046317,
      "balance_loss_clip": 1.05432057,
      "balance_loss_mlp": 1.03497982,
      "epoch": 0.12409066313954187,
      "flos": 20412343854720.0,
      "grad_norm": 1.8873261131681616,
      "language_loss": 0.82790679,
      "learning_rate": 3.907947234225871e-06,
      "loss": 0.84996206,
      "num_input_tokens_seen": 21803260,
      "step": 1032,
      "time_per_iteration": 2.625009775161743
    },
    {
      "auxiliary_loss_clip": 0.01160627,
      "auxiliary_loss_mlp": 0.01035329,
      "balance_loss_clip": 1.0585804,
      "balance_loss_mlp": 1.02444494,
      "epoch": 0.12421090603018096,
      "flos": 20736688688640.0,
      "grad_norm": 2.190576644275307,
      "language_loss": 0.87038624,
      "learning_rate": 3.907713483284352e-06,
      "loss": 0.89234579,
      "num_input_tokens_seen": 21822735,
      "step": 1033,
      "time_per_iteration": 2.6443607807159424
    },
    {
      "auxiliary_loss_clip": 0.01140039,
      "auxiliary_loss_mlp": 0.01044361,
      "balance_loss_clip": 1.05256724,
      "balance_loss_mlp": 1.03128386,
      "epoch": 0.12433114892082006,
      "flos": 24498834353280.0,
      "grad_norm": 2.2901131175290224,
      "language_loss": 0.97754049,
      "learning_rate": 3.907479442944216e-06,
      "loss": 0.99938446,
      "num_input_tokens_seen": 21841140,
      "step": 1034,
      "time_per_iteration": 3.532508134841919
    },
    {
      "auxiliary_loss_clip": 0.01225909,
      "auxiliary_loss_mlp": 0.01036581,
      "balance_loss_clip": 1.0662756,
      "balance_loss_mlp": 1.02637112,
      "epoch": 0.12445139181145914,
      "flos": 19682315838720.0,
      "grad_norm": 2.198523362472246,
      "language_loss": 0.9256736,
      "learning_rate": 3.907245113240963e-06,
      "loss": 0.94829845,
      "num_input_tokens_seen": 21859260,
      "step": 1035,
      "time_per_iteration": 2.4889605045318604
    },
    {
      "auxiliary_loss_clip": 0.01192853,
      "auxiliary_loss_mlp": 0.01035134,
      "balance_loss_clip": 1.05667853,
      "balance_loss_mlp": 1.0236001,
      "epoch": 0.12457163470209824,
      "flos": 46423087522560.0,
      "grad_norm": 1.6752296986508524,
      "language_loss": 0.73605829,
      "learning_rate": 3.907010494210144e-06,
      "loss": 0.75833815,
      "num_input_tokens_seen": 21881920,
      "step": 1036,
      "time_per_iteration": 2.764355421066284
    },
    {
      "auxiliary_loss_clip": 0.01230011,
      "auxiliary_loss_mlp": 0.01046237,
      "balance_loss_clip": 1.06778121,
      "balance_loss_mlp": 1.03355885,
      "epoch": 0.12469187759273732,
      "flos": 20376289578240.0,
      "grad_norm": 2.0552989901705367,
      "language_loss": 0.92072618,
      "learning_rate": 3.9067755858873495e-06,
      "loss": 0.9434886,
      "num_input_tokens_seen": 21898720,
      "step": 1037,
      "time_per_iteration": 2.509488344192505
    },
    {
      "auxiliary_loss_clip": 0.01088206,
      "auxiliary_loss_mlp": 0.01006039,
      "balance_loss_clip": 1.01946926,
      "balance_loss_mlp": 1.00302303,
      "epoch": 0.12481212048337642,
      "flos": 69224641447680.0,
      "grad_norm": 0.8631232454559965,
      "language_loss": 0.62778598,
      "learning_rate": 3.906540388308214e-06,
      "loss": 0.64872843,
      "num_input_tokens_seen": 21958305,
      "step": 1038,
      "time_per_iteration": 3.111362934112549
    },
    {
      "auxiliary_loss_clip": 0.01166583,
      "auxiliary_loss_mlp": 0.01047939,
      "balance_loss_clip": 1.05954885,
      "balance_loss_mlp": 1.03619719,
      "epoch": 0.12493236337401552,
      "flos": 18223696350720.0,
      "grad_norm": 1.7252077372252637,
      "language_loss": 0.81430256,
      "learning_rate": 3.906304901508417e-06,
      "loss": 0.83644783,
      "num_input_tokens_seen": 21977205,
      "step": 1039,
      "time_per_iteration": 3.4055750370025635
    },
    {
      "auxiliary_loss_clip": 0.01230046,
      "auxiliary_loss_mlp": 0.01043586,
      "balance_loss_clip": 1.07025886,
      "balance_loss_mlp": 1.03304839,
      "epoch": 0.12505260626465461,
      "flos": 30044375303040.0,
      "grad_norm": 2.030673336104599,
      "language_loss": 0.7545746,
      "learning_rate": 3.9060691255236835e-06,
      "loss": 0.77731097,
      "num_input_tokens_seen": 21997770,
      "step": 1040,
      "time_per_iteration": 2.5736887454986572
    },
    {
      "auxiliary_loss_clip": 0.01219124,
      "auxiliary_loss_mlp": 0.01041874,
      "balance_loss_clip": 1.05977941,
      "balance_loss_mlp": 1.02981019,
      "epoch": 0.1251728491552937,
      "flos": 24433980347520.0,
      "grad_norm": 1.7378624766192161,
      "language_loss": 0.80589688,
      "learning_rate": 3.905833060389778e-06,
      "loss": 0.82850689,
      "num_input_tokens_seen": 22021890,
      "step": 1041,
      "time_per_iteration": 4.35455584526062
    },
    {
      "auxiliary_loss_clip": 0.01242913,
      "auxiliary_loss_mlp": 0.00765922,
      "balance_loss_clip": 1.06769443,
      "balance_loss_mlp": 1.00015247,
      "epoch": 0.12529309204593278,
      "flos": 27119809952640.0,
      "grad_norm": 2.4006404917695154,
      "language_loss": 0.78460449,
      "learning_rate": 3.905596706142513e-06,
      "loss": 0.80469286,
      "num_input_tokens_seen": 22043300,
      "step": 1042,
      "time_per_iteration": 2.5455574989318848
    },
    {
      "auxiliary_loss_clip": 0.0119042,
      "auxiliary_loss_mlp": 0.01042626,
      "balance_loss_clip": 1.05873978,
      "balance_loss_mlp": 1.03075874,
      "epoch": 0.12541333493657186,
      "flos": 30774151923840.0,
      "grad_norm": 1.9265047748491917,
      "language_loss": 0.85775608,
      "learning_rate": 3.9053600628177435e-06,
      "loss": 0.88008654,
      "num_input_tokens_seen": 22062910,
      "step": 1043,
      "time_per_iteration": 2.6622276306152344
    },
    {
      "auxiliary_loss_clip": 0.01240352,
      "auxiliary_loss_mlp": 0.01037129,
      "balance_loss_clip": 1.06610239,
      "balance_loss_mlp": 1.02621567,
      "epoch": 0.12553357782721097,
      "flos": 23659566099840.0,
      "grad_norm": 2.189959295462693,
      "language_loss": 0.84603149,
      "learning_rate": 3.905123130451367e-06,
      "loss": 0.86880624,
      "num_input_tokens_seen": 22084010,
      "step": 1044,
      "time_per_iteration": 2.6156985759735107
    },
    {
      "auxiliary_loss_clip": 0.01243552,
      "auxiliary_loss_mlp": 0.01038057,
      "balance_loss_clip": 1.06874752,
      "balance_loss_mlp": 1.02632654,
      "epoch": 0.12565382071785006,
      "flos": 24863758577280.0,
      "grad_norm": 1.8574377599064194,
      "language_loss": 0.79525554,
      "learning_rate": 3.904885909079326e-06,
      "loss": 0.8180716,
      "num_input_tokens_seen": 22102795,
      "step": 1045,
      "time_per_iteration": 2.5214390754699707
    },
    {
      "auxiliary_loss_clip": 0.0122688,
      "auxiliary_loss_mlp": 0.01037631,
      "balance_loss_clip": 1.06402993,
      "balance_loss_mlp": 1.02630591,
      "epoch": 0.12577406360848914,
      "flos": 21360780518400.0,
      "grad_norm": 2.1491868964460044,
      "language_loss": 0.77576238,
      "learning_rate": 3.904648398737607e-06,
      "loss": 0.79840744,
      "num_input_tokens_seen": 22121360,
      "step": 1046,
      "time_per_iteration": 2.5263524055480957
    },
    {
      "auxiliary_loss_clip": 0.01241834,
      "auxiliary_loss_mlp": 0.01042549,
      "balance_loss_clip": 1.06716454,
      "balance_loss_mlp": 1.03160596,
      "epoch": 0.12589430649912825,
      "flos": 36138056774400.0,
      "grad_norm": 1.7367113541557497,
      "language_loss": 0.78241658,
      "learning_rate": 3.9044105994622406e-06,
      "loss": 0.80526048,
      "num_input_tokens_seen": 22142505,
      "step": 1047,
      "time_per_iteration": 2.6438097953796387
    },
    {
      "auxiliary_loss_clip": 0.01213877,
      "auxiliary_loss_mlp": 0.00766208,
      "balance_loss_clip": 1.06240201,
      "balance_loss_mlp": 1.00026429,
      "epoch": 0.12601454938976733,
      "flos": 25337671643520.0,
      "grad_norm": 1.9834209488074352,
      "language_loss": 0.81556469,
      "learning_rate": 3.9041725112893005e-06,
      "loss": 0.83536553,
      "num_input_tokens_seen": 22163730,
      "step": 1048,
      "time_per_iteration": 2.6039910316467285
    },
    {
      "auxiliary_loss_clip": 0.01191212,
      "auxiliary_loss_mlp": 0.0103965,
      "balance_loss_clip": 1.06269813,
      "balance_loss_mlp": 1.02858782,
      "epoch": 0.12613479228040642,
      "flos": 15560094286080.0,
      "grad_norm": 2.0108690803891704,
      "language_loss": 0.74804461,
      "learning_rate": 3.903934134254904e-06,
      "loss": 0.77035326,
      "num_input_tokens_seen": 22181520,
      "step": 1049,
      "time_per_iteration": 2.567673444747925
    },
    {
      "auxiliary_loss_clip": 0.01230172,
      "auxiliary_loss_mlp": 0.01043875,
      "balance_loss_clip": 1.06453228,
      "balance_loss_mlp": 1.03203142,
      "epoch": 0.1262550351710455,
      "flos": 21470595373440.0,
      "grad_norm": 2.921216138891059,
      "language_loss": 0.84975553,
      "learning_rate": 3.903695468395213e-06,
      "loss": 0.87249595,
      "num_input_tokens_seen": 22199390,
      "step": 1050,
      "time_per_iteration": 2.5238101482391357
    },
    {
      "auxiliary_loss_clip": 0.01213068,
      "auxiliary_loss_mlp": 0.01042308,
      "balance_loss_clip": 1.05948162,
      "balance_loss_mlp": 1.03168654,
      "epoch": 0.1263752780616846,
      "flos": 31576719456000.0,
      "grad_norm": 2.6318880292129165,
      "language_loss": 0.55679047,
      "learning_rate": 3.903456513746434e-06,
      "loss": 0.57934421,
      "num_input_tokens_seen": 22220365,
      "step": 1051,
      "time_per_iteration": 2.613668203353882
    },
    {
      "auxiliary_loss_clip": 0.01238207,
      "auxiliary_loss_mlp": 0.01038733,
      "balance_loss_clip": 1.06540203,
      "balance_loss_mlp": 1.02813554,
      "epoch": 0.1264955209523237,
      "flos": 28768217927040.0,
      "grad_norm": 1.7206193974943775,
      "language_loss": 0.87629473,
      "learning_rate": 3.903217270344815e-06,
      "loss": 0.89906412,
      "num_input_tokens_seen": 22240615,
      "step": 1052,
      "time_per_iteration": 2.537637710571289
    },
    {
      "auxiliary_loss_clip": 0.01184548,
      "auxiliary_loss_mlp": 0.01037958,
      "balance_loss_clip": 1.05635595,
      "balance_loss_mlp": 1.02667487,
      "epoch": 0.12661576384296278,
      "flos": 29241125412480.0,
      "grad_norm": 2.076397698030303,
      "language_loss": 0.82413191,
      "learning_rate": 3.902977738226648e-06,
      "loss": 0.84635699,
      "num_input_tokens_seen": 22261350,
      "step": 1053,
      "time_per_iteration": 2.6436336040496826
    },
    {
      "auxiliary_loss_clip": 0.01229076,
      "auxiliary_loss_mlp": 0.0104186,
      "balance_loss_clip": 1.06581616,
      "balance_loss_mlp": 1.02984357,
      "epoch": 0.12673600673360189,
      "flos": 20850346298880.0,
      "grad_norm": 2.012519176240002,
      "language_loss": 0.91379714,
      "learning_rate": 3.902737917428273e-06,
      "loss": 0.93650657,
      "num_input_tokens_seen": 22279515,
      "step": 1054,
      "time_per_iteration": 2.490696668624878
    },
    {
      "auxiliary_loss_clip": 0.01239093,
      "auxiliary_loss_mlp": 0.01039842,
      "balance_loss_clip": 1.06527889,
      "balance_loss_mlp": 1.02880883,
      "epoch": 0.12685624962424097,
      "flos": 25263695583360.0,
      "grad_norm": 1.6635908928488745,
      "language_loss": 0.83977073,
      "learning_rate": 3.902497807986068e-06,
      "loss": 0.86256003,
      "num_input_tokens_seen": 22299535,
      "step": 1055,
      "time_per_iteration": 2.502523422241211
    },
    {
      "auxiliary_loss_clip": 0.01194806,
      "auxiliary_loss_mlp": 0.01039005,
      "balance_loss_clip": 1.05784416,
      "balance_loss_mlp": 1.02744198,
      "epoch": 0.12697649251488005,
      "flos": 27527109246720.0,
      "grad_norm": 1.5842689481459875,
      "language_loss": 0.83632183,
      "learning_rate": 3.902257409936458e-06,
      "loss": 0.85865998,
      "num_input_tokens_seen": 22320300,
      "step": 1056,
      "time_per_iteration": 2.641488552093506
    },
    {
      "auxiliary_loss_clip": 0.01209253,
      "auxiliary_loss_mlp": 0.01039576,
      "balance_loss_clip": 1.06527722,
      "balance_loss_mlp": 1.0288713,
      "epoch": 0.12709673540551916,
      "flos": 21251863503360.0,
      "grad_norm": 1.8396098698391874,
      "language_loss": 0.83933568,
      "learning_rate": 3.902016723315912e-06,
      "loss": 0.86182398,
      "num_input_tokens_seen": 22338240,
      "step": 1057,
      "time_per_iteration": 2.5296690464019775
    },
    {
      "auxiliary_loss_clip": 0.01221255,
      "auxiliary_loss_mlp": 0.01040542,
      "balance_loss_clip": 1.06176949,
      "balance_loss_mlp": 1.02964067,
      "epoch": 0.12721697829615825,
      "flos": 25337707557120.0,
      "grad_norm": 7.034821121062021,
      "language_loss": 0.69578528,
      "learning_rate": 3.901775748160941e-06,
      "loss": 0.71840322,
      "num_input_tokens_seen": 22357420,
      "step": 1058,
      "time_per_iteration": 2.536893606185913
    },
    {
      "auxiliary_loss_clip": 0.01100003,
      "auxiliary_loss_mlp": 0.01008418,
      "balance_loss_clip": 1.02518487,
      "balance_loss_mlp": 1.00559258,
      "epoch": 0.12733722118679733,
      "flos": 61943287754880.0,
      "grad_norm": 0.7970972117639246,
      "language_loss": 0.60888958,
      "learning_rate": 3.901534484508101e-06,
      "loss": 0.62997377,
      "num_input_tokens_seen": 22420095,
      "step": 1059,
      "time_per_iteration": 3.098818302154541
    },
    {
      "auxiliary_loss_clip": 0.01200318,
      "auxiliary_loss_mlp": 0.01037541,
      "balance_loss_clip": 1.0596056,
      "balance_loss_mlp": 1.0263294,
      "epoch": 0.1274574640774364,
      "flos": 26976742081920.0,
      "grad_norm": 2.7969127552917716,
      "language_loss": 0.74671823,
      "learning_rate": 3.901292932393991e-06,
      "loss": 0.76909685,
      "num_input_tokens_seen": 22438975,
      "step": 1060,
      "time_per_iteration": 2.5649211406707764
    },
    {
      "auxiliary_loss_clip": 0.0123987,
      "auxiliary_loss_mlp": 0.01042676,
      "balance_loss_clip": 1.06724727,
      "balance_loss_mlp": 1.03141642,
      "epoch": 0.12757770696807552,
      "flos": 22236318529920.0,
      "grad_norm": 2.818224049382878,
      "language_loss": 0.85229158,
      "learning_rate": 3.9010510918552555e-06,
      "loss": 0.87511706,
      "num_input_tokens_seen": 22458050,
      "step": 1061,
      "time_per_iteration": 3.32651424407959
    },
    {
      "auxiliary_loss_clip": 0.01206446,
      "auxiliary_loss_mlp": 0.01045209,
      "balance_loss_clip": 1.06109846,
      "balance_loss_mlp": 1.03220916,
      "epoch": 0.1276979498587146,
      "flos": 28547905858560.0,
      "grad_norm": 3.845495566493531,
      "language_loss": 0.74482942,
      "learning_rate": 3.900808962928581e-06,
      "loss": 0.76734602,
      "num_input_tokens_seen": 22475665,
      "step": 1062,
      "time_per_iteration": 2.6039249897003174
    },
    {
      "auxiliary_loss_clip": 0.01241533,
      "auxiliary_loss_mlp": 0.01043355,
      "balance_loss_clip": 1.069332,
      "balance_loss_mlp": 1.03210807,
      "epoch": 0.1278181927493537,
      "flos": 17420338719360.0,
      "grad_norm": 2.3556815949442447,
      "language_loss": 0.89496851,
      "learning_rate": 3.900566545650698e-06,
      "loss": 0.91781747,
      "num_input_tokens_seen": 22493335,
      "step": 1063,
      "time_per_iteration": 2.5018365383148193
    },
    {
      "auxiliary_loss_clip": 0.01224384,
      "auxiliary_loss_mlp": 0.01038408,
      "balance_loss_clip": 1.06568766,
      "balance_loss_mlp": 1.02628446,
      "epoch": 0.1279384356399928,
      "flos": 21138636856320.0,
      "grad_norm": 2.4870396700110913,
      "language_loss": 0.81683242,
      "learning_rate": 3.900323840058381e-06,
      "loss": 0.83946037,
      "num_input_tokens_seen": 22511045,
      "step": 1064,
      "time_per_iteration": 2.5124664306640625
    },
    {
      "auxiliary_loss_clip": 0.01222741,
      "auxiliary_loss_mlp": 0.01039625,
      "balance_loss_clip": 1.06175506,
      "balance_loss_mlp": 1.02920628,
      "epoch": 0.12805867853063188,
      "flos": 26576733248640.0,
      "grad_norm": 3.1755074166154538,
      "language_loss": 0.81772017,
      "learning_rate": 3.900080846188449e-06,
      "loss": 0.84034383,
      "num_input_tokens_seen": 22529635,
      "step": 1065,
      "time_per_iteration": 2.5471057891845703
    },
    {
      "auxiliary_loss_clip": 0.01238892,
      "auxiliary_loss_mlp": 0.01036302,
      "balance_loss_clip": 1.06594896,
      "balance_loss_mlp": 1.02464962,
      "epoch": 0.12817892142127096,
      "flos": 16436206915200.0,
      "grad_norm": 2.080051520696307,
      "language_loss": 0.8166948,
      "learning_rate": 3.8998375640777625e-06,
      "loss": 0.83944678,
      "num_input_tokens_seen": 22547505,
      "step": 1066,
      "time_per_iteration": 3.271850347518921
    },
    {
      "auxiliary_loss_clip": 0.01106057,
      "auxiliary_loss_mlp": 0.01002177,
      "balance_loss_clip": 1.03350782,
      "balance_loss_mlp": 0.99933952,
      "epoch": 0.12829916431191005,
      "flos": 60757049099520.0,
      "grad_norm": 0.7070557646598391,
      "language_loss": 0.52648526,
      "learning_rate": 3.899593993763229e-06,
      "loss": 0.54756761,
      "num_input_tokens_seen": 22608465,
      "step": 1067,
      "time_per_iteration": 4.689392805099487
    },
    {
      "auxiliary_loss_clip": 0.01189222,
      "auxiliary_loss_mlp": 0.01041807,
      "balance_loss_clip": 1.06122744,
      "balance_loss_mlp": 1.02853906,
      "epoch": 0.12841940720254916,
      "flos": 29786895636480.0,
      "grad_norm": 4.176731565411882,
      "language_loss": 0.81389064,
      "learning_rate": 3.899350135281796e-06,
      "loss": 0.83620095,
      "num_input_tokens_seen": 22629465,
      "step": 1068,
      "time_per_iteration": 2.6616933345794678
    },
    {
      "auxiliary_loss_clip": 0.01196689,
      "auxiliary_loss_mlp": 0.01038846,
      "balance_loss_clip": 1.06291044,
      "balance_loss_mlp": 1.02799249,
      "epoch": 0.12853965009318824,
      "flos": 25951851319680.0,
      "grad_norm": 2.0782611993186997,
      "language_loss": 0.7963084,
      "learning_rate": 3.8991059886704585e-06,
      "loss": 0.81866372,
      "num_input_tokens_seen": 22648970,
      "step": 1069,
      "time_per_iteration": 2.618229866027832
    },
    {
      "auxiliary_loss_clip": 0.01186604,
      "auxiliary_loss_mlp": 0.01045415,
      "balance_loss_clip": 1.0594517,
      "balance_loss_mlp": 1.03403068,
      "epoch": 0.12865989298382732,
      "flos": 30846871008000.0,
      "grad_norm": 2.1171472038622725,
      "language_loss": 0.8299666,
      "learning_rate": 3.898861553966252e-06,
      "loss": 0.85228682,
      "num_input_tokens_seen": 22668620,
      "step": 1070,
      "time_per_iteration": 2.625675916671753
    },
    {
      "auxiliary_loss_clip": 0.01150602,
      "auxiliary_loss_mlp": 0.0104496,
      "balance_loss_clip": 1.05454755,
      "balance_loss_mlp": 1.03380799,
      "epoch": 0.12878013587446643,
      "flos": 25885776251520.0,
      "grad_norm": 1.767287465185013,
      "language_loss": 0.88117641,
      "learning_rate": 3.898616831206257e-06,
      "loss": 0.90313202,
      "num_input_tokens_seen": 22689045,
      "step": 1071,
      "time_per_iteration": 2.7152910232543945
    },
    {
      "auxiliary_loss_clip": 0.01190213,
      "auxiliary_loss_mlp": 0.01039473,
      "balance_loss_clip": 1.05717158,
      "balance_loss_mlp": 1.025841,
      "epoch": 0.12890037876510552,
      "flos": 23333138277120.0,
      "grad_norm": 2.7753628309298186,
      "language_loss": 0.76919007,
      "learning_rate": 3.8983718204276e-06,
      "loss": 0.79148692,
      "num_input_tokens_seen": 22711265,
      "step": 1072,
      "time_per_iteration": 2.6214137077331543
    },
    {
      "auxiliary_loss_clip": 0.01205405,
      "auxiliary_loss_mlp": 0.0104505,
      "balance_loss_clip": 1.06063318,
      "balance_loss_mlp": 1.03459585,
      "epoch": 0.1290206216557446,
      "flos": 23587242065280.0,
      "grad_norm": 1.801344787788328,
      "language_loss": 0.82660186,
      "learning_rate": 3.898126521667446e-06,
      "loss": 0.84910643,
      "num_input_tokens_seen": 22731420,
      "step": 1073,
      "time_per_iteration": 2.5689280033111572
    },
    {
      "auxiliary_loss_clip": 0.01220513,
      "auxiliary_loss_mlp": 0.01048484,
      "balance_loss_clip": 1.06045818,
      "balance_loss_mlp": 1.03642035,
      "epoch": 0.12914086454638368,
      "flos": 24170610850560.0,
      "grad_norm": 1.6126434237526102,
      "language_loss": 0.83289838,
      "learning_rate": 3.897880934963007e-06,
      "loss": 0.85558832,
      "num_input_tokens_seen": 22750970,
      "step": 1074,
      "time_per_iteration": 2.523643732070923
    },
    {
      "auxiliary_loss_clip": 0.01203398,
      "auxiliary_loss_mlp": 0.01038013,
      "balance_loss_clip": 1.05829477,
      "balance_loss_mlp": 1.02604461,
      "epoch": 0.1292611074370228,
      "flos": 20267157081600.0,
      "grad_norm": 2.45808645228896,
      "language_loss": 0.7873584,
      "learning_rate": 3.89763506035154e-06,
      "loss": 0.80977255,
      "num_input_tokens_seen": 22768820,
      "step": 1075,
      "time_per_iteration": 2.5363309383392334
    },
    {
      "auxiliary_loss_clip": 0.01210482,
      "auxiliary_loss_mlp": 0.01035575,
      "balance_loss_clip": 1.06172299,
      "balance_loss_mlp": 1.02454185,
      "epoch": 0.12938135032766188,
      "flos": 27377684668800.0,
      "grad_norm": 3.0214997472114447,
      "language_loss": 0.80906409,
      "learning_rate": 3.897388897870343e-06,
      "loss": 0.83152467,
      "num_input_tokens_seen": 22789460,
      "step": 1076,
      "time_per_iteration": 2.5580832958221436
    },
    {
      "auxiliary_loss_clip": 0.01220461,
      "auxiliary_loss_mlp": 0.01036571,
      "balance_loss_clip": 1.06149411,
      "balance_loss_mlp": 1.02417898,
      "epoch": 0.12950159321830096,
      "flos": 29277107861760.0,
      "grad_norm": 1.9033456538678497,
      "language_loss": 0.74950051,
      "learning_rate": 3.89714244755676e-06,
      "loss": 0.77207083,
      "num_input_tokens_seen": 22810820,
      "step": 1077,
      "time_per_iteration": 2.6054811477661133
    },
    {
      "auxiliary_loss_clip": 0.01163658,
      "auxiliary_loss_mlp": 0.0104029,
      "balance_loss_clip": 1.05325067,
      "balance_loss_mlp": 1.02897716,
      "epoch": 0.12962183610894007,
      "flos": 24534888629760.0,
      "grad_norm": 2.5521679309733747,
      "language_loss": 0.86465424,
      "learning_rate": 3.896895709448175e-06,
      "loss": 0.88669372,
      "num_input_tokens_seen": 22830570,
      "step": 1078,
      "time_per_iteration": 2.6051576137542725
    },
    {
      "auxiliary_loss_clip": 0.01154197,
      "auxiliary_loss_mlp": 0.01041021,
      "balance_loss_clip": 1.05152571,
      "balance_loss_mlp": 1.02999401,
      "epoch": 0.12974207899957915,
      "flos": 11215944552960.0,
      "grad_norm": 2.658614775245721,
      "language_loss": 0.76757759,
      "learning_rate": 3.896648683582019e-06,
      "loss": 0.78952974,
      "num_input_tokens_seen": 22845905,
      "step": 1079,
      "time_per_iteration": 2.603343963623047
    },
    {
      "auxiliary_loss_clip": 0.01175617,
      "auxiliary_loss_mlp": 0.01036716,
      "balance_loss_clip": 1.06057978,
      "balance_loss_mlp": 1.02589798,
      "epoch": 0.12986232189021824,
      "flos": 24717889445760.0,
      "grad_norm": 2.1070772137520026,
      "language_loss": 0.80655158,
      "learning_rate": 3.896401369995766e-06,
      "loss": 0.82867491,
      "num_input_tokens_seen": 22865710,
      "step": 1080,
      "time_per_iteration": 2.643333673477173
    },
    {
      "auxiliary_loss_clip": 0.01241174,
      "auxiliary_loss_mlp": 0.0104661,
      "balance_loss_clip": 1.06868219,
      "balance_loss_mlp": 1.03552365,
      "epoch": 0.12998256478085732,
      "flos": 23915357827200.0,
      "grad_norm": 1.8554135263392066,
      "language_loss": 0.79311562,
      "learning_rate": 3.896153768726932e-06,
      "loss": 0.81599343,
      "num_input_tokens_seen": 22886020,
      "step": 1081,
      "time_per_iteration": 2.5395965576171875
    },
    {
      "auxiliary_loss_clip": 0.0122403,
      "auxiliary_loss_mlp": 0.01040681,
      "balance_loss_clip": 1.06594539,
      "balance_loss_mlp": 1.02932608,
      "epoch": 0.13010280767149643,
      "flos": 18624207974400.0,
      "grad_norm": 1.9891574188386492,
      "language_loss": 0.88152111,
      "learning_rate": 3.8959058798130806e-06,
      "loss": 0.90416819,
      "num_input_tokens_seen": 22903995,
      "step": 1082,
      "time_per_iteration": 2.474423408508301
    },
    {
      "auxiliary_loss_clip": 0.01211475,
      "auxiliary_loss_mlp": 0.00766485,
      "balance_loss_clip": 1.06288767,
      "balance_loss_mlp": 1.00034475,
      "epoch": 0.1302230505621355,
      "flos": 22783992174720.0,
      "grad_norm": 1.93527029883986,
      "language_loss": 0.74766552,
      "learning_rate": 3.895657703291814e-06,
      "loss": 0.76744515,
      "num_input_tokens_seen": 22924100,
      "step": 1083,
      "time_per_iteration": 2.55025053024292
    },
    {
      "auxiliary_loss_clip": 0.01216616,
      "auxiliary_loss_mlp": 0.01035848,
      "balance_loss_clip": 1.06058097,
      "balance_loss_mlp": 1.02440393,
      "epoch": 0.1303432934527746,
      "flos": 21323612920320.0,
      "grad_norm": 2.8579568162587927,
      "language_loss": 0.7955972,
      "learning_rate": 3.895409239200781e-06,
      "loss": 0.81812179,
      "num_input_tokens_seen": 22939985,
      "step": 1084,
      "time_per_iteration": 2.5215818881988525
    },
    {
      "auxiliary_loss_clip": 0.0121806,
      "auxiliary_loss_mlp": 0.01041378,
      "balance_loss_clip": 1.0621314,
      "balance_loss_mlp": 1.02850354,
      "epoch": 0.1304635363434137,
      "flos": 20922490765440.0,
      "grad_norm": 2.390041997093012,
      "language_loss": 0.91477942,
      "learning_rate": 3.895160487577673e-06,
      "loss": 0.93737376,
      "num_input_tokens_seen": 22957555,
      "step": 1085,
      "time_per_iteration": 2.496581554412842
    },
    {
      "auxiliary_loss_clip": 0.01114649,
      "auxiliary_loss_mlp": 0.01003939,
      "balance_loss_clip": 1.02569234,
      "balance_loss_mlp": 1.00112534,
      "epoch": 0.1305837792340528,
      "flos": 63245659080960.0,
      "grad_norm": 0.7858939093723247,
      "language_loss": 0.60890037,
      "learning_rate": 3.894911448460226e-06,
      "loss": 0.63008624,
      "num_input_tokens_seen": 23016870,
      "step": 1086,
      "time_per_iteration": 2.9670746326446533
    },
    {
      "auxiliary_loss_clip": 0.01126529,
      "auxiliary_loss_mlp": 0.01046833,
      "balance_loss_clip": 1.05105567,
      "balance_loss_mlp": 1.03490007,
      "epoch": 0.13070402212469187,
      "flos": 26428852955520.0,
      "grad_norm": 1.9298341026671,
      "language_loss": 0.72878098,
      "learning_rate": 3.8946621218862195e-06,
      "loss": 0.75051457,
      "num_input_tokens_seen": 23037870,
      "step": 1087,
      "time_per_iteration": 3.5803439617156982
    },
    {
      "auxiliary_loss_clip": 0.01189059,
      "auxiliary_loss_mlp": 0.01043075,
      "balance_loss_clip": 1.06005859,
      "balance_loss_mlp": 1.03217936,
      "epoch": 0.13082426501533098,
      "flos": 27673409341440.0,
      "grad_norm": 3.9347373992496233,
      "language_loss": 0.89026272,
      "learning_rate": 3.894412507893475e-06,
      "loss": 0.91258407,
      "num_input_tokens_seen": 23058150,
      "step": 1088,
      "time_per_iteration": 2.6474409103393555
    },
    {
      "auxiliary_loss_clip": 0.01184951,
      "auxiliary_loss_mlp": 0.0104793,
      "balance_loss_clip": 1.05821359,
      "balance_loss_mlp": 1.03605092,
      "epoch": 0.13094450790597006,
      "flos": 24826770547200.0,
      "grad_norm": 2.040015778418802,
      "language_loss": 0.71958792,
      "learning_rate": 3.894162606519859e-06,
      "loss": 0.74191678,
      "num_input_tokens_seen": 23077100,
      "step": 1089,
      "time_per_iteration": 2.671130418777466
    },
    {
      "auxiliary_loss_clip": 0.01176615,
      "auxiliary_loss_mlp": 0.01041977,
      "balance_loss_clip": 1.05913639,
      "balance_loss_mlp": 1.03121877,
      "epoch": 0.13106475079660915,
      "flos": 19062605468160.0,
      "grad_norm": 1.9492241161256092,
      "language_loss": 0.76893365,
      "learning_rate": 3.893912417803282e-06,
      "loss": 0.79111958,
      "num_input_tokens_seen": 23096815,
      "step": 1090,
      "time_per_iteration": 2.6335387229919434
    },
    {
      "auxiliary_loss_clip": 0.01178566,
      "auxiliary_loss_mlp": 0.01042553,
      "balance_loss_clip": 1.05422139,
      "balance_loss_mlp": 1.03019667,
      "epoch": 0.13118499368724823,
      "flos": 28913189218560.0,
      "grad_norm": 2.079338919257438,
      "language_loss": 0.77189744,
      "learning_rate": 3.8936619417816975e-06,
      "loss": 0.79410869,
      "num_input_tokens_seen": 23117145,
      "step": 1091,
      "time_per_iteration": 2.689709424972534
    },
    {
      "auxiliary_loss_clip": 0.01192748,
      "auxiliary_loss_mlp": 0.01033427,
      "balance_loss_clip": 1.06201732,
      "balance_loss_mlp": 1.02257299,
      "epoch": 0.13130523657788734,
      "flos": 14283398206080.0,
      "grad_norm": 1.7588624226971892,
      "language_loss": 0.71889734,
      "learning_rate": 3.8934111784931015e-06,
      "loss": 0.74115902,
      "num_input_tokens_seen": 23134595,
      "step": 1092,
      "time_per_iteration": 2.6658742427825928
    },
    {
      "auxiliary_loss_clip": 0.01104425,
      "auxiliary_loss_mlp": 0.01002692,
      "balance_loss_clip": 1.0229907,
      "balance_loss_mlp": 0.99985522,
      "epoch": 0.13142547946852642,
      "flos": 70174155519360.0,
      "grad_norm": 0.9243659688392305,
      "language_loss": 0.59138483,
      "learning_rate": 3.893160127975535e-06,
      "loss": 0.61245602,
      "num_input_tokens_seen": 23195285,
      "step": 1093,
      "time_per_iteration": 3.997832775115967
    },
    {
      "auxiliary_loss_clip": 0.01180875,
      "auxiliary_loss_mlp": 0.01037954,
      "balance_loss_clip": 1.05656636,
      "balance_loss_mlp": 1.02686143,
      "epoch": 0.1315457223591655,
      "flos": 45805998844800.0,
      "grad_norm": 2.248025414806439,
      "language_loss": 0.81055689,
      "learning_rate": 3.8929087902670826e-06,
      "loss": 0.83274513,
      "num_input_tokens_seen": 23216915,
      "step": 1094,
      "time_per_iteration": 4.473284959793091
    },
    {
      "auxiliary_loss_clip": 0.01115156,
      "auxiliary_loss_mlp": 0.01002481,
      "balance_loss_clip": 1.02234173,
      "balance_loss_mlp": 0.9998461,
      "epoch": 0.13166596524980462,
      "flos": 62881165820160.0,
      "grad_norm": 0.9316042087994315,
      "language_loss": 0.60643041,
      "learning_rate": 3.8926571654058715e-06,
      "loss": 0.62760675,
      "num_input_tokens_seen": 23273560,
      "step": 1095,
      "time_per_iteration": 3.001385450363159
    },
    {
      "auxiliary_loss_clip": 0.01189067,
      "auxiliary_loss_mlp": 0.01036534,
      "balance_loss_clip": 1.06040812,
      "balance_loss_mlp": 1.0253458,
      "epoch": 0.1317862081404437,
      "flos": 23586523793280.0,
      "grad_norm": 2.3690148783628695,
      "language_loss": 0.77153385,
      "learning_rate": 3.892405253430074e-06,
      "loss": 0.79378986,
      "num_input_tokens_seen": 23291080,
      "step": 1096,
      "time_per_iteration": 2.585012435913086
    },
    {
      "auxiliary_loss_clip": 0.01211897,
      "auxiliary_loss_mlp": 0.00766704,
      "balance_loss_clip": 1.06427073,
      "balance_loss_mlp": 1.00037026,
      "epoch": 0.13190645103108278,
      "flos": 20260764460800.0,
      "grad_norm": 2.6380488360395065,
      "language_loss": 0.82375747,
      "learning_rate": 3.892153054377904e-06,
      "loss": 0.84354347,
      "num_input_tokens_seen": 23308485,
      "step": 1097,
      "time_per_iteration": 2.505908250808716
    },
    {
      "auxiliary_loss_clip": 0.01053572,
      "auxiliary_loss_mlp": 0.01006238,
      "balance_loss_clip": 1.02045012,
      "balance_loss_mlp": 1.00341308,
      "epoch": 0.13202669392172187,
      "flos": 53455440136320.0,
      "grad_norm": 0.9438268029637948,
      "language_loss": 0.59427595,
      "learning_rate": 3.891900568287619e-06,
      "loss": 0.61487406,
      "num_input_tokens_seen": 23360870,
      "step": 1098,
      "time_per_iteration": 3.0244534015655518
    },
    {
      "auxiliary_loss_clip": 0.01196647,
      "auxiliary_loss_mlp": 0.01037809,
      "balance_loss_clip": 1.06021881,
      "balance_loss_mlp": 1.02593517,
      "epoch": 0.13214693681236098,
      "flos": 15851293845120.0,
      "grad_norm": 2.7464925446688007,
      "language_loss": 0.72030115,
      "learning_rate": 3.891647795197523e-06,
      "loss": 0.74264568,
      "num_input_tokens_seen": 23376910,
      "step": 1099,
      "time_per_iteration": 2.5363311767578125
    },
    {
      "auxiliary_loss_clip": 0.01199107,
      "auxiliary_loss_mlp": 0.01046583,
      "balance_loss_clip": 1.05809522,
      "balance_loss_mlp": 1.03436375,
      "epoch": 0.13226717970300006,
      "flos": 19353840940800.0,
      "grad_norm": 1.9702353367692258,
      "language_loss": 0.68540627,
      "learning_rate": 3.8913947351459605e-06,
      "loss": 0.70786315,
      "num_input_tokens_seen": 23394450,
      "step": 1100,
      "time_per_iteration": 2.570089817047119
    },
    {
      "auxiliary_loss_clip": 0.01242258,
      "auxiliary_loss_mlp": 0.01038752,
      "balance_loss_clip": 1.06853461,
      "balance_loss_mlp": 1.02806497,
      "epoch": 0.13238742259363914,
      "flos": 20698084546560.0,
      "grad_norm": 1.992411835738522,
      "language_loss": 0.67721313,
      "learning_rate": 3.89114138817132e-06,
      "loss": 0.70002329,
      "num_input_tokens_seen": 23411115,
      "step": 1101,
      "time_per_iteration": 2.4617741107940674
    },
    {
      "auxiliary_loss_clip": 0.0122472,
      "auxiliary_loss_mlp": 0.01033473,
      "balance_loss_clip": 1.06689572,
      "balance_loss_mlp": 1.02222586,
      "epoch": 0.13250766548427825,
      "flos": 21032449274880.0,
      "grad_norm": 1.7244034211692043,
      "language_loss": 0.84244347,
      "learning_rate": 3.890887754312035e-06,
      "loss": 0.8650254,
      "num_input_tokens_seen": 23429360,
      "step": 1102,
      "time_per_iteration": 2.5234038829803467
    },
    {
      "auxiliary_loss_clip": 0.01198335,
      "auxiliary_loss_mlp": 0.01047433,
      "balance_loss_clip": 1.05613685,
      "balance_loss_mlp": 1.03599524,
      "epoch": 0.13262790837491734,
      "flos": 22637871648000.0,
      "grad_norm": 1.8425585182217774,
      "language_loss": 0.87634182,
      "learning_rate": 3.890633833606581e-06,
      "loss": 0.89879954,
      "num_input_tokens_seen": 23449050,
      "step": 1103,
      "time_per_iteration": 2.519550323486328
    },
    {
      "auxiliary_loss_clip": 0.01223589,
      "auxiliary_loss_mlp": 0.01033289,
      "balance_loss_clip": 1.06723905,
      "balance_loss_mlp": 1.0228045,
      "epoch": 0.13274815126555642,
      "flos": 19683141851520.0,
      "grad_norm": 1.8864708105263763,
      "language_loss": 0.69656742,
      "learning_rate": 3.890379626093477e-06,
      "loss": 0.71913624,
      "num_input_tokens_seen": 23468800,
      "step": 1104,
      "time_per_iteration": 2.5130162239074707
    },
    {
      "auxiliary_loss_clip": 0.01164821,
      "auxiliary_loss_mlp": 0.01037858,
      "balance_loss_clip": 1.0563271,
      "balance_loss_mlp": 1.02599669,
      "epoch": 0.1328683941561955,
      "flos": 21317687176320.0,
      "grad_norm": 3.0201325999247755,
      "language_loss": 0.92674553,
      "learning_rate": 3.890125131811287e-06,
      "loss": 0.94877231,
      "num_input_tokens_seen": 23486850,
      "step": 1105,
      "time_per_iteration": 2.5744683742523193
    },
    {
      "auxiliary_loss_clip": 0.01193684,
      "auxiliary_loss_mlp": 0.01037857,
      "balance_loss_clip": 1.055655,
      "balance_loss_mlp": 1.02708602,
      "epoch": 0.1329886370468346,
      "flos": 13699131580800.0,
      "grad_norm": 2.0178759255710434,
      "language_loss": 0.75340331,
      "learning_rate": 3.889870350798618e-06,
      "loss": 0.77571869,
      "num_input_tokens_seen": 23504195,
      "step": 1106,
      "time_per_iteration": 2.5424411296844482
    },
    {
      "auxiliary_loss_clip": 0.01241392,
      "auxiliary_loss_mlp": 0.01038581,
      "balance_loss_clip": 1.06606436,
      "balance_loss_mlp": 1.02779841,
      "epoch": 0.1331088799374737,
      "flos": 21032413361280.0,
      "grad_norm": 1.6954560955448363,
      "language_loss": 0.78544891,
      "learning_rate": 3.889615283094119e-06,
      "loss": 0.80824864,
      "num_input_tokens_seen": 23523385,
      "step": 1107,
      "time_per_iteration": 2.4959022998809814
    },
    {
      "auxiliary_loss_clip": 0.01244503,
      "auxiliary_loss_mlp": 0.01041223,
      "balance_loss_clip": 1.06660509,
      "balance_loss_mlp": 1.02921891,
      "epoch": 0.13322912282811278,
      "flos": 18260432985600.0,
      "grad_norm": 2.1217730320413266,
      "language_loss": 0.84691358,
      "learning_rate": 3.889359928736485e-06,
      "loss": 0.86977088,
      "num_input_tokens_seen": 23541330,
      "step": 1108,
      "time_per_iteration": 2.5023608207702637
    },
    {
      "auxiliary_loss_clip": 0.01201133,
      "auxiliary_loss_mlp": 0.00766386,
      "balance_loss_clip": 1.06272173,
      "balance_loss_mlp": 1.00047123,
      "epoch": 0.1333493657187519,
      "flos": 24460876656000.0,
      "grad_norm": 2.030399835573967,
      "language_loss": 0.90990919,
      "learning_rate": 3.889104287764451e-06,
      "loss": 0.92958438,
      "num_input_tokens_seen": 23561705,
      "step": 1109,
      "time_per_iteration": 2.5452005863189697
    },
    {
      "auxiliary_loss_clip": 0.01208617,
      "auxiliary_loss_mlp": 0.01042573,
      "balance_loss_clip": 1.0652591,
      "balance_loss_mlp": 1.03161192,
      "epoch": 0.13346960860939097,
      "flos": 22158930677760.0,
      "grad_norm": 2.0471571648656246,
      "language_loss": 0.90363342,
      "learning_rate": 3.888848360216798e-06,
      "loss": 0.92614532,
      "num_input_tokens_seen": 23579350,
      "step": 1110,
      "time_per_iteration": 2.536431312561035
    },
    {
      "auxiliary_loss_clip": 0.01103863,
      "auxiliary_loss_mlp": 0.01004289,
      "balance_loss_clip": 1.02094758,
      "balance_loss_mlp": 1.00163019,
      "epoch": 0.13358985150003005,
      "flos": 67931212608000.0,
      "grad_norm": 0.7998990758872707,
      "language_loss": 0.5653863,
      "learning_rate": 3.888592146132351e-06,
      "loss": 0.58646786,
      "num_input_tokens_seen": 23640620,
      "step": 1111,
      "time_per_iteration": 3.199981689453125
    },
    {
      "auxiliary_loss_clip": 0.01224742,
      "auxiliary_loss_mlp": 0.01043393,
      "balance_loss_clip": 1.06680393,
      "balance_loss_mlp": 1.03215134,
      "epoch": 0.13371009439066917,
      "flos": 26834284742400.0,
      "grad_norm": 2.09199863814541,
      "language_loss": 0.78491414,
      "learning_rate": 3.888335645549978e-06,
      "loss": 0.80759543,
      "num_input_tokens_seen": 23661040,
      "step": 1112,
      "time_per_iteration": 2.5374910831451416
    },
    {
      "auxiliary_loss_clip": 0.01242404,
      "auxiliary_loss_mlp": 0.0104559,
      "balance_loss_clip": 1.07057321,
      "balance_loss_mlp": 1.03442645,
      "epoch": 0.13383033728130825,
      "flos": 26322844942080.0,
      "grad_norm": 2.8271973731830724,
      "language_loss": 0.81292695,
      "learning_rate": 3.888078858508588e-06,
      "loss": 0.83580691,
      "num_input_tokens_seen": 23680900,
      "step": 1113,
      "time_per_iteration": 2.512892246246338
    },
    {
      "auxiliary_loss_clip": 0.01209126,
      "auxiliary_loss_mlp": 0.01037102,
      "balance_loss_clip": 1.06670332,
      "balance_loss_mlp": 1.02578866,
      "epoch": 0.13395058017194733,
      "flos": 22563931501440.0,
      "grad_norm": 1.9607594972561997,
      "language_loss": 0.84491694,
      "learning_rate": 3.8878217850471365e-06,
      "loss": 0.86737925,
      "num_input_tokens_seen": 23700815,
      "step": 1114,
      "time_per_iteration": 3.590632200241089
    },
    {
      "auxiliary_loss_clip": 0.01244709,
      "auxiliary_loss_mlp": 0.0104511,
      "balance_loss_clip": 1.07039833,
      "balance_loss_mlp": 1.03270042,
      "epoch": 0.13407082306258641,
      "flos": 25810938264960.0,
      "grad_norm": 1.7815688823615061,
      "language_loss": 0.74153411,
      "learning_rate": 3.887564425204621e-06,
      "loss": 0.76443231,
      "num_input_tokens_seen": 23722500,
      "step": 1115,
      "time_per_iteration": 2.5157196521759033
    },
    {
      "auxiliary_loss_clip": 0.01082572,
      "auxiliary_loss_mlp": 0.01002906,
      "balance_loss_clip": 1.02276957,
      "balance_loss_mlp": 1.00024784,
      "epoch": 0.13419106595322552,
      "flos": 68338365269760.0,
      "grad_norm": 0.8454408615764806,
      "language_loss": 0.54641867,
      "learning_rate": 3.887306779020083e-06,
      "loss": 0.56727344,
      "num_input_tokens_seen": 23777155,
      "step": 1116,
      "time_per_iteration": 3.055781602859497
    },
    {
      "auxiliary_loss_clip": 0.01229169,
      "auxiliary_loss_mlp": 0.01043548,
      "balance_loss_clip": 1.06806564,
      "balance_loss_mlp": 1.03185344,
      "epoch": 0.1343113088438646,
      "flos": 20449080489600.0,
      "grad_norm": 2.609570743827039,
      "language_loss": 0.7029599,
      "learning_rate": 3.887048846532608e-06,
      "loss": 0.72568709,
      "num_input_tokens_seen": 23794130,
      "step": 1117,
      "time_per_iteration": 2.49936842918396
    },
    {
      "auxiliary_loss_clip": 0.0108691,
      "auxiliary_loss_mlp": 0.01003492,
      "balance_loss_clip": 1.01915431,
      "balance_loss_mlp": 1.00084519,
      "epoch": 0.1344315517345037,
      "flos": 67389784951680.0,
      "grad_norm": 0.7589417508545027,
      "language_loss": 0.58111024,
      "learning_rate": 3.8867906277813224e-06,
      "loss": 0.6020143,
      "num_input_tokens_seen": 23852285,
      "step": 1118,
      "time_per_iteration": 3.0013554096221924
    },
    {
      "auxiliary_loss_clip": 0.01226343,
      "auxiliary_loss_mlp": 0.00766284,
      "balance_loss_clip": 1.06450415,
      "balance_loss_mlp": 1.00029206,
      "epoch": 0.1345517946251428,
      "flos": 40734442788480.0,
      "grad_norm": 3.41867014912935,
      "language_loss": 0.74377131,
      "learning_rate": 3.886532122805399e-06,
      "loss": 0.76369756,
      "num_input_tokens_seen": 23874765,
      "step": 1119,
      "time_per_iteration": 3.567920207977295
    },
    {
      "auxiliary_loss_clip": 0.0114868,
      "auxiliary_loss_mlp": 0.01045096,
      "balance_loss_clip": 1.05268717,
      "balance_loss_mlp": 1.03314567,
      "epoch": 0.13467203751578188,
      "flos": 22816850140800.0,
      "grad_norm": 1.6925202653526292,
      "language_loss": 0.89782441,
      "learning_rate": 3.886273331644053e-06,
      "loss": 0.91976219,
      "num_input_tokens_seen": 23893635,
      "step": 1120,
      "time_per_iteration": 3.566516637802124
    },
    {
      "auxiliary_loss_clip": 0.01174579,
      "auxiliary_loss_mlp": 0.01035851,
      "balance_loss_clip": 1.06019986,
      "balance_loss_mlp": 1.02468705,
      "epoch": 0.13479228040642097,
      "flos": 17091576512640.0,
      "grad_norm": 2.018650031964999,
      "language_loss": 0.82421106,
      "learning_rate": 3.886014254336542e-06,
      "loss": 0.84631538,
      "num_input_tokens_seen": 23910110,
      "step": 1121,
      "time_per_iteration": 2.569540500640869
    },
    {
      "auxiliary_loss_clip": 0.0122203,
      "auxiliary_loss_mlp": 0.01035701,
      "balance_loss_clip": 1.06409669,
      "balance_loss_mlp": 1.02470994,
      "epoch": 0.13491252329706005,
      "flos": 23730525417600.0,
      "grad_norm": 1.810207558557642,
      "language_loss": 0.92445028,
      "learning_rate": 3.885754890922168e-06,
      "loss": 0.94702762,
      "num_input_tokens_seen": 23930440,
      "step": 1122,
      "time_per_iteration": 2.5304903984069824
    },
    {
      "auxiliary_loss_clip": 0.01131994,
      "auxiliary_loss_mlp": 0.01047555,
      "balance_loss_clip": 1.05270624,
      "balance_loss_mlp": 1.03595567,
      "epoch": 0.13503276618769916,
      "flos": 34127058960000.0,
      "grad_norm": 1.9392186077251008,
      "language_loss": 0.78558898,
      "learning_rate": 3.885495241440277e-06,
      "loss": 0.80738449,
      "num_input_tokens_seen": 23954535,
      "step": 1123,
      "time_per_iteration": 2.7606773376464844
    },
    {
      "auxiliary_loss_clip": 0.01242893,
      "auxiliary_loss_mlp": 0.01043186,
      "balance_loss_clip": 1.06803179,
      "balance_loss_mlp": 1.03191471,
      "epoch": 0.13515300907833824,
      "flos": 17712328377600.0,
      "grad_norm": 1.8355905977515146,
      "language_loss": 0.74198043,
      "learning_rate": 3.885235305930257e-06,
      "loss": 0.7648412,
      "num_input_tokens_seen": 23972735,
      "step": 1124,
      "time_per_iteration": 2.4871087074279785
    },
    {
      "auxiliary_loss_clip": 0.01190885,
      "auxiliary_loss_mlp": 0.01048269,
      "balance_loss_clip": 1.06436086,
      "balance_loss_mlp": 1.0352211,
      "epoch": 0.13527325196897733,
      "flos": 20260872201600.0,
      "grad_norm": 1.847931973002981,
      "language_loss": 0.85501164,
      "learning_rate": 3.884975084431539e-06,
      "loss": 0.87740314,
      "num_input_tokens_seen": 23987685,
      "step": 1125,
      "time_per_iteration": 2.5507395267486572
    },
    {
      "auxiliary_loss_clip": 0.01215421,
      "auxiliary_loss_mlp": 0.00766433,
      "balance_loss_clip": 1.06335557,
      "balance_loss_mlp": 1.00042605,
      "epoch": 0.13539349485961644,
      "flos": 18186492839040.0,
      "grad_norm": 2.2416136256575228,
      "language_loss": 0.91655236,
      "learning_rate": 3.8847145769836e-06,
      "loss": 0.93637091,
      "num_input_tokens_seen": 24004105,
      "step": 1126,
      "time_per_iteration": 2.515225648880005
    },
    {
      "auxiliary_loss_clip": 0.01242819,
      "auxiliary_loss_mlp": 0.01041202,
      "balance_loss_clip": 1.06684661,
      "balance_loss_mlp": 1.02956116,
      "epoch": 0.13551373775025552,
      "flos": 19317463441920.0,
      "grad_norm": 2.386962710680505,
      "language_loss": 0.66415197,
      "learning_rate": 3.884453783625959e-06,
      "loss": 0.68699217,
      "num_input_tokens_seen": 24021715,
      "step": 1127,
      "time_per_iteration": 2.482673168182373
    },
    {
      "auxiliary_loss_clip": 0.01203213,
      "auxiliary_loss_mlp": 0.01035733,
      "balance_loss_clip": 1.06316209,
      "balance_loss_mlp": 1.02518249,
      "epoch": 0.1356339806408946,
      "flos": 20850813175680.0,
      "grad_norm": 2.1108926415072697,
      "language_loss": 0.84781194,
      "learning_rate": 3.884192704398176e-06,
      "loss": 0.87020141,
      "num_input_tokens_seen": 24038915,
      "step": 1128,
      "time_per_iteration": 2.531822443008423
    },
    {
      "auxiliary_loss_clip": 0.012242,
      "auxiliary_loss_mlp": 0.01051782,
      "balance_loss_clip": 1.06316972,
      "balance_loss_mlp": 1.04063559,
      "epoch": 0.13575422353153369,
      "flos": 50476037696640.0,
      "grad_norm": 1.6739670975124537,
      "language_loss": 0.74570715,
      "learning_rate": 3.883931339339858e-06,
      "loss": 0.76846701,
      "num_input_tokens_seen": 24063300,
      "step": 1129,
      "time_per_iteration": 2.761939287185669
    },
    {
      "auxiliary_loss_clip": 0.01227772,
      "auxiliary_loss_mlp": 0.01039168,
      "balance_loss_clip": 1.06450844,
      "balance_loss_mlp": 1.02742577,
      "epoch": 0.1358744664221728,
      "flos": 18150797698560.0,
      "grad_norm": 2.4110153977600044,
      "language_loss": 0.78667426,
      "learning_rate": 3.883669688490654e-06,
      "loss": 0.8093437,
      "num_input_tokens_seen": 24081070,
      "step": 1130,
      "time_per_iteration": 2.5068349838256836
    },
    {
      "auxiliary_loss_clip": 0.01195893,
      "auxiliary_loss_mlp": 0.00766058,
      "balance_loss_clip": 1.05866027,
      "balance_loss_mlp": 1.00041699,
      "epoch": 0.13599470931281188,
      "flos": 18442966924800.0,
      "grad_norm": 1.9424638081951011,
      "language_loss": 0.85507464,
      "learning_rate": 3.883407751890256e-06,
      "loss": 0.87469411,
      "num_input_tokens_seen": 24099675,
      "step": 1131,
      "time_per_iteration": 2.5076582431793213
    },
    {
      "auxiliary_loss_clip": 0.01191876,
      "auxiliary_loss_mlp": 0.01048915,
      "balance_loss_clip": 1.05910599,
      "balance_loss_mlp": 1.03619492,
      "epoch": 0.13611495220345096,
      "flos": 26680766014080.0,
      "grad_norm": 1.7181343736842447,
      "language_loss": 0.85642576,
      "learning_rate": 3.8831455295783994e-06,
      "loss": 0.87883371,
      "num_input_tokens_seen": 24118925,
      "step": 1132,
      "time_per_iteration": 2.627047061920166
    },
    {
      "auxiliary_loss_clip": 0.01203861,
      "auxiliary_loss_mlp": 0.0104093,
      "balance_loss_clip": 1.06280267,
      "balance_loss_mlp": 1.02939045,
      "epoch": 0.13623519509409007,
      "flos": 21686238673920.0,
      "grad_norm": 1.643416965574082,
      "language_loss": 0.73959291,
      "learning_rate": 3.882883021594864e-06,
      "loss": 0.76204079,
      "num_input_tokens_seen": 24137065,
      "step": 1133,
      "time_per_iteration": 2.579265594482422
    },
    {
      "auxiliary_loss_clip": 0.01181733,
      "auxiliary_loss_mlp": 0.01036871,
      "balance_loss_clip": 1.06068063,
      "balance_loss_mlp": 1.02583218,
      "epoch": 0.13635543798472916,
      "flos": 14830389492480.0,
      "grad_norm": 2.4077745151837826,
      "language_loss": 0.86901402,
      "learning_rate": 3.8826202279794705e-06,
      "loss": 0.89120007,
      "num_input_tokens_seen": 24154125,
      "step": 1134,
      "time_per_iteration": 2.5350449085235596
    },
    {
      "auxiliary_loss_clip": 0.0124275,
      "auxiliary_loss_mlp": 0.01039736,
      "balance_loss_clip": 1.06911612,
      "balance_loss_mlp": 1.0288341,
      "epoch": 0.13647568087536824,
      "flos": 22890323410560.0,
      "grad_norm": 1.9634094323757916,
      "language_loss": 0.70318204,
      "learning_rate": 3.882357148772085e-06,
      "loss": 0.72600693,
      "num_input_tokens_seen": 24171550,
      "step": 1135,
      "time_per_iteration": 2.4762027263641357
    },
    {
      "auxiliary_loss_clip": 0.01175443,
      "auxiliary_loss_mlp": 0.01045917,
      "balance_loss_clip": 1.05802798,
      "balance_loss_mlp": 1.03416872,
      "epoch": 0.13659592376600732,
      "flos": 19937927998080.0,
      "grad_norm": 2.7922594386765605,
      "language_loss": 0.84492689,
      "learning_rate": 3.882093784012617e-06,
      "loss": 0.86714047,
      "num_input_tokens_seen": 24190190,
      "step": 1136,
      "time_per_iteration": 2.534083127975464
    },
    {
      "auxiliary_loss_clip": 0.01205692,
      "auxiliary_loss_mlp": 0.0103658,
      "balance_loss_clip": 1.06324697,
      "balance_loss_mlp": 1.02511835,
      "epoch": 0.13671616665664643,
      "flos": 21428579439360.0,
      "grad_norm": 1.9171071513542501,
      "language_loss": 0.84301311,
      "learning_rate": 3.881830133741019e-06,
      "loss": 0.86543584,
      "num_input_tokens_seen": 24209055,
      "step": 1137,
      "time_per_iteration": 2.5401196479797363
    },
    {
      "auxiliary_loss_clip": 0.01190995,
      "auxiliary_loss_mlp": 0.010478,
      "balance_loss_clip": 1.06460023,
      "balance_loss_mlp": 1.03609347,
      "epoch": 0.13683640954728551,
      "flos": 22778138257920.0,
      "grad_norm": 1.9410813736814505,
      "language_loss": 0.76070166,
      "learning_rate": 3.881566197997285e-06,
      "loss": 0.78308958,
      "num_input_tokens_seen": 24225490,
      "step": 1138,
      "time_per_iteration": 2.585272789001465
    },
    {
      "auxiliary_loss_clip": 0.01204776,
      "auxiliary_loss_mlp": 0.01039517,
      "balance_loss_clip": 1.06676984,
      "balance_loss_mlp": 1.02851439,
      "epoch": 0.1369566524379246,
      "flos": 21725884310400.0,
      "grad_norm": 1.5339118104084037,
      "language_loss": 0.75060034,
      "learning_rate": 3.881301976821456e-06,
      "loss": 0.77304322,
      "num_input_tokens_seen": 24245520,
      "step": 1139,
      "time_per_iteration": 2.5421037673950195
    },
    {
      "auxiliary_loss_clip": 0.0122006,
      "auxiliary_loss_mlp": 0.0104191,
      "balance_loss_clip": 1.06502652,
      "balance_loss_mlp": 1.03129435,
      "epoch": 0.1370768953285637,
      "flos": 18624459369600.0,
      "grad_norm": 1.8970164879730396,
      "language_loss": 0.90650415,
      "learning_rate": 3.881037470253612e-06,
      "loss": 0.92912388,
      "num_input_tokens_seen": 24265035,
      "step": 1140,
      "time_per_iteration": 3.3628947734832764
    },
    {
      "auxiliary_loss_clip": 0.01175835,
      "auxiliary_loss_mlp": 0.01040442,
      "balance_loss_clip": 1.06009603,
      "balance_loss_mlp": 1.02960598,
      "epoch": 0.1371971382192028,
      "flos": 14939521989120.0,
      "grad_norm": 2.6303102786651906,
      "language_loss": 0.79046839,
      "learning_rate": 3.88077267833388e-06,
      "loss": 0.81263113,
      "num_input_tokens_seen": 24281550,
      "step": 1141,
      "time_per_iteration": 2.544905424118042
    },
    {
      "auxiliary_loss_clip": 0.0117028,
      "auxiliary_loss_mlp": 0.01044457,
      "balance_loss_clip": 1.05718565,
      "balance_loss_mlp": 1.0331862,
      "epoch": 0.13731738110984187,
      "flos": 19023785844480.0,
      "grad_norm": 2.096575567553329,
      "language_loss": 0.83701593,
      "learning_rate": 3.880507601102427e-06,
      "loss": 0.85916328,
      "num_input_tokens_seen": 24299485,
      "step": 1142,
      "time_per_iteration": 2.6105563640594482
    },
    {
      "auxiliary_loss_clip": 0.01239174,
      "auxiliary_loss_mlp": 0.01047938,
      "balance_loss_clip": 1.06891823,
      "balance_loss_mlp": 1.03702462,
      "epoch": 0.13743762400048098,
      "flos": 18187462506240.0,
      "grad_norm": 2.8469880353884602,
      "language_loss": 0.82199681,
      "learning_rate": 3.880242238599467e-06,
      "loss": 0.84486794,
      "num_input_tokens_seen": 24316010,
      "step": 1143,
      "time_per_iteration": 2.445014715194702
    },
    {
      "auxiliary_loss_clip": 0.01234912,
      "auxiliary_loss_mlp": 0.01047229,
      "balance_loss_clip": 1.06560016,
      "balance_loss_mlp": 1.03576708,
      "epoch": 0.13755786689112007,
      "flos": 21031982398080.0,
      "grad_norm": 1.8033794513004204,
      "language_loss": 0.83189905,
      "learning_rate": 3.879976590865254e-06,
      "loss": 0.85472047,
      "num_input_tokens_seen": 24335465,
      "step": 1144,
      "time_per_iteration": 2.5212836265563965
    },
    {
      "auxiliary_loss_clip": 0.01207388,
      "auxiliary_loss_mlp": 0.01045335,
      "balance_loss_clip": 1.06589866,
      "balance_loss_mlp": 1.03406358,
      "epoch": 0.13767810978175915,
      "flos": 21360636864000.0,
      "grad_norm": 1.8640830388365308,
      "language_loss": 0.87317169,
      "learning_rate": 3.879710657940087e-06,
      "loss": 0.8956989,
      "num_input_tokens_seen": 24354415,
      "step": 1145,
      "time_per_iteration": 2.547109842300415
    },
    {
      "auxiliary_loss_clip": 0.01226205,
      "auxiliary_loss_mlp": 0.01054492,
      "balance_loss_clip": 1.06520343,
      "balance_loss_mlp": 1.04227889,
      "epoch": 0.13779835267239823,
      "flos": 30592084861440.0,
      "grad_norm": 2.015525290913156,
      "language_loss": 0.70455736,
      "learning_rate": 3.879444439864308e-06,
      "loss": 0.7273643,
      "num_input_tokens_seen": 24373990,
      "step": 1146,
      "time_per_iteration": 3.3156747817993164
    },
    {
      "auxiliary_loss_clip": 0.01221469,
      "auxiliary_loss_mlp": 0.00766313,
      "balance_loss_clip": 1.06280899,
      "balance_loss_mlp": 1.00066495,
      "epoch": 0.13791859556303734,
      "flos": 22669867687680.0,
      "grad_norm": 1.9964562402872512,
      "language_loss": 0.86180943,
      "learning_rate": 3.879177936678301e-06,
      "loss": 0.88168716,
      "num_input_tokens_seen": 24392995,
      "step": 1147,
      "time_per_iteration": 4.178260564804077
    },
    {
      "auxiliary_loss_clip": 0.01211583,
      "auxiliary_loss_mlp": 0.01043046,
      "balance_loss_clip": 1.06459701,
      "balance_loss_mlp": 1.03132725,
      "epoch": 0.13803883845367643,
      "flos": 35224166016000.0,
      "grad_norm": 2.2988233421089848,
      "language_loss": 0.76987356,
      "learning_rate": 3.878911148422496e-06,
      "loss": 0.79241979,
      "num_input_tokens_seen": 24414470,
      "step": 1148,
      "time_per_iteration": 2.643392324447632
    },
    {
      "auxiliary_loss_clip": 0.01222322,
      "auxiliary_loss_mlp": 0.01041263,
      "balance_loss_clip": 1.06442523,
      "balance_loss_mlp": 1.02956891,
      "epoch": 0.1381590813443155,
      "flos": 32014542332160.0,
      "grad_norm": 2.189990400431333,
      "language_loss": 0.70677912,
      "learning_rate": 3.878644075137364e-06,
      "loss": 0.72941506,
      "num_input_tokens_seen": 24435120,
      "step": 1149,
      "time_per_iteration": 2.5886855125427246
    },
    {
      "auxiliary_loss_clip": 0.0116837,
      "auxiliary_loss_mlp": 0.01037692,
      "balance_loss_clip": 1.0549655,
      "balance_loss_mlp": 1.02666509,
      "epoch": 0.13827932423495462,
      "flos": 17821855923840.0,
      "grad_norm": 2.1550576330570514,
      "language_loss": 0.79334706,
      "learning_rate": 3.878376716863418e-06,
      "loss": 0.81540769,
      "num_input_tokens_seen": 24451420,
      "step": 1150,
      "time_per_iteration": 2.51723051071167
    },
    {
      "auxiliary_loss_clip": 0.01202777,
      "auxiliary_loss_mlp": 0.01045085,
      "balance_loss_clip": 1.05982995,
      "balance_loss_mlp": 1.03306866,
      "epoch": 0.1383995671255937,
      "flos": 19427098728960.0,
      "grad_norm": 5.871303816597254,
      "language_loss": 0.71790552,
      "learning_rate": 3.878109073641219e-06,
      "loss": 0.74038422,
      "num_input_tokens_seen": 24470450,
      "step": 1151,
      "time_per_iteration": 2.5488815307617188
    },
    {
      "auxiliary_loss_clip": 0.01172043,
      "auxiliary_loss_mlp": 0.010417,
      "balance_loss_clip": 1.05994558,
      "balance_loss_mlp": 1.03064334,
      "epoch": 0.13851981001623279,
      "flos": 28296603331200.0,
      "grad_norm": 1.5508358547550867,
      "language_loss": 0.81265759,
      "learning_rate": 3.877841145511366e-06,
      "loss": 0.834795,
      "num_input_tokens_seen": 24493190,
      "step": 1152,
      "time_per_iteration": 2.6709396839141846
    },
    {
      "auxiliary_loss_clip": 0.01226703,
      "auxiliary_loss_mlp": 0.01041941,
      "balance_loss_clip": 1.06567085,
      "balance_loss_mlp": 1.03045535,
      "epoch": 0.13864005290687187,
      "flos": 21213079793280.0,
      "grad_norm": 1.8762178696647758,
      "language_loss": 0.82642603,
      "learning_rate": 3.8775729325145035e-06,
      "loss": 0.84911251,
      "num_input_tokens_seen": 24512425,
      "step": 1153,
      "time_per_iteration": 2.5234885215759277
    },
    {
      "auxiliary_loss_clip": 0.01073429,
      "auxiliary_loss_mlp": 0.01012187,
      "balance_loss_clip": 1.01950455,
      "balance_loss_mlp": 1.00952899,
      "epoch": 0.13876029579751098,
      "flos": 71653389413760.0,
      "grad_norm": 0.7938055868828094,
      "language_loss": 0.6472044,
      "learning_rate": 3.877304434691321e-06,
      "loss": 0.66806054,
      "num_input_tokens_seen": 24579275,
      "step": 1154,
      "time_per_iteration": 3.2339136600494385
    },
    {
      "auxiliary_loss_clip": 0.01188929,
      "auxiliary_loss_mlp": 0.01032813,
      "balance_loss_clip": 1.06273544,
      "balance_loss_mlp": 1.02256095,
      "epoch": 0.13888053868815006,
      "flos": 21941348042880.0,
      "grad_norm": 1.6842712277662588,
      "language_loss": 0.79914737,
      "learning_rate": 3.877035652082548e-06,
      "loss": 0.82136476,
      "num_input_tokens_seen": 24598720,
      "step": 1155,
      "time_per_iteration": 2.562546491622925
    },
    {
      "auxiliary_loss_clip": 0.01196628,
      "auxiliary_loss_mlp": 0.01036279,
      "balance_loss_clip": 1.06213915,
      "balance_loss_mlp": 1.02471614,
      "epoch": 0.13900078157878915,
      "flos": 19608627087360.0,
      "grad_norm": 1.9114433231332668,
      "language_loss": 0.85481822,
      "learning_rate": 3.87676658472896e-06,
      "loss": 0.87714732,
      "num_input_tokens_seen": 24617530,
      "step": 1156,
      "time_per_iteration": 2.5305514335632324
    },
    {
      "auxiliary_loss_clip": 0.01220095,
      "auxiliary_loss_mlp": 0.0104503,
      "balance_loss_clip": 1.0616771,
      "balance_loss_mlp": 1.03439069,
      "epoch": 0.13912102446942826,
      "flos": 22638051216000.0,
      "grad_norm": 2.0059744219112257,
      "language_loss": 0.85106218,
      "learning_rate": 3.876497232671372e-06,
      "loss": 0.87371337,
      "num_input_tokens_seen": 24637485,
      "step": 1157,
      "time_per_iteration": 2.489564895629883
    },
    {
      "auxiliary_loss_clip": 0.0117955,
      "auxiliary_loss_mlp": 0.01039531,
      "balance_loss_clip": 1.05941677,
      "balance_loss_mlp": 1.02901089,
      "epoch": 0.13924126736006734,
      "flos": 29643324975360.0,
      "grad_norm": 1.9446276015636033,
      "language_loss": 0.83603609,
      "learning_rate": 3.876227595950647e-06,
      "loss": 0.8582269,
      "num_input_tokens_seen": 24656915,
      "step": 1158,
      "time_per_iteration": 2.6752874851226807
    },
    {
      "auxiliary_loss_clip": 0.01238211,
      "auxiliary_loss_mlp": 0.01041393,
      "balance_loss_clip": 1.06763482,
      "balance_loss_mlp": 1.02937031,
      "epoch": 0.13936151025070642,
      "flos": 27417653527680.0,
      "grad_norm": 1.6112974470249788,
      "language_loss": 0.79248452,
      "learning_rate": 3.875957674607686e-06,
      "loss": 0.81528056,
      "num_input_tokens_seen": 24679190,
      "step": 1159,
      "time_per_iteration": 2.5316503047943115
    },
    {
      "auxiliary_loss_clip": 0.01212441,
      "auxiliary_loss_mlp": 0.00766559,
      "balance_loss_clip": 1.05945969,
      "balance_loss_mlp": 1.00070596,
      "epoch": 0.1394817531413455,
      "flos": 16399326625920.0,
      "grad_norm": 2.745152521557462,
      "language_loss": 0.88152218,
      "learning_rate": 3.8756874686834386e-06,
      "loss": 0.90131223,
      "num_input_tokens_seen": 24697405,
      "step": 1160,
      "time_per_iteration": 2.4753894805908203
    },
    {
      "auxiliary_loss_clip": 0.01224718,
      "auxiliary_loss_mlp": 0.00766718,
      "balance_loss_clip": 1.06229985,
      "balance_loss_mlp": 1.00059319,
      "epoch": 0.13960199603198462,
      "flos": 30922319525760.0,
      "grad_norm": 1.5296851091031625,
      "language_loss": 0.80769718,
      "learning_rate": 3.875416978218893e-06,
      "loss": 0.82761157,
      "num_input_tokens_seen": 24720600,
      "step": 1161,
      "time_per_iteration": 2.580864191055298
    },
    {
      "auxiliary_loss_clip": 0.01198174,
      "auxiliary_loss_mlp": 0.01040919,
      "balance_loss_clip": 1.05661917,
      "balance_loss_mlp": 1.0294807,
      "epoch": 0.1397222389226237,
      "flos": 18113773754880.0,
      "grad_norm": 13.512801540782924,
      "language_loss": 0.82489759,
      "learning_rate": 3.8751462032550835e-06,
      "loss": 0.84728849,
      "num_input_tokens_seen": 24737605,
      "step": 1162,
      "time_per_iteration": 2.534541606903076
    },
    {
      "auxiliary_loss_clip": 0.01202995,
      "auxiliary_loss_mlp": 0.01027993,
      "balance_loss_clip": 1.06549644,
      "balance_loss_mlp": 1.01723433,
      "epoch": 0.13984248181326278,
      "flos": 16872772815360.0,
      "grad_norm": 2.5041874374000392,
      "language_loss": 0.82664227,
      "learning_rate": 3.874875143833085e-06,
      "loss": 0.84895217,
      "num_input_tokens_seen": 24755845,
      "step": 1163,
      "time_per_iteration": 2.5094411373138428
    },
    {
      "auxiliary_loss_clip": 0.0122095,
      "auxiliary_loss_mlp": 0.01047486,
      "balance_loss_clip": 1.06333816,
      "balance_loss_mlp": 1.03568423,
      "epoch": 0.1399627247039019,
      "flos": 54121401267840.0,
      "grad_norm": 1.809441658863343,
      "language_loss": 0.68863541,
      "learning_rate": 3.874603799994019e-06,
      "loss": 0.7113198,
      "num_input_tokens_seen": 24779380,
      "step": 1164,
      "time_per_iteration": 2.797175168991089
    },
    {
      "auxiliary_loss_clip": 0.01182786,
      "auxiliary_loss_mlp": 0.01036673,
      "balance_loss_clip": 1.05867195,
      "balance_loss_mlp": 1.02638566,
      "epoch": 0.14008296759454097,
      "flos": 11765521618560.0,
      "grad_norm": 2.0568116690227636,
      "language_loss": 0.87022305,
      "learning_rate": 3.874332171779046e-06,
      "loss": 0.89241767,
      "num_input_tokens_seen": 24794260,
      "step": 1165,
      "time_per_iteration": 2.568225145339966
    },
    {
      "auxiliary_loss_clip": 0.0118308,
      "auxiliary_loss_mlp": 0.0103346,
      "balance_loss_clip": 1.05651498,
      "balance_loss_mlp": 1.02255273,
      "epoch": 0.14020321048518006,
      "flos": 22017514832640.0,
      "grad_norm": 1.6720728256415833,
      "language_loss": 0.75627828,
      "learning_rate": 3.874060259229373e-06,
      "loss": 0.77844369,
      "num_input_tokens_seen": 24815835,
      "step": 1166,
      "time_per_iteration": 2.6282758712768555
    },
    {
      "auxiliary_loss_clip": 0.01224605,
      "auxiliary_loss_mlp": 0.0104494,
      "balance_loss_clip": 1.06599569,
      "balance_loss_mlp": 1.03320432,
      "epoch": 0.14032345337581917,
      "flos": 23404313076480.0,
      "grad_norm": 2.1005222844913387,
      "language_loss": 0.93716538,
      "learning_rate": 3.873788062386249e-06,
      "loss": 0.9598608,
      "num_input_tokens_seen": 24834095,
      "step": 1167,
      "time_per_iteration": 3.344567060470581
    },
    {
      "auxiliary_loss_clip": 0.01193847,
      "auxiliary_loss_mlp": 0.01042578,
      "balance_loss_clip": 1.06391454,
      "balance_loss_mlp": 1.03195691,
      "epoch": 0.14044369626645825,
      "flos": 29645767100160.0,
      "grad_norm": 1.9557510903710569,
      "language_loss": 0.82077014,
      "learning_rate": 3.873515581290965e-06,
      "loss": 0.8431344,
      "num_input_tokens_seen": 24858900,
      "step": 1168,
      "time_per_iteration": 2.672848701477051
    },
    {
      "auxiliary_loss_clip": 0.01191536,
      "auxiliary_loss_mlp": 0.01039288,
      "balance_loss_clip": 1.0639782,
      "balance_loss_mlp": 1.02794576,
      "epoch": 0.14056393915709733,
      "flos": 18332972501760.0,
      "grad_norm": 10.310937155097955,
      "language_loss": 0.75927639,
      "learning_rate": 3.8732428159848575e-06,
      "loss": 0.78158462,
      "num_input_tokens_seen": 24877875,
      "step": 1169,
      "time_per_iteration": 2.574352502822876
    },
    {
      "auxiliary_loss_clip": 0.01222151,
      "auxiliary_loss_mlp": 0.0103911,
      "balance_loss_clip": 1.0677588,
      "balance_loss_mlp": 1.02782106,
      "epoch": 0.14068418204773642,
      "flos": 26687517770880.0,
      "grad_norm": 1.914566701977521,
      "language_loss": 0.78181374,
      "learning_rate": 3.872969766509304e-06,
      "loss": 0.80442631,
      "num_input_tokens_seen": 24898430,
      "step": 1170,
      "time_per_iteration": 2.5370090007781982
    },
    {
      "auxiliary_loss_clip": 0.01078067,
      "auxiliary_loss_mlp": 0.01003821,
      "balance_loss_clip": 1.02103245,
      "balance_loss_mlp": 1.00079286,
      "epoch": 0.14080442493837553,
      "flos": 65259314501760.0,
      "grad_norm": 0.7658627073932307,
      "language_loss": 0.55671537,
      "learning_rate": 3.872696432905726e-06,
      "loss": 0.5775342,
      "num_input_tokens_seen": 24959250,
      "step": 1171,
      "time_per_iteration": 3.1267569065093994
    },
    {
      "auxiliary_loss_clip": 0.01222473,
      "auxiliary_loss_mlp": 0.01043889,
      "balance_loss_clip": 1.06142688,
      "balance_loss_mlp": 1.03257596,
      "epoch": 0.1409246678290146,
      "flos": 25776715582080.0,
      "grad_norm": 5.546998461052881,
      "language_loss": 0.71459758,
      "learning_rate": 3.872422815215589e-06,
      "loss": 0.73726118,
      "num_input_tokens_seen": 24978330,
      "step": 1172,
      "time_per_iteration": 2.5472028255462646
    },
    {
      "auxiliary_loss_clip": 0.01215745,
      "auxiliary_loss_mlp": 0.01044703,
      "balance_loss_clip": 1.05991089,
      "balance_loss_mlp": 1.03180456,
      "epoch": 0.1410449107196537,
      "flos": 21868521217920.0,
      "grad_norm": 1.7228580108787452,
      "language_loss": 0.74132156,
      "learning_rate": 3.8721489134803994e-06,
      "loss": 0.76392603,
      "num_input_tokens_seen": 24997120,
      "step": 1173,
      "time_per_iteration": 3.3893330097198486
    },
    {
      "auxiliary_loss_clip": 0.01218674,
      "auxiliary_loss_mlp": 0.0104568,
      "balance_loss_clip": 1.06441677,
      "balance_loss_mlp": 1.0335927,
      "epoch": 0.1411651536102928,
      "flos": 16684133564160.0,
      "grad_norm": 2.4506299516665164,
      "language_loss": 0.72398233,
      "learning_rate": 3.871874727741707e-06,
      "loss": 0.74662584,
      "num_input_tokens_seen": 25014350,
      "step": 1174,
      "time_per_iteration": 4.179301023483276
    },
    {
      "auxiliary_loss_clip": 0.0121732,
      "auxiliary_loss_mlp": 0.01039841,
      "balance_loss_clip": 1.06673551,
      "balance_loss_mlp": 1.02944636,
      "epoch": 0.1412853965009319,
      "flos": 20992264934400.0,
      "grad_norm": 2.0172332081482733,
      "language_loss": 0.96609676,
      "learning_rate": 3.871600258041108e-06,
      "loss": 0.98866832,
      "num_input_tokens_seen": 25033875,
      "step": 1175,
      "time_per_iteration": 2.601478099822998
    },
    {
      "auxiliary_loss_clip": 0.01201002,
      "auxiliary_loss_mlp": 0.01039342,
      "balance_loss_clip": 1.0600369,
      "balance_loss_mlp": 1.02734923,
      "epoch": 0.14140563939157097,
      "flos": 20335279224960.0,
      "grad_norm": 3.268061804043667,
      "language_loss": 0.85863793,
      "learning_rate": 3.871325504420238e-06,
      "loss": 0.88104135,
      "num_input_tokens_seen": 25052865,
      "step": 1176,
      "time_per_iteration": 2.562810182571411
    },
    {
      "auxiliary_loss_clip": 0.01236416,
      "auxiliary_loss_mlp": 0.01035924,
      "balance_loss_clip": 1.06784892,
      "balance_loss_mlp": 1.02533817,
      "epoch": 0.14152588228221005,
      "flos": 21068826773760.0,
      "grad_norm": 1.7145377081477498,
      "language_loss": 0.81782746,
      "learning_rate": 3.871050466920776e-06,
      "loss": 0.8405509,
      "num_input_tokens_seen": 25072770,
      "step": 1177,
      "time_per_iteration": 2.502150058746338
    },
    {
      "auxiliary_loss_clip": 0.01179289,
      "auxiliary_loss_mlp": 0.01031377,
      "balance_loss_clip": 1.05759549,
      "balance_loss_mlp": 1.02102399,
      "epoch": 0.14164612517284916,
      "flos": 18223157646720.0,
      "grad_norm": 1.8022537104679055,
      "language_loss": 0.79641569,
      "learning_rate": 3.870775145584447e-06,
      "loss": 0.81852233,
      "num_input_tokens_seen": 25090550,
      "step": 1178,
      "time_per_iteration": 2.632185459136963
    },
    {
      "auxiliary_loss_clip": 0.01212098,
      "auxiliary_loss_mlp": 0.01044947,
      "balance_loss_clip": 1.06302476,
      "balance_loss_mlp": 1.0333184,
      "epoch": 0.14176636806348825,
      "flos": 22744454279040.0,
      "grad_norm": 2.808521870278306,
      "language_loss": 0.64842236,
      "learning_rate": 3.8704995404530145e-06,
      "loss": 0.67099285,
      "num_input_tokens_seen": 25106175,
      "step": 1179,
      "time_per_iteration": 2.6408607959747314
    },
    {
      "auxiliary_loss_clip": 0.01233894,
      "auxiliary_loss_mlp": 0.01038018,
      "balance_loss_clip": 1.06740737,
      "balance_loss_mlp": 1.02773619,
      "epoch": 0.14188661095412733,
      "flos": 22091095843200.0,
      "grad_norm": 2.009320451255184,
      "language_loss": 0.84931207,
      "learning_rate": 3.87022365156829e-06,
      "loss": 0.87203121,
      "num_input_tokens_seen": 25126890,
      "step": 1180,
      "time_per_iteration": 2.526172161102295
    },
    {
      "auxiliary_loss_clip": 0.01142297,
      "auxiliary_loss_mlp": 0.01039185,
      "balance_loss_clip": 1.05483246,
      "balance_loss_mlp": 1.0281527,
      "epoch": 0.14200685384476644,
      "flos": 24352390604160.0,
      "grad_norm": 2.282580410868269,
      "language_loss": 0.81009901,
      "learning_rate": 3.869947478972123e-06,
      "loss": 0.83191383,
      "num_input_tokens_seen": 25147915,
      "step": 1181,
      "time_per_iteration": 2.83522629737854
    },
    {
      "auxiliary_loss_clip": 0.01214349,
      "auxiliary_loss_mlp": 0.01037039,
      "balance_loss_clip": 1.0625701,
      "balance_loss_mlp": 1.02524948,
      "epoch": 0.14212709673540552,
      "flos": 24022048199040.0,
      "grad_norm": 2.262742897864024,
      "language_loss": 0.82555723,
      "learning_rate": 3.869671022706412e-06,
      "loss": 0.84807116,
      "num_input_tokens_seen": 25166645,
      "step": 1182,
      "time_per_iteration": 2.696279525756836
    },
    {
      "auxiliary_loss_clip": 0.01158631,
      "auxiliary_loss_mlp": 0.0104602,
      "balance_loss_clip": 1.05308425,
      "balance_loss_mlp": 1.03517818,
      "epoch": 0.1422473396260446,
      "flos": 26431797870720.0,
      "grad_norm": 1.9557381545607537,
      "language_loss": 0.64627188,
      "learning_rate": 3.869394282813092e-06,
      "loss": 0.66831839,
      "num_input_tokens_seen": 25185845,
      "step": 1183,
      "time_per_iteration": 2.659381628036499
    },
    {
      "auxiliary_loss_clip": 0.01193628,
      "auxiliary_loss_mlp": 0.01043299,
      "balance_loss_clip": 1.05697608,
      "balance_loss_mlp": 1.03195632,
      "epoch": 0.1423675825166837,
      "flos": 17055306754560.0,
      "grad_norm": 3.1121010254052033,
      "language_loss": 0.89463937,
      "learning_rate": 3.869117259334147e-06,
      "loss": 0.91700864,
      "num_input_tokens_seen": 25203770,
      "step": 1184,
      "time_per_iteration": 2.5750486850738525
    },
    {
      "auxiliary_loss_clip": 0.01214635,
      "auxiliary_loss_mlp": 0.01042238,
      "balance_loss_clip": 1.0623312,
      "balance_loss_mlp": 1.03145552,
      "epoch": 0.1424878254073228,
      "flos": 17929480049280.0,
      "grad_norm": 1.853639622062382,
      "language_loss": 0.81930923,
      "learning_rate": 3.868839952311599e-06,
      "loss": 0.84187794,
      "num_input_tokens_seen": 25221725,
      "step": 1185,
      "time_per_iteration": 2.49806547164917
    },
    {
      "auxiliary_loss_clip": 0.01200315,
      "auxiliary_loss_mlp": 0.0103552,
      "balance_loss_clip": 1.06298769,
      "balance_loss_mlp": 1.02446318,
      "epoch": 0.14260806829796188,
      "flos": 20303606407680.0,
      "grad_norm": 2.323956828573697,
      "language_loss": 0.80340397,
      "learning_rate": 3.868562361787516e-06,
      "loss": 0.82576227,
      "num_input_tokens_seen": 25240855,
      "step": 1186,
      "time_per_iteration": 2.5370519161224365
    },
    {
      "auxiliary_loss_clip": 0.01136701,
      "auxiliary_loss_mlp": 0.01033927,
      "balance_loss_clip": 1.05215549,
      "balance_loss_mlp": 1.02290595,
      "epoch": 0.14272831118860096,
      "flos": 23185724860800.0,
      "grad_norm": 2.056503246063335,
      "language_loss": 0.691715,
      "learning_rate": 3.868284487804009e-06,
      "loss": 0.71342123,
      "num_input_tokens_seen": 25260085,
      "step": 1187,
      "time_per_iteration": 2.696873903274536
    },
    {
      "auxiliary_loss_clip": 0.01208476,
      "auxiliary_loss_mlp": 0.01046048,
      "balance_loss_clip": 1.06049669,
      "balance_loss_mlp": 1.03538513,
      "epoch": 0.14284855407924008,
      "flos": 27232210586880.0,
      "grad_norm": 1.6143287209150559,
      "language_loss": 0.78217274,
      "learning_rate": 3.86800633040323e-06,
      "loss": 0.80471802,
      "num_input_tokens_seen": 25280675,
      "step": 1188,
      "time_per_iteration": 2.6943676471710205
    },
    {
      "auxiliary_loss_clip": 0.01202893,
      "auxiliary_loss_mlp": 0.00765847,
      "balance_loss_clip": 1.06457949,
      "balance_loss_mlp": 1.00075531,
      "epoch": 0.14296879696987916,
      "flos": 28184202696960.0,
      "grad_norm": 2.001762600775245,
      "language_loss": 0.78061759,
      "learning_rate": 3.867727889627376e-06,
      "loss": 0.80030501,
      "num_input_tokens_seen": 25300290,
      "step": 1189,
      "time_per_iteration": 2.5867836475372314
    },
    {
      "auxiliary_loss_clip": 0.01178168,
      "auxiliary_loss_mlp": 0.01044245,
      "balance_loss_clip": 1.05849648,
      "balance_loss_mlp": 1.03204358,
      "epoch": 0.14308903986051824,
      "flos": 19390290266880.0,
      "grad_norm": 2.1207731494882895,
      "language_loss": 0.78223324,
      "learning_rate": 3.867449165518687e-06,
      "loss": 0.80445737,
      "num_input_tokens_seen": 25316760,
      "step": 1190,
      "time_per_iteration": 2.5264713764190674
    },
    {
      "auxiliary_loss_clip": 0.01238012,
      "auxiliary_loss_mlp": 0.00766565,
      "balance_loss_clip": 1.06639266,
      "balance_loss_mlp": 1.00079107,
      "epoch": 0.14320928275115732,
      "flos": 17457506317440.0,
      "grad_norm": 1.7759648261721543,
      "language_loss": 0.71031684,
      "learning_rate": 3.867170158119444e-06,
      "loss": 0.73036253,
      "num_input_tokens_seen": 25335760,
      "step": 1191,
      "time_per_iteration": 2.556504487991333
    },
    {
      "auxiliary_loss_clip": 0.01238046,
      "auxiliary_loss_mlp": 0.01037906,
      "balance_loss_clip": 1.06650412,
      "balance_loss_mlp": 1.02736795,
      "epoch": 0.14332952564179643,
      "flos": 21466070259840.0,
      "grad_norm": 2.214404937392677,
      "language_loss": 0.75551212,
      "learning_rate": 3.866890867471972e-06,
      "loss": 0.77827168,
      "num_input_tokens_seen": 25354230,
      "step": 1192,
      "time_per_iteration": 2.5354533195495605
    },
    {
      "auxiliary_loss_clip": 0.01198297,
      "auxiliary_loss_mlp": 0.01044425,
      "balance_loss_clip": 1.0549221,
      "balance_loss_mlp": 1.03257012,
      "epoch": 0.14344976853243552,
      "flos": 16396992241920.0,
      "grad_norm": 2.5540064589372715,
      "language_loss": 0.89672583,
      "learning_rate": 3.86661129361864e-06,
      "loss": 0.91915309,
      "num_input_tokens_seen": 25368720,
      "step": 1193,
      "time_per_iteration": 3.3329994678497314
    },
    {
      "auxiliary_loss_clip": 0.01202651,
      "auxiliary_loss_mlp": 0.01046793,
      "balance_loss_clip": 1.06281376,
      "balance_loss_mlp": 1.03503346,
      "epoch": 0.1435700114230746,
      "flos": 18916736336640.0,
      "grad_norm": 1.9344724869248264,
      "language_loss": 0.86119479,
      "learning_rate": 3.866331436601859e-06,
      "loss": 0.88368922,
      "num_input_tokens_seen": 25386715,
      "step": 1194,
      "time_per_iteration": 2.5208323001861572
    },
    {
      "auxiliary_loss_clip": 0.01235557,
      "auxiliary_loss_mlp": 0.01042975,
      "balance_loss_clip": 1.06596208,
      "balance_loss_mlp": 1.03156674,
      "epoch": 0.1436902543137137,
      "flos": 19755394058880.0,
      "grad_norm": 2.205488070809292,
      "language_loss": 0.73481584,
      "learning_rate": 3.866051296464083e-06,
      "loss": 0.75760114,
      "num_input_tokens_seen": 25405550,
      "step": 1195,
      "time_per_iteration": 2.485048770904541
    },
    {
      "auxiliary_loss_clip": 0.01234827,
      "auxiliary_loss_mlp": 0.00765843,
      "balance_loss_clip": 1.06353295,
      "balance_loss_mlp": 1.00076652,
      "epoch": 0.1438104972043528,
      "flos": 14684807669760.0,
      "grad_norm": 2.8025494673110067,
      "language_loss": 0.85218567,
      "learning_rate": 3.86577087324781e-06,
      "loss": 0.87219238,
      "num_input_tokens_seen": 25422040,
      "step": 1196,
      "time_per_iteration": 2.424133539199829
    },
    {
      "auxiliary_loss_clip": 0.01217736,
      "auxiliary_loss_mlp": 0.01037417,
      "balance_loss_clip": 1.06584454,
      "balance_loss_mlp": 1.02664042,
      "epoch": 0.14393074009499188,
      "flos": 17092330698240.0,
      "grad_norm": 1.9612024199306752,
      "language_loss": 0.77624083,
      "learning_rate": 3.865490166995578e-06,
      "loss": 0.79879236,
      "num_input_tokens_seen": 25440270,
      "step": 1197,
      "time_per_iteration": 2.529550313949585
    },
    {
      "auxiliary_loss_clip": 0.01219585,
      "auxiliary_loss_mlp": 0.01041657,
      "balance_loss_clip": 1.06455755,
      "balance_loss_mlp": 1.03010535,
      "epoch": 0.144050982985631,
      "flos": 30476200608000.0,
      "grad_norm": 4.971996642695616,
      "language_loss": 0.8433696,
      "learning_rate": 3.86520917774997e-06,
      "loss": 0.86598206,
      "num_input_tokens_seen": 25459705,
      "step": 1198,
      "time_per_iteration": 2.6028761863708496
    },
    {
      "auxiliary_loss_clip": 0.01213289,
      "auxiliary_loss_mlp": 0.01044164,
      "balance_loss_clip": 1.06324959,
      "balance_loss_mlp": 1.03384066,
      "epoch": 0.14417122587627007,
      "flos": 17858484817920.0,
      "grad_norm": 2.2202358873986574,
      "language_loss": 0.74863768,
      "learning_rate": 3.864927905553614e-06,
      "loss": 0.77121222,
      "num_input_tokens_seen": 25477615,
      "step": 1199,
      "time_per_iteration": 2.5557374954223633
    },
    {
      "auxiliary_loss_clip": 0.01181641,
      "auxiliary_loss_mlp": 0.01042246,
      "balance_loss_clip": 1.05734086,
      "balance_loss_mlp": 1.03181517,
      "epoch": 0.14429146876690915,
      "flos": 21613914639360.0,
      "grad_norm": 1.6120325061678757,
      "language_loss": 0.88999832,
      "learning_rate": 3.8646463504491765e-06,
      "loss": 0.91223723,
      "num_input_tokens_seen": 25497750,
      "step": 1200,
      "time_per_iteration": 4.302061557769775
    },
    {
      "auxiliary_loss_clip": 0.01222008,
      "auxiliary_loss_mlp": 0.0104002,
      "balance_loss_clip": 1.06677556,
      "balance_loss_mlp": 1.02803946,
      "epoch": 0.14441171165754824,
      "flos": 23258120722560.0,
      "grad_norm": 1.7643310954077014,
      "language_loss": 0.8278898,
      "learning_rate": 3.8643645124793705e-06,
      "loss": 0.85051,
      "num_input_tokens_seen": 25516650,
      "step": 1201,
      "time_per_iteration": 2.5275492668151855
    },
    {
      "auxiliary_loss_clip": 0.01215386,
      "auxiliary_loss_mlp": 0.01035574,
      "balance_loss_clip": 1.06172729,
      "balance_loss_mlp": 1.02488685,
      "epoch": 0.14453195454818735,
      "flos": 42854213963520.0,
      "grad_norm": 1.5969281970308902,
      "language_loss": 0.74687552,
      "learning_rate": 3.8640823916869515e-06,
      "loss": 0.7693851,
      "num_input_tokens_seen": 25540960,
      "step": 1202,
      "time_per_iteration": 2.702115297317505
    },
    {
      "auxiliary_loss_clip": 0.01233115,
      "auxiliary_loss_mlp": 0.01036422,
      "balance_loss_clip": 1.06433201,
      "balance_loss_mlp": 1.02570486,
      "epoch": 0.14465219743882643,
      "flos": 27235873774080.0,
      "grad_norm": 1.745997800890538,
      "language_loss": 0.78473842,
      "learning_rate": 3.863799988114714e-06,
      "loss": 0.80743378,
      "num_input_tokens_seen": 25562990,
      "step": 1203,
      "time_per_iteration": 2.5356040000915527
    },
    {
      "auxiliary_loss_clip": 0.01239593,
      "auxiliary_loss_mlp": 0.01034699,
      "balance_loss_clip": 1.06682611,
      "balance_loss_mlp": 1.02324939,
      "epoch": 0.1447724403294655,
      "flos": 16690705752960.0,
      "grad_norm": 5.995993969021272,
      "language_loss": 0.70618761,
      "learning_rate": 3.863517301805502e-06,
      "loss": 0.72893059,
      "num_input_tokens_seen": 25581380,
      "step": 1204,
      "time_per_iteration": 2.4148178100585938
    },
    {
      "auxiliary_loss_clip": 0.01190975,
      "auxiliary_loss_mlp": 0.01040444,
      "balance_loss_clip": 1.06332421,
      "balance_loss_mlp": 1.02926266,
      "epoch": 0.14489268322010462,
      "flos": 20073741321600.0,
      "grad_norm": 2.501096555932035,
      "language_loss": 0.9669441,
      "learning_rate": 3.863234332802196e-06,
      "loss": 0.98925823,
      "num_input_tokens_seen": 25593585,
      "step": 1205,
      "time_per_iteration": 2.5622525215148926
    },
    {
      "auxiliary_loss_clip": 0.01197562,
      "auxiliary_loss_mlp": 0.01043914,
      "balance_loss_clip": 1.05821252,
      "balance_loss_mlp": 1.0336864,
      "epoch": 0.1450129261107437,
      "flos": 27125627955840.0,
      "grad_norm": 2.1030886228043117,
      "language_loss": 0.74215102,
      "learning_rate": 3.862951081147723e-06,
      "loss": 0.76456577,
      "num_input_tokens_seen": 25613750,
      "step": 1206,
      "time_per_iteration": 2.5627236366271973
    },
    {
      "auxiliary_loss_clip": 0.01218229,
      "auxiliary_loss_mlp": 0.01040788,
      "balance_loss_clip": 1.06570017,
      "balance_loss_mlp": 1.03087544,
      "epoch": 0.1451331690013828,
      "flos": 25702344472320.0,
      "grad_norm": 2.3329062118639916,
      "language_loss": 0.77881825,
      "learning_rate": 3.862667546885053e-06,
      "loss": 0.80140841,
      "num_input_tokens_seen": 25632300,
      "step": 1207,
      "time_per_iteration": 2.5468392372131348
    },
    {
      "auxiliary_loss_clip": 0.01207141,
      "auxiliary_loss_mlp": 0.01042052,
      "balance_loss_clip": 1.06144905,
      "balance_loss_mlp": 1.03090012,
      "epoch": 0.14525341189202187,
      "flos": 25737393168000.0,
      "grad_norm": 2.0588635540689166,
      "language_loss": 0.73146099,
      "learning_rate": 3.8623837300571965e-06,
      "loss": 0.75395298,
      "num_input_tokens_seen": 25651285,
      "step": 1208,
      "time_per_iteration": 2.5630273818969727
    },
    {
      "auxiliary_loss_clip": 0.01236599,
      "auxiliary_loss_mlp": 0.01038061,
      "balance_loss_clip": 1.06619263,
      "balance_loss_mlp": 1.02670598,
      "epoch": 0.14537365478266098,
      "flos": 23073898844160.0,
      "grad_norm": 2.3296526383850877,
      "language_loss": 0.84067225,
      "learning_rate": 3.8620996307072085e-06,
      "loss": 0.86341882,
      "num_input_tokens_seen": 25671990,
      "step": 1209,
      "time_per_iteration": 2.474330425262451
    },
    {
      "auxiliary_loss_clip": 0.01188648,
      "auxiliary_loss_mlp": 0.01038214,
      "balance_loss_clip": 1.05639768,
      "balance_loss_mlp": 1.02694249,
      "epoch": 0.14549389767330007,
      "flos": 20595021448320.0,
      "grad_norm": 1.8315405527030273,
      "language_loss": 0.64536703,
      "learning_rate": 3.861815248878188e-06,
      "loss": 0.66763568,
      "num_input_tokens_seen": 25689475,
      "step": 1210,
      "time_per_iteration": 2.593482255935669
    },
    {
      "auxiliary_loss_clip": 0.0120048,
      "auxiliary_loss_mlp": 0.01043552,
      "balance_loss_clip": 1.06309366,
      "balance_loss_mlp": 1.03321075,
      "epoch": 0.14561414056393915,
      "flos": 15121804533120.0,
      "grad_norm": 5.521584986918201,
      "language_loss": 0.79655075,
      "learning_rate": 3.861530584613274e-06,
      "loss": 0.81899112,
      "num_input_tokens_seen": 25707475,
      "step": 1211,
      "time_per_iteration": 2.5235373973846436
    },
    {
      "auxiliary_loss_clip": 0.01222332,
      "auxiliary_loss_mlp": 0.00766027,
      "balance_loss_clip": 1.06636417,
      "balance_loss_mlp": 1.00107944,
      "epoch": 0.14573438345457826,
      "flos": 19427493778560.0,
      "grad_norm": 2.199432454665693,
      "language_loss": 0.82192487,
      "learning_rate": 3.86124563795565e-06,
      "loss": 0.84180844,
      "num_input_tokens_seen": 25726290,
      "step": 1212,
      "time_per_iteration": 2.523937702178955
    },
    {
      "auxiliary_loss_clip": 0.01233505,
      "auxiliary_loss_mlp": 0.0103872,
      "balance_loss_clip": 1.0668366,
      "balance_loss_mlp": 1.02824187,
      "epoch": 0.14585462634521734,
      "flos": 24828422572800.0,
      "grad_norm": 1.6804086105383762,
      "language_loss": 0.70204425,
      "learning_rate": 3.860960408948543e-06,
      "loss": 0.72476649,
      "num_input_tokens_seen": 25748040,
      "step": 1213,
      "time_per_iteration": 2.5231709480285645
    },
    {
      "auxiliary_loss_clip": 0.01207665,
      "auxiliary_loss_mlp": 0.01042277,
      "balance_loss_clip": 1.06314898,
      "balance_loss_mlp": 1.03201938,
      "epoch": 0.14597486923585642,
      "flos": 15448627405440.0,
      "grad_norm": 2.4291383192565514,
      "language_loss": 0.89876348,
      "learning_rate": 3.860674897635222e-06,
      "loss": 0.92126286,
      "num_input_tokens_seen": 25764525,
      "step": 1214,
      "time_per_iteration": 2.5019190311431885
    },
    {
      "auxiliary_loss_clip": 0.01218165,
      "auxiliary_loss_mlp": 0.01048291,
      "balance_loss_clip": 1.0659411,
      "balance_loss_mlp": 1.03723979,
      "epoch": 0.1460951121264955,
      "flos": 16655154266880.0,
      "grad_norm": 1.9940973093713716,
      "language_loss": 0.83366507,
      "learning_rate": 3.860389104058998e-06,
      "loss": 0.85632968,
      "num_input_tokens_seen": 25782755,
      "step": 1215,
      "time_per_iteration": 2.559870481491089
    },
    {
      "auxiliary_loss_clip": 0.01201474,
      "auxiliary_loss_mlp": 0.01034726,
      "balance_loss_clip": 1.06231809,
      "balance_loss_mlp": 1.02429473,
      "epoch": 0.14621535501713462,
      "flos": 24863291700480.0,
      "grad_norm": 1.8641895390783614,
      "language_loss": 0.72593713,
      "learning_rate": 3.860103028263227e-06,
      "loss": 0.74829912,
      "num_input_tokens_seen": 25805860,
      "step": 1216,
      "time_per_iteration": 2.6286466121673584
    },
    {
      "auxiliary_loss_clip": 0.0116329,
      "auxiliary_loss_mlp": 0.01035046,
      "balance_loss_clip": 1.05190015,
      "balance_loss_mlp": 1.02422214,
      "epoch": 0.1463355979077737,
      "flos": 25228000442880.0,
      "grad_norm": 2.333938901783611,
      "language_loss": 0.70038521,
      "learning_rate": 3.859816670291304e-06,
      "loss": 0.7223686,
      "num_input_tokens_seen": 25824955,
      "step": 1217,
      "time_per_iteration": 2.682464838027954
    },
    {
      "auxiliary_loss_clip": 0.01150706,
      "auxiliary_loss_mlp": 0.01032483,
      "balance_loss_clip": 1.05657268,
      "balance_loss_mlp": 1.02086031,
      "epoch": 0.14645584079841278,
      "flos": 22054143726720.0,
      "grad_norm": 2.3336884630007955,
      "language_loss": 0.8993752,
      "learning_rate": 3.859530030186672e-06,
      "loss": 0.92120707,
      "num_input_tokens_seen": 25841965,
      "step": 1218,
      "time_per_iteration": 2.666663408279419
    },
    {
      "auxiliary_loss_clip": 0.01207673,
      "auxiliary_loss_mlp": 0.01037527,
      "balance_loss_clip": 1.06523824,
      "balance_loss_mlp": 1.02678597,
      "epoch": 0.1465760836890519,
      "flos": 23623870959360.0,
      "grad_norm": 2.1699847963898073,
      "language_loss": 0.8243829,
      "learning_rate": 3.859243107992813e-06,
      "loss": 0.84683496,
      "num_input_tokens_seen": 25860770,
      "step": 1219,
      "time_per_iteration": 2.598076820373535
    },
    {
      "auxiliary_loss_clip": 0.01188952,
      "auxiliary_loss_mlp": 0.01038613,
      "balance_loss_clip": 1.05639017,
      "balance_loss_mlp": 1.02721632,
      "epoch": 0.14669632657969098,
      "flos": 37407893356800.0,
      "grad_norm": 2.087799042459627,
      "language_loss": 0.77893454,
      "learning_rate": 3.858955903753252e-06,
      "loss": 0.80121017,
      "num_input_tokens_seen": 25879410,
      "step": 1220,
      "time_per_iteration": 2.7478127479553223
    },
    {
      "auxiliary_loss_clip": 0.01216577,
      "auxiliary_loss_mlp": 0.01037962,
      "balance_loss_clip": 1.06097937,
      "balance_loss_mlp": 1.02846062,
      "epoch": 0.14681656947033006,
      "flos": 28365910623360.0,
      "grad_norm": 1.5556749032978585,
      "language_loss": 0.83639264,
      "learning_rate": 3.858668417511559e-06,
      "loss": 0.85893798,
      "num_input_tokens_seen": 25902160,
      "step": 1221,
      "time_per_iteration": 3.4299418926239014
    },
    {
      "auxiliary_loss_clip": 0.01206603,
      "auxiliary_loss_mlp": 0.01032313,
      "balance_loss_clip": 1.0648303,
      "balance_loss_mlp": 1.02134025,
      "epoch": 0.14693681236096917,
      "flos": 18479488078080.0,
      "grad_norm": 2.1766633783015403,
      "language_loss": 0.76281548,
      "learning_rate": 3.8583806493113445e-06,
      "loss": 0.78520465,
      "num_input_tokens_seen": 25920505,
      "step": 1222,
      "time_per_iteration": 2.6012563705444336
    },
    {
      "auxiliary_loss_clip": 0.01215196,
      "auxiliary_loss_mlp": 0.01041007,
      "balance_loss_clip": 1.06405354,
      "balance_loss_mlp": 1.03082013,
      "epoch": 0.14705705525160825,
      "flos": 20777806782720.0,
      "grad_norm": 2.2225291822672437,
      "language_loss": 0.82385731,
      "learning_rate": 3.858092599196263e-06,
      "loss": 0.84641933,
      "num_input_tokens_seen": 25938460,
      "step": 1223,
      "time_per_iteration": 2.4934475421905518
    },
    {
      "auxiliary_loss_clip": 0.01218014,
      "auxiliary_loss_mlp": 0.01032236,
      "balance_loss_clip": 1.06439936,
      "balance_loss_mlp": 1.02188921,
      "epoch": 0.14717729814224734,
      "flos": 29932944336000.0,
      "grad_norm": 2.5930044814038324,
      "language_loss": 0.82323581,
      "learning_rate": 3.857804267210012e-06,
      "loss": 0.84573829,
      "num_input_tokens_seen": 25957760,
      "step": 1224,
      "time_per_iteration": 2.541569232940674
    },
    {
      "auxiliary_loss_clip": 0.01170889,
      "auxiliary_loss_mlp": 0.01043631,
      "balance_loss_clip": 1.05348575,
      "balance_loss_mlp": 1.03323007,
      "epoch": 0.14729754103288642,
      "flos": 20047491457920.0,
      "grad_norm": 2.1280711188396375,
      "language_loss": 0.88104141,
      "learning_rate": 3.857515653396331e-06,
      "loss": 0.90318668,
      "num_input_tokens_seen": 25974970,
      "step": 1225,
      "time_per_iteration": 2.5456156730651855
    },
    {
      "auxiliary_loss_clip": 0.01171425,
      "auxiliary_loss_mlp": 0.01038743,
      "balance_loss_clip": 1.05644774,
      "balance_loss_mlp": 1.02866352,
      "epoch": 0.14741778392352553,
      "flos": 19281516906240.0,
      "grad_norm": 2.3431463258731995,
      "language_loss": 0.87043232,
      "learning_rate": 3.857226757799002e-06,
      "loss": 0.89253402,
      "num_input_tokens_seen": 25992525,
      "step": 1226,
      "time_per_iteration": 3.3609275817871094
    },
    {
      "auxiliary_loss_clip": 0.01199118,
      "auxiliary_loss_mlp": 0.01037997,
      "balance_loss_clip": 1.05969715,
      "balance_loss_mlp": 1.02741694,
      "epoch": 0.1475380268141646,
      "flos": 25411108999680.0,
      "grad_norm": 2.3536418791766964,
      "language_loss": 0.74155647,
      "learning_rate": 3.85693758046185e-06,
      "loss": 0.76392758,
      "num_input_tokens_seen": 26010815,
      "step": 1227,
      "time_per_iteration": 4.199660301208496
    },
    {
      "auxiliary_loss_clip": 0.01232908,
      "auxiliary_loss_mlp": 0.01041756,
      "balance_loss_clip": 1.06786013,
      "balance_loss_mlp": 1.03155172,
      "epoch": 0.1476582697048037,
      "flos": 20847652778880.0,
      "grad_norm": 1.926404603545611,
      "language_loss": 0.82754517,
      "learning_rate": 3.8566481214287435e-06,
      "loss": 0.85029173,
      "num_input_tokens_seen": 26028935,
      "step": 1228,
      "time_per_iteration": 2.489389181137085
    },
    {
      "auxiliary_loss_clip": 0.01177275,
      "auxiliary_loss_mlp": 0.01044358,
      "balance_loss_clip": 1.05549455,
      "balance_loss_mlp": 1.0336647,
      "epoch": 0.1477785125954428,
      "flos": 14028109269120.0,
      "grad_norm": 2.0631050130531796,
      "language_loss": 0.90664279,
      "learning_rate": 3.8563583807435935e-06,
      "loss": 0.92885911,
      "num_input_tokens_seen": 26045080,
      "step": 1229,
      "time_per_iteration": 2.5459182262420654
    },
    {
      "auxiliary_loss_clip": 0.01220487,
      "auxiliary_loss_mlp": 0.00765693,
      "balance_loss_clip": 1.06373549,
      "balance_loss_mlp": 1.00118995,
      "epoch": 0.1478987554860819,
      "flos": 20516699842560.0,
      "grad_norm": 2.151583407052703,
      "language_loss": 0.77519083,
      "learning_rate": 3.856068358450353e-06,
      "loss": 0.79505265,
      "num_input_tokens_seen": 26065030,
      "step": 1230,
      "time_per_iteration": 2.523956298828125
    },
    {
      "auxiliary_loss_clip": 0.01199544,
      "auxiliary_loss_mlp": 0.01041675,
      "balance_loss_clip": 1.06643867,
      "balance_loss_mlp": 1.031358,
      "epoch": 0.14801899837672097,
      "flos": 17857012360320.0,
      "grad_norm": 1.7727411085181164,
      "language_loss": 0.855335,
      "learning_rate": 3.8557780545930186e-06,
      "loss": 0.87774724,
      "num_input_tokens_seen": 26083445,
      "step": 1231,
      "time_per_iteration": 2.5099692344665527
    },
    {
      "auxiliary_loss_clip": 0.01201041,
      "auxiliary_loss_mlp": 0.0104045,
      "balance_loss_clip": 1.06306303,
      "balance_loss_mlp": 1.03039432,
      "epoch": 0.14813924126736006,
      "flos": 20881408584960.0,
      "grad_norm": 1.7840170347637485,
      "language_loss": 0.7953316,
      "learning_rate": 3.855487469215628e-06,
      "loss": 0.81774652,
      "num_input_tokens_seen": 26102375,
      "step": 1232,
      "time_per_iteration": 2.579627752304077
    },
    {
      "auxiliary_loss_clip": 0.01186958,
      "auxiliary_loss_mlp": 0.01034884,
      "balance_loss_clip": 1.06098998,
      "balance_loss_mlp": 1.02443576,
      "epoch": 0.14825948415799917,
      "flos": 37414070496000.0,
      "grad_norm": 2.1075343167426537,
      "language_loss": 0.72536087,
      "learning_rate": 3.855196602362264e-06,
      "loss": 0.74757934,
      "num_input_tokens_seen": 26125295,
      "step": 1233,
      "time_per_iteration": 2.8180525302886963
    },
    {
      "auxiliary_loss_clip": 0.01214793,
      "auxiliary_loss_mlp": 0.01031907,
      "balance_loss_clip": 1.06178856,
      "balance_loss_mlp": 1.02136278,
      "epoch": 0.14837972704863825,
      "flos": 22014641744640.0,
      "grad_norm": 1.9838775211896933,
      "language_loss": 0.94167888,
      "learning_rate": 3.854905454077051e-06,
      "loss": 0.9641459,
      "num_input_tokens_seen": 26142905,
      "step": 1234,
      "time_per_iteration": 2.5686495304107666
    },
    {
      "auxiliary_loss_clip": 0.01137481,
      "auxiliary_loss_mlp": 0.01037529,
      "balance_loss_clip": 1.05279207,
      "balance_loss_mlp": 1.02734292,
      "epoch": 0.14849996993927733,
      "flos": 20996323171200.0,
      "grad_norm": 2.2573491934859526,
      "language_loss": 0.88214552,
      "learning_rate": 3.854614024404155e-06,
      "loss": 0.90389562,
      "num_input_tokens_seen": 26161215,
      "step": 1235,
      "time_per_iteration": 2.7196803092956543
    },
    {
      "auxiliary_loss_clip": 0.01187531,
      "auxiliary_loss_mlp": 0.01033257,
      "balance_loss_clip": 1.05716324,
      "balance_loss_mlp": 1.02280831,
      "epoch": 0.14862021282991644,
      "flos": 20047994248320.0,
      "grad_norm": 1.9089457635837919,
      "language_loss": 0.89282167,
      "learning_rate": 3.8543223133877865e-06,
      "loss": 0.91502953,
      "num_input_tokens_seen": 26179810,
      "step": 1236,
      "time_per_iteration": 2.5940613746643066
    },
    {
      "auxiliary_loss_clip": 0.01183638,
      "auxiliary_loss_mlp": 0.01042046,
      "balance_loss_clip": 1.05688262,
      "balance_loss_mlp": 1.02977979,
      "epoch": 0.14874045572055553,
      "flos": 22712027276160.0,
      "grad_norm": 1.8445128142860934,
      "language_loss": 0.88202894,
      "learning_rate": 3.854030321072198e-06,
      "loss": 0.90428585,
      "num_input_tokens_seen": 26199715,
      "step": 1237,
      "time_per_iteration": 2.8678267002105713
    },
    {
      "auxiliary_loss_clip": 0.01191781,
      "auxiliary_loss_mlp": 0.01030139,
      "balance_loss_clip": 1.06005001,
      "balance_loss_mlp": 1.01998234,
      "epoch": 0.1488606986111946,
      "flos": 25411288567680.0,
      "grad_norm": 1.9483647369410164,
      "language_loss": 0.73584008,
      "learning_rate": 3.853738047501682e-06,
      "loss": 0.75805926,
      "num_input_tokens_seen": 26220275,
      "step": 1238,
      "time_per_iteration": 2.7049202919006348
    },
    {
      "auxiliary_loss_clip": 0.01218715,
      "auxiliary_loss_mlp": 0.01039142,
      "balance_loss_clip": 1.06586349,
      "balance_loss_mlp": 1.02797818,
      "epoch": 0.1489809415018337,
      "flos": 17018749687680.0,
      "grad_norm": 2.075972479156352,
      "language_loss": 0.77499306,
      "learning_rate": 3.85344549272058e-06,
      "loss": 0.7975716,
      "num_input_tokens_seen": 26238255,
      "step": 1239,
      "time_per_iteration": 2.5351147651672363
    },
    {
      "auxiliary_loss_clip": 0.01212706,
      "auxiliary_loss_mlp": 0.01039232,
      "balance_loss_clip": 1.06142783,
      "balance_loss_mlp": 1.02824092,
      "epoch": 0.1491011843924728,
      "flos": 33659394860160.0,
      "grad_norm": 1.7969874009261801,
      "language_loss": 0.82827985,
      "learning_rate": 3.853152656773269e-06,
      "loss": 0.8507992,
      "num_input_tokens_seen": 26259690,
      "step": 1240,
      "time_per_iteration": 2.667109966278076
    },
    {
      "auxiliary_loss_clip": 0.01198511,
      "auxiliary_loss_mlp": 0.01034156,
      "balance_loss_clip": 1.06150651,
      "balance_loss_mlp": 1.02347493,
      "epoch": 0.14922142728311188,
      "flos": 21179000764800.0,
      "grad_norm": 1.9606264955825021,
      "language_loss": 0.84800541,
      "learning_rate": 3.852859539704174e-06,
      "loss": 0.87033212,
      "num_input_tokens_seen": 26278990,
      "step": 1241,
      "time_per_iteration": 2.6013031005859375
    },
    {
      "auxiliary_loss_clip": 0.01166085,
      "auxiliary_loss_mlp": 0.01040563,
      "balance_loss_clip": 1.05418491,
      "balance_loss_mlp": 1.02991736,
      "epoch": 0.14934167017375097,
      "flos": 29860548474240.0,
      "grad_norm": 1.894130687108162,
      "language_loss": 0.76595157,
      "learning_rate": 3.85256614155776e-06,
      "loss": 0.78801805,
      "num_input_tokens_seen": 26299120,
      "step": 1242,
      "time_per_iteration": 2.697136878967285
    },
    {
      "auxiliary_loss_clip": 0.01212995,
      "auxiliary_loss_mlp": 0.01035343,
      "balance_loss_clip": 1.05970931,
      "balance_loss_mlp": 1.024912,
      "epoch": 0.14946191306439008,
      "flos": 17019216564480.0,
      "grad_norm": 1.9377910472175266,
      "language_loss": 0.74569196,
      "learning_rate": 3.852272462378535e-06,
      "loss": 0.76817536,
      "num_input_tokens_seen": 26316995,
      "step": 1243,
      "time_per_iteration": 2.5294992923736572
    },
    {
      "auxiliary_loss_clip": 0.0119959,
      "auxiliary_loss_mlp": 0.01043554,
      "balance_loss_clip": 1.06105196,
      "balance_loss_mlp": 1.03357601,
      "epoch": 0.14958215595502916,
      "flos": 15669047214720.0,
      "grad_norm": 1.9167691044361883,
      "language_loss": 0.77718198,
      "learning_rate": 3.85197850221105e-06,
      "loss": 0.79961348,
      "num_input_tokens_seen": 26333295,
      "step": 1244,
      "time_per_iteration": 2.5356898307800293
    },
    {
      "auxiliary_loss_clip": 0.01217267,
      "auxiliary_loss_mlp": 0.01038893,
      "balance_loss_clip": 1.06765032,
      "balance_loss_mlp": 1.02871215,
      "epoch": 0.14970239884566824,
      "flos": 33108560818560.0,
      "grad_norm": 1.846309031505951,
      "language_loss": 0.76009542,
      "learning_rate": 3.851684261099899e-06,
      "loss": 0.78265703,
      "num_input_tokens_seen": 26355035,
      "step": 1245,
      "time_per_iteration": 2.6334850788116455
    },
    {
      "auxiliary_loss_clip": 0.01195769,
      "auxiliary_loss_mlp": 0.01035537,
      "balance_loss_clip": 1.05859756,
      "balance_loss_mlp": 1.0236876,
      "epoch": 0.14982264173630733,
      "flos": 17821245392640.0,
      "grad_norm": 1.9784966353138738,
      "language_loss": 0.86618328,
      "learning_rate": 3.851389739089718e-06,
      "loss": 0.88849628,
      "num_input_tokens_seen": 26371655,
      "step": 1246,
      "time_per_iteration": 2.4992992877960205
    },
    {
      "auxiliary_loss_clip": 0.01222927,
      "auxiliary_loss_mlp": 0.01037831,
      "balance_loss_clip": 1.06952918,
      "balance_loss_mlp": 1.02729273,
      "epoch": 0.14994288462694644,
      "flos": 32409559175040.0,
      "grad_norm": 1.8646401595558197,
      "language_loss": 0.80390334,
      "learning_rate": 3.851094936225186e-06,
      "loss": 0.82651091,
      "num_input_tokens_seen": 26392540,
      "step": 1247,
      "time_per_iteration": 3.3245184421539307
    },
    {
      "auxiliary_loss_clip": 0.01198179,
      "auxiliary_loss_mlp": 0.01030544,
      "balance_loss_clip": 1.06495309,
      "balance_loss_mlp": 1.02032173,
      "epoch": 0.15006312751758552,
      "flos": 31794661226880.0,
      "grad_norm": 1.4361056906965086,
      "language_loss": 0.76639128,
      "learning_rate": 3.850799852551024e-06,
      "loss": 0.78867853,
      "num_input_tokens_seen": 26414960,
      "step": 1248,
      "time_per_iteration": 2.604133129119873
    },
    {
      "auxiliary_loss_clip": 0.01208518,
      "auxiliary_loss_mlp": 0.01042822,
      "balance_loss_clip": 1.06055403,
      "balance_loss_mlp": 1.03155136,
      "epoch": 0.1501833704082246,
      "flos": 16618022582400.0,
      "grad_norm": 2.3968315918848697,
      "language_loss": 0.86450148,
      "learning_rate": 3.850504488111995e-06,
      "loss": 0.88701487,
      "num_input_tokens_seen": 26431635,
      "step": 1249,
      "time_per_iteration": 2.4781830310821533
    },
    {
      "auxiliary_loss_clip": 0.01191947,
      "auxiliary_loss_mlp": 0.01032365,
      "balance_loss_clip": 1.05761659,
      "balance_loss_mlp": 1.02242303,
      "epoch": 0.15030361329886371,
      "flos": 23471178243840.0,
      "grad_norm": 2.1618187401357,
      "language_loss": 0.82751626,
      "learning_rate": 3.850208842952907e-06,
      "loss": 0.84975946,
      "num_input_tokens_seen": 26450440,
      "step": 1250,
      "time_per_iteration": 2.5932819843292236
    },
    {
      "auxiliary_loss_clip": 0.01177491,
      "auxiliary_loss_mlp": 0.01041265,
      "balance_loss_clip": 1.05656052,
      "balance_loss_mlp": 1.03056002,
      "epoch": 0.1504238561895028,
      "flos": 25629409906560.0,
      "grad_norm": 1.9155241691215383,
      "language_loss": 0.79397964,
      "learning_rate": 3.849912917118608e-06,
      "loss": 0.81616724,
      "num_input_tokens_seen": 26471480,
      "step": 1251,
      "time_per_iteration": 2.6426584720611572
    },
    {
      "auxiliary_loss_clip": 0.01128717,
      "auxiliary_loss_mlp": 0.01010275,
      "balance_loss_clip": 1.03730643,
      "balance_loss_mlp": 1.00724685,
      "epoch": 0.15054409908014188,
      "flos": 52095146129280.0,
      "grad_norm": 0.8852069280610807,
      "language_loss": 0.59294569,
      "learning_rate": 3.849616710653992e-06,
      "loss": 0.61433554,
      "num_input_tokens_seen": 26532950,
      "step": 1252,
      "time_per_iteration": 3.051146984100342
    },
    {
      "auxiliary_loss_clip": 0.01213775,
      "auxiliary_loss_mlp": 0.01039175,
      "balance_loss_clip": 1.06332695,
      "balance_loss_mlp": 1.02819586,
      "epoch": 0.150664341970781,
      "flos": 18880251096960.0,
      "grad_norm": 1.8306079649858018,
      "language_loss": 0.75011694,
      "learning_rate": 3.84932022360399e-06,
      "loss": 0.77264643,
      "num_input_tokens_seen": 26551615,
      "step": 1253,
      "time_per_iteration": 3.309061288833618
    },
    {
      "auxiliary_loss_clip": 0.0120186,
      "auxiliary_loss_mlp": 0.01047958,
      "balance_loss_clip": 1.06609726,
      "balance_loss_mlp": 1.03640103,
      "epoch": 0.15078458486142007,
      "flos": 22163240309760.0,
      "grad_norm": 3.1467694429227144,
      "language_loss": 0.84721869,
      "learning_rate": 3.849023456013581e-06,
      "loss": 0.86971688,
      "num_input_tokens_seen": 26569175,
      "step": 1254,
      "time_per_iteration": 3.4654369354248047
    },
    {
      "auxiliary_loss_clip": 0.01223025,
      "auxiliary_loss_mlp": 0.01040879,
      "balance_loss_clip": 1.06529975,
      "balance_loss_mlp": 1.02985799,
      "epoch": 0.15090482775205916,
      "flos": 26651894457600.0,
      "grad_norm": 2.206111092908473,
      "language_loss": 0.62324739,
      "learning_rate": 3.848726407927784e-06,
      "loss": 0.64588642,
      "num_input_tokens_seen": 26589560,
      "step": 1255,
      "time_per_iteration": 2.5752763748168945
    },
    {
      "auxiliary_loss_clip": 0.01204494,
      "auxiliary_loss_mlp": 0.01040541,
      "balance_loss_clip": 1.06490159,
      "balance_loss_mlp": 1.03045022,
      "epoch": 0.15102507064269824,
      "flos": 21798998444160.0,
      "grad_norm": 2.455917277957209,
      "language_loss": 0.86136806,
      "learning_rate": 3.84842907939166e-06,
      "loss": 0.88381845,
      "num_input_tokens_seen": 26608785,
      "step": 1256,
      "time_per_iteration": 2.579418182373047
    },
    {
      "auxiliary_loss_clip": 0.01179641,
      "auxiliary_loss_mlp": 0.01044904,
      "balance_loss_clip": 1.05955935,
      "balance_loss_mlp": 1.03450906,
      "epoch": 0.15114531353333735,
      "flos": 22820908377600.0,
      "grad_norm": 2.8060968145051794,
      "language_loss": 0.71663916,
      "learning_rate": 3.8481314704503146e-06,
      "loss": 0.73888457,
      "num_input_tokens_seen": 26628615,
      "step": 1257,
      "time_per_iteration": 2.5872514247894287
    },
    {
      "auxiliary_loss_clip": 0.01219811,
      "auxiliary_loss_mlp": 0.01040407,
      "balance_loss_clip": 1.0691011,
      "balance_loss_mlp": 1.03022075,
      "epoch": 0.15126555642397643,
      "flos": 19682674974720.0,
      "grad_norm": 2.0882302997137914,
      "language_loss": 0.87752086,
      "learning_rate": 3.847833581148895e-06,
      "loss": 0.90012306,
      "num_input_tokens_seen": 26647525,
      "step": 1258,
      "time_per_iteration": 2.5161356925964355
    },
    {
      "auxiliary_loss_clip": 0.01230845,
      "auxiliary_loss_mlp": 0.01035757,
      "balance_loss_clip": 1.06317234,
      "balance_loss_mlp": 1.02459311,
      "epoch": 0.15138579931461552,
      "flos": 28726022424960.0,
      "grad_norm": 5.407241101664347,
      "language_loss": 0.81182969,
      "learning_rate": 3.84753541153259e-06,
      "loss": 0.83449572,
      "num_input_tokens_seen": 26667095,
      "step": 1259,
      "time_per_iteration": 2.545649766921997
    },
    {
      "auxiliary_loss_clip": 0.01218021,
      "auxiliary_loss_mlp": 0.01037919,
      "balance_loss_clip": 1.06639695,
      "balance_loss_mlp": 1.02786338,
      "epoch": 0.15150604220525463,
      "flos": 22127006465280.0,
      "grad_norm": 1.5734462770400977,
      "language_loss": 0.83185923,
      "learning_rate": 3.847236961646633e-06,
      "loss": 0.85441858,
      "num_input_tokens_seen": 26686075,
      "step": 1260,
      "time_per_iteration": 2.5548789501190186
    },
    {
      "auxiliary_loss_clip": 0.01194536,
      "auxiliary_loss_mlp": 0.01042146,
      "balance_loss_clip": 1.05972373,
      "balance_loss_mlp": 1.03080893,
      "epoch": 0.1516262850958937,
      "flos": 12968708515200.0,
      "grad_norm": 2.144056087054954,
      "language_loss": 0.77974689,
      "learning_rate": 3.846938231536296e-06,
      "loss": 0.80211365,
      "num_input_tokens_seen": 26701695,
      "step": 1261,
      "time_per_iteration": 2.5173611640930176
    },
    {
      "auxiliary_loss_clip": 0.01222641,
      "auxiliary_loss_mlp": 0.01032574,
      "balance_loss_clip": 1.06780601,
      "balance_loss_mlp": 1.02248275,
      "epoch": 0.1517465279865328,
      "flos": 21797130936960.0,
      "grad_norm": 1.9857101710018616,
      "language_loss": 0.81057531,
      "learning_rate": 3.8466392212468995e-06,
      "loss": 0.8331275,
      "num_input_tokens_seen": 26721885,
      "step": 1262,
      "time_per_iteration": 2.545497179031372
    },
    {
      "auxiliary_loss_clip": 0.01102087,
      "auxiliary_loss_mlp": 0.01007892,
      "balance_loss_clip": 1.02883697,
      "balance_loss_mlp": 1.00476837,
      "epoch": 0.15186677087717187,
      "flos": 58174569901440.0,
      "grad_norm": 0.82025512844847,
      "language_loss": 0.61884898,
      "learning_rate": 3.8463399308238e-06,
      "loss": 0.63994879,
      "num_input_tokens_seen": 26780990,
      "step": 1263,
      "time_per_iteration": 3.1565325260162354
    },
    {
      "auxiliary_loss_clip": 0.01216945,
      "auxiliary_loss_mlp": 0.01040488,
      "balance_loss_clip": 1.06655836,
      "balance_loss_mlp": 1.02850199,
      "epoch": 0.15198701376781099,
      "flos": 32669696448000.0,
      "grad_norm": 1.950180375412627,
      "language_loss": 0.63794374,
      "learning_rate": 3.846040360312402e-06,
      "loss": 0.66051811,
      "num_input_tokens_seen": 26804250,
      "step": 1264,
      "time_per_iteration": 2.6542649269104004
    },
    {
      "auxiliary_loss_clip": 0.01232222,
      "auxiliary_loss_mlp": 0.01042431,
      "balance_loss_clip": 1.06537294,
      "balance_loss_mlp": 1.03154683,
      "epoch": 0.15210725665845007,
      "flos": 28402575431040.0,
      "grad_norm": 5.505775451890121,
      "language_loss": 0.81541437,
      "learning_rate": 3.8457405097581485e-06,
      "loss": 0.83816087,
      "num_input_tokens_seen": 26823240,
      "step": 1265,
      "time_per_iteration": 2.6206254959106445
    },
    {
      "auxiliary_loss_clip": 0.01172616,
      "auxiliary_loss_mlp": 0.01040508,
      "balance_loss_clip": 1.05505085,
      "balance_loss_mlp": 1.0293622,
      "epoch": 0.15222749954908915,
      "flos": 19938179393280.0,
      "grad_norm": 2.0813915623907375,
      "language_loss": 0.77849263,
      "learning_rate": 3.8454403792065275e-06,
      "loss": 0.80062389,
      "num_input_tokens_seen": 26842060,
      "step": 1266,
      "time_per_iteration": 2.6828672885894775
    },
    {
      "auxiliary_loss_clip": 0.01175289,
      "auxiliary_loss_mlp": 0.01048096,
      "balance_loss_clip": 1.05693722,
      "balance_loss_mlp": 1.03721261,
      "epoch": 0.15234774243972826,
      "flos": 21324223451520.0,
      "grad_norm": 1.9520953429284653,
      "language_loss": 0.85729051,
      "learning_rate": 3.845139968703068e-06,
      "loss": 0.87952435,
      "num_input_tokens_seen": 26859580,
      "step": 1267,
      "time_per_iteration": 2.5823848247528076
    },
    {
      "auxiliary_loss_clip": 0.01169603,
      "auxiliary_loss_mlp": 0.01044353,
      "balance_loss_clip": 1.0563761,
      "balance_loss_mlp": 1.03303993,
      "epoch": 0.15246798533036734,
      "flos": 25957812977280.0,
      "grad_norm": 1.9090533275148713,
      "language_loss": 0.82801747,
      "learning_rate": 3.844839278293342e-06,
      "loss": 0.85015702,
      "num_input_tokens_seen": 26880430,
      "step": 1268,
      "time_per_iteration": 2.7268917560577393
    },
    {
      "auxiliary_loss_clip": 0.01236778,
      "auxiliary_loss_mlp": 0.0103976,
      "balance_loss_clip": 1.0693264,
      "balance_loss_mlp": 1.02874529,
      "epoch": 0.15258822822100643,
      "flos": 25811907932160.0,
      "grad_norm": 2.8893714409464484,
      "language_loss": 0.7668348,
      "learning_rate": 3.8445383080229654e-06,
      "loss": 0.78960019,
      "num_input_tokens_seen": 26896445,
      "step": 1269,
      "time_per_iteration": 2.538482189178467
    },
    {
      "auxiliary_loss_clip": 0.01194403,
      "auxiliary_loss_mlp": 0.01034566,
      "balance_loss_clip": 1.05814433,
      "balance_loss_mlp": 1.02336049,
      "epoch": 0.1527084711116455,
      "flos": 25265455349760.0,
      "grad_norm": 2.1131678030585768,
      "language_loss": 0.7376042,
      "learning_rate": 3.844237057937593e-06,
      "loss": 0.75989389,
      "num_input_tokens_seen": 26915450,
      "step": 1270,
      "time_per_iteration": 2.585883378982544
    },
    {
      "auxiliary_loss_clip": 0.0122409,
      "auxiliary_loss_mlp": 0.0103638,
      "balance_loss_clip": 1.06435156,
      "balance_loss_mlp": 1.02497745,
      "epoch": 0.15282871400228462,
      "flos": 29240227572480.0,
      "grad_norm": 2.400238039857949,
      "language_loss": 0.77860111,
      "learning_rate": 3.843935528082926e-06,
      "loss": 0.80120587,
      "num_input_tokens_seen": 26936475,
      "step": 1271,
      "time_per_iteration": 2.593721628189087
    },
    {
      "auxiliary_loss_clip": 0.01218084,
      "auxiliary_loss_mlp": 0.01033306,
      "balance_loss_clip": 1.06450868,
      "balance_loss_mlp": 1.0230782,
      "epoch": 0.1529489568929237,
      "flos": 20882952869760.0,
      "grad_norm": 1.8662503511665753,
      "language_loss": 0.8498466,
      "learning_rate": 3.843633718504704e-06,
      "loss": 0.87236059,
      "num_input_tokens_seen": 26954920,
      "step": 1272,
      "time_per_iteration": 2.5233373641967773
    },
    {
      "auxiliary_loss_clip": 0.01184753,
      "auxiliary_loss_mlp": 0.01037267,
      "balance_loss_clip": 1.05964136,
      "balance_loss_mlp": 1.0268724,
      "epoch": 0.1530691997835628,
      "flos": 20083833043200.0,
      "grad_norm": 2.500035629287759,
      "language_loss": 0.90378654,
      "learning_rate": 3.843331629248715e-06,
      "loss": 0.92600679,
      "num_input_tokens_seen": 26972520,
      "step": 1273,
      "time_per_iteration": 2.6000401973724365
    },
    {
      "auxiliary_loss_clip": 0.01234526,
      "auxiliary_loss_mlp": 0.01033767,
      "balance_loss_clip": 1.06784344,
      "balance_loss_mlp": 1.02362239,
      "epoch": 0.1531894426742019,
      "flos": 28759814144640.0,
      "grad_norm": 2.3991455516462037,
      "language_loss": 0.76452708,
      "learning_rate": 3.843029260360782e-06,
      "loss": 0.78721005,
      "num_input_tokens_seen": 26990890,
      "step": 1274,
      "time_per_iteration": 3.3469324111938477
    },
    {
      "auxiliary_loss_clip": 0.01218224,
      "auxiliary_loss_mlp": 0.01045229,
      "balance_loss_clip": 1.06559873,
      "balance_loss_mlp": 1.03507292,
      "epoch": 0.15330968556484098,
      "flos": 22236282616320.0,
      "grad_norm": 2.5379867313463187,
      "language_loss": 0.79083681,
      "learning_rate": 3.8427266118867755e-06,
      "loss": 0.81347132,
      "num_input_tokens_seen": 27010640,
      "step": 1275,
      "time_per_iteration": 2.564117908477783
    },
    {
      "auxiliary_loss_clip": 0.01201257,
      "auxiliary_loss_mlp": 0.01036847,
      "balance_loss_clip": 1.06319606,
      "balance_loss_mlp": 1.02596307,
      "epoch": 0.15342992845548006,
      "flos": 27527504296320.0,
      "grad_norm": 2.1819681684979897,
      "language_loss": 0.82657218,
      "learning_rate": 3.842423683872608e-06,
      "loss": 0.84895325,
      "num_input_tokens_seen": 27031215,
      "step": 1276,
      "time_per_iteration": 2.5950303077697754
    },
    {
      "auxiliary_loss_clip": 0.01216163,
      "auxiliary_loss_mlp": 0.01043493,
      "balance_loss_clip": 1.0629673,
      "balance_loss_mlp": 1.03274584,
      "epoch": 0.15355017134611917,
      "flos": 19609596754560.0,
      "grad_norm": 2.445758739801298,
      "language_loss": 0.77784663,
      "learning_rate": 3.842120476364232e-06,
      "loss": 0.80044317,
      "num_input_tokens_seen": 27049665,
      "step": 1277,
      "time_per_iteration": 2.517185688018799
    },
    {
      "auxiliary_loss_clip": 0.01222064,
      "auxiliary_loss_mlp": 0.01031818,
      "balance_loss_clip": 1.06284201,
      "balance_loss_mlp": 1.02114904,
      "epoch": 0.15367041423675826,
      "flos": 18478590238080.0,
      "grad_norm": 2.086664832911439,
      "language_loss": 0.83740085,
      "learning_rate": 3.841816989407644e-06,
      "loss": 0.85993969,
      "num_input_tokens_seen": 27065155,
      "step": 1278,
      "time_per_iteration": 2.4948816299438477
    },
    {
      "auxiliary_loss_clip": 0.0118492,
      "auxiliary_loss_mlp": 0.01046029,
      "balance_loss_clip": 1.06155968,
      "balance_loss_mlp": 1.03506184,
      "epoch": 0.15379065712739734,
      "flos": 41427662342400.0,
      "grad_norm": 1.9984647715335626,
      "language_loss": 0.7709744,
      "learning_rate": 3.841513223048884e-06,
      "loss": 0.79328394,
      "num_input_tokens_seen": 27085840,
      "step": 1279,
      "time_per_iteration": 2.7593600749969482
    },
    {
      "auxiliary_loss_clip": 0.01182383,
      "auxiliary_loss_mlp": 0.0103826,
      "balance_loss_clip": 1.05660594,
      "balance_loss_mlp": 1.02737617,
      "epoch": 0.15391090001803642,
      "flos": 22054215553920.0,
      "grad_norm": 2.4294632891232535,
      "language_loss": 0.78472275,
      "learning_rate": 3.841209177334031e-06,
      "loss": 0.80692917,
      "num_input_tokens_seen": 27104200,
      "step": 1280,
      "time_per_iteration": 3.4639816284179688
    },
    {
      "auxiliary_loss_clip": 0.01213993,
      "auxiliary_loss_mlp": 0.0103739,
      "balance_loss_clip": 1.06417727,
      "balance_loss_mlp": 1.02682781,
      "epoch": 0.15403114290867553,
      "flos": 15450351258240.0,
      "grad_norm": 2.066657805289406,
      "language_loss": 0.74962389,
      "learning_rate": 3.84090485230921e-06,
      "loss": 0.7721377,
      "num_input_tokens_seen": 27122440,
      "step": 1281,
      "time_per_iteration": 3.363729238510132
    },
    {
      "auxiliary_loss_clip": 0.01233154,
      "auxiliary_loss_mlp": 0.0103607,
      "balance_loss_clip": 1.06654787,
      "balance_loss_mlp": 1.02524614,
      "epoch": 0.15415138579931462,
      "flos": 17929156826880.0,
      "grad_norm": 3.197919723147775,
      "language_loss": 0.76139849,
      "learning_rate": 3.840600248020588e-06,
      "loss": 0.78409076,
      "num_input_tokens_seen": 27139380,
      "step": 1282,
      "time_per_iteration": 2.443704843521118
    },
    {
      "auxiliary_loss_clip": 0.01206283,
      "auxiliary_loss_mlp": 0.01047181,
      "balance_loss_clip": 1.05950594,
      "balance_loss_mlp": 1.0359571,
      "epoch": 0.1542716286899537,
      "flos": 11429325296640.0,
      "grad_norm": 2.218352223811508,
      "language_loss": 0.79768419,
      "learning_rate": 3.840295364514371e-06,
      "loss": 0.82021886,
      "num_input_tokens_seen": 27156760,
      "step": 1283,
      "time_per_iteration": 2.545259952545166
    },
    {
      "auxiliary_loss_clip": 0.01202859,
      "auxiliary_loss_mlp": 0.01041116,
      "balance_loss_clip": 1.06284356,
      "balance_loss_mlp": 1.02999425,
      "epoch": 0.1543918715805928,
      "flos": 17420338719360.0,
      "grad_norm": 2.3094999921301653,
      "language_loss": 0.7854867,
      "learning_rate": 3.83999020183681e-06,
      "loss": 0.80792642,
      "num_input_tokens_seen": 27175455,
      "step": 1284,
      "time_per_iteration": 2.6204116344451904
    },
    {
      "auxiliary_loss_clip": 0.01146617,
      "auxiliary_loss_mlp": 0.01043022,
      "balance_loss_clip": 1.05383277,
      "balance_loss_mlp": 1.03269887,
      "epoch": 0.1545121144712319,
      "flos": 17786376264960.0,
      "grad_norm": 2.061416944553707,
      "language_loss": 0.78716886,
      "learning_rate": 3.839684760034199e-06,
      "loss": 0.80906522,
      "num_input_tokens_seen": 27193660,
      "step": 1285,
      "time_per_iteration": 2.7224276065826416
    },
    {
      "auxiliary_loss_clip": 0.01181168,
      "auxiliary_loss_mlp": 0.01038995,
      "balance_loss_clip": 1.05995119,
      "balance_loss_mlp": 1.02752066,
      "epoch": 0.15463235736187098,
      "flos": 28220185146240.0,
      "grad_norm": 2.1749821451854863,
      "language_loss": 0.65758193,
      "learning_rate": 3.8393790391528716e-06,
      "loss": 0.67978358,
      "num_input_tokens_seen": 27214355,
      "step": 1286,
      "time_per_iteration": 2.6035068035125732
    },
    {
      "auxiliary_loss_clip": 0.01198753,
      "auxiliary_loss_mlp": 0.01037027,
      "balance_loss_clip": 1.05927157,
      "balance_loss_mlp": 1.02650094,
      "epoch": 0.15475260025251006,
      "flos": 22856890826880.0,
      "grad_norm": 2.0392908280263033,
      "language_loss": 0.89314538,
      "learning_rate": 3.8390730392392075e-06,
      "loss": 0.9155032,
      "num_input_tokens_seen": 27234335,
      "step": 1287,
      "time_per_iteration": 2.540876626968384
    },
    {
      "auxiliary_loss_clip": 0.01234376,
      "auxiliary_loss_mlp": 0.01036589,
      "balance_loss_clip": 1.06759024,
      "balance_loss_mlp": 1.02630162,
      "epoch": 0.15487284314314917,
      "flos": 17602872658560.0,
      "grad_norm": 3.9021632639204054,
      "language_loss": 0.79294878,
      "learning_rate": 3.838766760339626e-06,
      "loss": 0.81565845,
      "num_input_tokens_seen": 27252860,
      "step": 1288,
      "time_per_iteration": 2.4512548446655273
    },
    {
      "auxiliary_loss_clip": 0.01166882,
      "auxiliary_loss_mlp": 0.01039393,
      "balance_loss_clip": 1.05561733,
      "balance_loss_mlp": 1.02815199,
      "epoch": 0.15499308603378825,
      "flos": 20082037363200.0,
      "grad_norm": 2.477251727838228,
      "language_loss": 0.79087478,
      "learning_rate": 3.838460202500587e-06,
      "loss": 0.8129375,
      "num_input_tokens_seen": 27268650,
      "step": 1289,
      "time_per_iteration": 2.575636863708496
    },
    {
      "auxiliary_loss_clip": 0.01181334,
      "auxiliary_loss_mlp": 0.01034357,
      "balance_loss_clip": 1.06307626,
      "balance_loss_mlp": 1.02257359,
      "epoch": 0.15511332892442733,
      "flos": 15918051271680.0,
      "grad_norm": 2.5256556364037914,
      "language_loss": 0.74148095,
      "learning_rate": 3.838153365768599e-06,
      "loss": 0.76363784,
      "num_input_tokens_seen": 27285160,
      "step": 1290,
      "time_per_iteration": 2.6385819911956787
    },
    {
      "auxiliary_loss_clip": 0.01185779,
      "auxiliary_loss_mlp": 0.01050897,
      "balance_loss_clip": 1.0661453,
      "balance_loss_mlp": 1.03948307,
      "epoch": 0.15523357181506645,
      "flos": 41282475569280.0,
      "grad_norm": 2.4787875437808387,
      "language_loss": 0.7563808,
      "learning_rate": 3.837846250190206e-06,
      "loss": 0.77874762,
      "num_input_tokens_seen": 27308025,
      "step": 1291,
      "time_per_iteration": 2.7535786628723145
    },
    {
      "auxiliary_loss_clip": 0.01164503,
      "auxiliary_loss_mlp": 0.0076618,
      "balance_loss_clip": 1.05694509,
      "balance_loss_mlp": 1.00158119,
      "epoch": 0.15535381470570553,
      "flos": 18478769806080.0,
      "grad_norm": 1.9988032328664185,
      "language_loss": 0.76965094,
      "learning_rate": 3.837538855811998e-06,
      "loss": 0.78895772,
      "num_input_tokens_seen": 27326200,
      "step": 1292,
      "time_per_iteration": 2.5844223499298096
    },
    {
      "auxiliary_loss_clip": 0.01209864,
      "auxiliary_loss_mlp": 0.01043264,
      "balance_loss_clip": 1.06541908,
      "balance_loss_mlp": 1.03286862,
      "epoch": 0.1554740575963446,
      "flos": 13918150759680.0,
      "grad_norm": 2.072007059468015,
      "language_loss": 0.71081936,
      "learning_rate": 3.837231182680606e-06,
      "loss": 0.73335069,
      "num_input_tokens_seen": 27344165,
      "step": 1293,
      "time_per_iteration": 2.5352165699005127
    },
    {
      "auxiliary_loss_clip": 0.01225103,
      "auxiliary_loss_mlp": 0.01037405,
      "balance_loss_clip": 1.06754553,
      "balance_loss_mlp": 1.02632999,
      "epoch": 0.1555943004869837,
      "flos": 20847078161280.0,
      "grad_norm": 1.6758742727645968,
      "language_loss": 0.75904238,
      "learning_rate": 3.836923230842706e-06,
      "loss": 0.78166747,
      "num_input_tokens_seen": 27363280,
      "step": 1294,
      "time_per_iteration": 2.5206072330474854
    },
    {
      "auxiliary_loss_clip": 0.01171255,
      "auxiliary_loss_mlp": 0.01040005,
      "balance_loss_clip": 1.05404019,
      "balance_loss_mlp": 1.02873349,
      "epoch": 0.1557145433776228,
      "flos": 22085888371200.0,
      "grad_norm": 1.8952306131757557,
      "language_loss": 0.81068122,
      "learning_rate": 3.836615000345011e-06,
      "loss": 0.83279383,
      "num_input_tokens_seen": 27381460,
      "step": 1295,
      "time_per_iteration": 2.6306960582733154
    },
    {
      "auxiliary_loss_clip": 0.01229092,
      "auxiliary_loss_mlp": 0.01036545,
      "balance_loss_clip": 1.06455302,
      "balance_loss_mlp": 1.02669811,
      "epoch": 0.1558347862682619,
      "flos": 19791987039360.0,
      "grad_norm": 1.9693740919876468,
      "language_loss": 0.77962816,
      "learning_rate": 3.836306491234282e-06,
      "loss": 0.8022846,
      "num_input_tokens_seen": 27399310,
      "step": 1296,
      "time_per_iteration": 2.4918596744537354
    },
    {
      "auxiliary_loss_clip": 0.01197237,
      "auxiliary_loss_mlp": 0.01040416,
      "balance_loss_clip": 1.06636786,
      "balance_loss_mlp": 1.03065872,
      "epoch": 0.15595502915890097,
      "flos": 17237086508160.0,
      "grad_norm": 2.256355953925597,
      "language_loss": 0.75914252,
      "learning_rate": 3.835997703557317e-06,
      "loss": 0.78151906,
      "num_input_tokens_seen": 27416050,
      "step": 1297,
      "time_per_iteration": 2.54742431640625
    },
    {
      "auxiliary_loss_clip": 0.01169121,
      "auxiliary_loss_mlp": 0.01038782,
      "balance_loss_clip": 1.05299473,
      "balance_loss_mlp": 1.02845311,
      "epoch": 0.15607527204954008,
      "flos": 19719519350400.0,
      "grad_norm": 1.645801302749267,
      "language_loss": 0.80161953,
      "learning_rate": 3.83568863736096e-06,
      "loss": 0.82369852,
      "num_input_tokens_seen": 27434920,
      "step": 1298,
      "time_per_iteration": 2.6482622623443604
    },
    {
      "auxiliary_loss_clip": 0.01187663,
      "auxiliary_loss_mlp": 0.01037645,
      "balance_loss_clip": 1.05865836,
      "balance_loss_mlp": 1.02741694,
      "epoch": 0.15619551494017916,
      "flos": 18515650095360.0,
      "grad_norm": 2.8277417804540192,
      "language_loss": 0.89143074,
      "learning_rate": 3.8353792926920975e-06,
      "loss": 0.91368371,
      "num_input_tokens_seen": 27453570,
      "step": 1299,
      "time_per_iteration": 2.560147285461426
    },
    {
      "auxiliary_loss_clip": 0.01225637,
      "auxiliary_loss_mlp": 0.01045142,
      "balance_loss_clip": 1.0686295,
      "balance_loss_mlp": 1.03401339,
      "epoch": 0.15631575783081825,
      "flos": 19902125116800.0,
      "grad_norm": 2.1531110795537747,
      "language_loss": 0.81672341,
      "learning_rate": 3.835069669597655e-06,
      "loss": 0.83943117,
      "num_input_tokens_seen": 27471960,
      "step": 1300,
      "time_per_iteration": 3.353548526763916
    },
    {
      "auxiliary_loss_clip": 0.01222735,
      "auxiliary_loss_mlp": 0.00766352,
      "balance_loss_clip": 1.0662626,
      "balance_loss_mlp": 1.0013653,
      "epoch": 0.15643600072145733,
      "flos": 20777663128320.0,
      "grad_norm": 2.1402735527529555,
      "language_loss": 0.79648507,
      "learning_rate": 3.834759768124603e-06,
      "loss": 0.81637597,
      "num_input_tokens_seen": 27490835,
      "step": 1301,
      "time_per_iteration": 2.532723903656006
    },
    {
      "auxiliary_loss_clip": 0.01192352,
      "auxiliary_loss_mlp": 0.01035551,
      "balance_loss_clip": 1.06492293,
      "balance_loss_mlp": 1.0251143,
      "epoch": 0.15655624361209644,
      "flos": 18546389159040.0,
      "grad_norm": 2.223335978249787,
      "language_loss": 0.76437336,
      "learning_rate": 3.834449588319953e-06,
      "loss": 0.78665245,
      "num_input_tokens_seen": 27508870,
      "step": 1302,
      "time_per_iteration": 2.549192190170288
    },
    {
      "auxiliary_loss_clip": 0.01215807,
      "auxiliary_loss_mlp": 0.01040587,
      "balance_loss_clip": 1.06791782,
      "balance_loss_mlp": 1.0304544,
      "epoch": 0.15667648650273552,
      "flos": 25229544727680.0,
      "grad_norm": 1.7577237529635878,
      "language_loss": 0.85125482,
      "learning_rate": 3.834139130230758e-06,
      "loss": 0.87381887,
      "num_input_tokens_seen": 27528175,
      "step": 1303,
      "time_per_iteration": 2.5474674701690674
    },
    {
      "auxiliary_loss_clip": 0.01204918,
      "auxiliary_loss_mlp": 0.01039047,
      "balance_loss_clip": 1.06090903,
      "balance_loss_mlp": 1.02859271,
      "epoch": 0.1567967293933746,
      "flos": 24827093769600.0,
      "grad_norm": 1.8089787601871836,
      "language_loss": 0.81391156,
      "learning_rate": 3.833828393904117e-06,
      "loss": 0.83635116,
      "num_input_tokens_seen": 27548455,
      "step": 1304,
      "time_per_iteration": 2.595574378967285
    },
    {
      "auxiliary_loss_clip": 0.01166488,
      "auxiliary_loss_mlp": 0.01033154,
      "balance_loss_clip": 1.05539382,
      "balance_loss_mlp": 1.02206767,
      "epoch": 0.15691697228401372,
      "flos": 19164555244800.0,
      "grad_norm": 2.3217233323079047,
      "language_loss": 0.77731085,
      "learning_rate": 3.833517379387165e-06,
      "loss": 0.79930729,
      "num_input_tokens_seen": 27564910,
      "step": 1305,
      "time_per_iteration": 2.5775556564331055
    },
    {
      "auxiliary_loss_clip": 0.01222565,
      "auxiliary_loss_mlp": 0.01041748,
      "balance_loss_clip": 1.06769645,
      "balance_loss_mlp": 1.03103673,
      "epoch": 0.1570372151746528,
      "flos": 24790931752320.0,
      "grad_norm": 3.092072864454218,
      "language_loss": 0.88825744,
      "learning_rate": 3.833206086727085e-06,
      "loss": 0.91090059,
      "num_input_tokens_seen": 27584260,
      "step": 1306,
      "time_per_iteration": 2.5487661361694336
    },
    {
      "auxiliary_loss_clip": 0.01190021,
      "auxiliary_loss_mlp": 0.01037477,
      "balance_loss_clip": 1.05850053,
      "balance_loss_mlp": 1.02743971,
      "epoch": 0.15715745806529188,
      "flos": 24863650836480.0,
      "grad_norm": 1.963992388615502,
      "language_loss": 0.70863497,
      "learning_rate": 3.8328945159710994e-06,
      "loss": 0.73090994,
      "num_input_tokens_seen": 27604440,
      "step": 1307,
      "time_per_iteration": 4.979633331298828
    },
    {
      "auxiliary_loss_clip": 0.01226948,
      "auxiliary_loss_mlp": 0.00765368,
      "balance_loss_clip": 1.06996155,
      "balance_loss_mlp": 1.00123274,
      "epoch": 0.157277700955931,
      "flos": 21872148491520.0,
      "grad_norm": 2.3597321218678435,
      "language_loss": 0.88840675,
      "learning_rate": 3.832582667166473e-06,
      "loss": 0.90832996,
      "num_input_tokens_seen": 27624250,
      "step": 1308,
      "time_per_iteration": 2.510880947113037
    },
    {
      "auxiliary_loss_clip": 0.01205869,
      "auxiliary_loss_mlp": 0.01038475,
      "balance_loss_clip": 1.06444168,
      "balance_loss_mlp": 1.02699482,
      "epoch": 0.15739794384657008,
      "flos": 24533344344960.0,
      "grad_norm": 1.741141653743624,
      "language_loss": 0.82020366,
      "learning_rate": 3.8322705403605125e-06,
      "loss": 0.84264708,
      "num_input_tokens_seen": 27644595,
      "step": 1309,
      "time_per_iteration": 2.5459489822387695
    },
    {
      "auxiliary_loss_clip": 0.01195501,
      "auxiliary_loss_mlp": 0.01037319,
      "balance_loss_clip": 1.06246519,
      "balance_loss_mlp": 1.02796674,
      "epoch": 0.15751818673720916,
      "flos": 17745329998080.0,
      "grad_norm": 2.433016627240624,
      "language_loss": 0.80914116,
      "learning_rate": 3.831958135600568e-06,
      "loss": 0.8314693,
      "num_input_tokens_seen": 27662145,
      "step": 1310,
      "time_per_iteration": 2.481736421585083
    },
    {
      "auxiliary_loss_clip": 0.01218928,
      "auxiliary_loss_mlp": 0.01034927,
      "balance_loss_clip": 1.06626499,
      "balance_loss_mlp": 1.02568829,
      "epoch": 0.15763842962784824,
      "flos": 17858520731520.0,
      "grad_norm": 1.9674667067269547,
      "language_loss": 0.79463673,
      "learning_rate": 3.831645452934032e-06,
      "loss": 0.81717527,
      "num_input_tokens_seen": 27680575,
      "step": 1311,
      "time_per_iteration": 2.461761474609375
    },
    {
      "auxiliary_loss_clip": 0.0123525,
      "auxiliary_loss_mlp": 0.01044459,
      "balance_loss_clip": 1.07023692,
      "balance_loss_mlp": 1.03440952,
      "epoch": 0.15775867251848735,
      "flos": 26980908059520.0,
      "grad_norm": 1.8567165983109104,
      "language_loss": 0.80104721,
      "learning_rate": 3.831332492408336e-06,
      "loss": 0.82384431,
      "num_input_tokens_seen": 27701985,
      "step": 1312,
      "time_per_iteration": 2.5449059009552
    },
    {
      "auxiliary_loss_clip": 0.01199619,
      "auxiliary_loss_mlp": 0.01030943,
      "balance_loss_clip": 1.06210828,
      "balance_loss_mlp": 1.0205543,
      "epoch": 0.15787891540912644,
      "flos": 19240398812160.0,
      "grad_norm": 2.0189939879799477,
      "language_loss": 0.69460416,
      "learning_rate": 3.831019254070957e-06,
      "loss": 0.71690983,
      "num_input_tokens_seen": 27719770,
      "step": 1313,
      "time_per_iteration": 2.5115060806274414
    },
    {
      "auxiliary_loss_clip": 0.01175294,
      "auxiliary_loss_mlp": 0.01033725,
      "balance_loss_clip": 1.05974579,
      "balance_loss_mlp": 1.02364564,
      "epoch": 0.15799915829976552,
      "flos": 27271102037760.0,
      "grad_norm": 2.720418961676346,
      "language_loss": 0.95244241,
      "learning_rate": 3.8307057379694135e-06,
      "loss": 0.9745326,
      "num_input_tokens_seen": 27739105,
      "step": 1314,
      "time_per_iteration": 2.6371397972106934
    },
    {
      "auxiliary_loss_clip": 0.01231751,
      "auxiliary_loss_mlp": 0.01041856,
      "balance_loss_clip": 1.06557298,
      "balance_loss_mlp": 1.03184867,
      "epoch": 0.15811940119040463,
      "flos": 20405520270720.0,
      "grad_norm": 2.0725347502638236,
      "language_loss": 0.82151711,
      "learning_rate": 3.830391944151264e-06,
      "loss": 0.84425312,
      "num_input_tokens_seen": 27754985,
      "step": 1315,
      "time_per_iteration": 2.4328408241271973
    },
    {
      "auxiliary_loss_clip": 0.01200777,
      "auxiliary_loss_mlp": 0.01043284,
      "balance_loss_clip": 1.06002641,
      "balance_loss_mlp": 1.03357458,
      "epoch": 0.1582396440810437,
      "flos": 32599347661440.0,
      "grad_norm": 1.932466968115719,
      "language_loss": 0.67339623,
      "learning_rate": 3.830077872664114e-06,
      "loss": 0.6958369,
      "num_input_tokens_seen": 27776110,
      "step": 1316,
      "time_per_iteration": 2.6023385524749756
    },
    {
      "auxiliary_loss_clip": 0.01154208,
      "auxiliary_loss_mlp": 0.01042564,
      "balance_loss_clip": 1.05398107,
      "balance_loss_mlp": 1.03291392,
      "epoch": 0.1583598869716828,
      "flos": 33800559310080.0,
      "grad_norm": 1.7667233206382023,
      "language_loss": 0.72862113,
      "learning_rate": 3.829763523555604e-06,
      "loss": 0.75058889,
      "num_input_tokens_seen": 27796510,
      "step": 1317,
      "time_per_iteration": 2.7155137062072754
    },
    {
      "auxiliary_loss_clip": 0.01211408,
      "auxiliary_loss_mlp": 0.01035761,
      "balance_loss_clip": 1.06838012,
      "balance_loss_mlp": 1.02619505,
      "epoch": 0.15848012986232188,
      "flos": 24681332378880.0,
      "grad_norm": 2.114874555651769,
      "language_loss": 0.77905083,
      "learning_rate": 3.829448896873423e-06,
      "loss": 0.80152249,
      "num_input_tokens_seen": 27815610,
      "step": 1318,
      "time_per_iteration": 2.543682813644409
    },
    {
      "auxiliary_loss_clip": 0.01159997,
      "auxiliary_loss_mlp": 0.00765157,
      "balance_loss_clip": 1.06019151,
      "balance_loss_mlp": 1.00129008,
      "epoch": 0.158600372752961,
      "flos": 22602068766720.0,
      "grad_norm": 1.7690122907861627,
      "language_loss": 0.79127955,
      "learning_rate": 3.829133992665299e-06,
      "loss": 0.81053114,
      "num_input_tokens_seen": 27834735,
      "step": 1319,
      "time_per_iteration": 2.610743761062622
    },
    {
      "auxiliary_loss_clip": 0.01205086,
      "auxiliary_loss_mlp": 0.01038033,
      "balance_loss_clip": 1.06355608,
      "balance_loss_mlp": 1.02825737,
      "epoch": 0.15872061564360007,
      "flos": 27927944092800.0,
      "grad_norm": 2.1452207925442734,
      "language_loss": 0.88678765,
      "learning_rate": 3.828818810979002e-06,
      "loss": 0.90921885,
      "num_input_tokens_seen": 27853065,
      "step": 1320,
      "time_per_iteration": 2.5379014015197754
    },
    {
      "auxiliary_loss_clip": 0.01232221,
      "auxiliary_loss_mlp": 0.01041007,
      "balance_loss_clip": 1.07075691,
      "balance_loss_mlp": 1.0312196,
      "epoch": 0.15884085853423915,
      "flos": 23696805525120.0,
      "grad_norm": 1.8274815272764542,
      "language_loss": 0.80584955,
      "learning_rate": 3.8285033518623454e-06,
      "loss": 0.82858193,
      "num_input_tokens_seen": 27873315,
      "step": 1321,
      "time_per_iteration": 2.4854278564453125
    },
    {
      "auxiliary_loss_clip": 0.01221472,
      "auxiliary_loss_mlp": 0.01040578,
      "balance_loss_clip": 1.06742227,
      "balance_loss_mlp": 1.02958679,
      "epoch": 0.15896110142487826,
      "flos": 23112359331840.0,
      "grad_norm": 2.692626492663524,
      "language_loss": 0.81532472,
      "learning_rate": 3.8281876153631845e-06,
      "loss": 0.83794522,
      "num_input_tokens_seen": 27890070,
      "step": 1322,
      "time_per_iteration": 2.507112979888916
    },
    {
      "auxiliary_loss_clip": 0.01166667,
      "auxiliary_loss_mlp": 0.0104213,
      "balance_loss_clip": 1.05711365,
      "balance_loss_mlp": 1.03118706,
      "epoch": 0.15908134431551735,
      "flos": 14685238632960.0,
      "grad_norm": 1.883383624501825,
      "language_loss": 0.64697874,
      "learning_rate": 3.827871601529416e-06,
      "loss": 0.66906679,
      "num_input_tokens_seen": 27908590,
      "step": 1323,
      "time_per_iteration": 2.5850954055786133
    },
    {
      "auxiliary_loss_clip": 0.01179731,
      "auxiliary_loss_mlp": 0.01040591,
      "balance_loss_clip": 1.05840039,
      "balance_loss_mlp": 1.0305593,
      "epoch": 0.15920158720615643,
      "flos": 20193611984640.0,
      "grad_norm": 1.7632585409258437,
      "language_loss": 0.80515623,
      "learning_rate": 3.827555310408979e-06,
      "loss": 0.82735944,
      "num_input_tokens_seen": 27927985,
      "step": 1324,
      "time_per_iteration": 2.5570547580718994
    },
    {
      "auxiliary_loss_clip": 0.01181059,
      "auxiliary_loss_mlp": 0.01038098,
      "balance_loss_clip": 1.06348395,
      "balance_loss_mlp": 1.02760148,
      "epoch": 0.1593218300967955,
      "flos": 24826626892800.0,
      "grad_norm": 1.7186704152473238,
      "language_loss": 0.82870293,
      "learning_rate": 3.827238742049854e-06,
      "loss": 0.85089451,
      "num_input_tokens_seen": 27948280,
      "step": 1325,
      "time_per_iteration": 2.654414415359497
    },
    {
      "auxiliary_loss_clip": 0.01229299,
      "auxiliary_loss_mlp": 0.01037082,
      "balance_loss_clip": 1.06540775,
      "balance_loss_mlp": 1.02651358,
      "epoch": 0.15944207298743462,
      "flos": 28328707111680.0,
      "grad_norm": 1.949110916574243,
      "language_loss": 0.52068645,
      "learning_rate": 3.826921896500066e-06,
      "loss": 0.54335022,
      "num_input_tokens_seen": 27969565,
      "step": 1326,
      "time_per_iteration": 2.5206069946289062
    },
    {
      "auxiliary_loss_clip": 0.01190421,
      "auxiliary_loss_mlp": 0.01036531,
      "balance_loss_clip": 1.06250894,
      "balance_loss_mlp": 1.02545023,
      "epoch": 0.1595623158780737,
      "flos": 22964838174720.0,
      "grad_norm": 2.819476890216086,
      "language_loss": 0.7808429,
      "learning_rate": 3.826604773807678e-06,
      "loss": 0.80311245,
      "num_input_tokens_seen": 27987540,
      "step": 1327,
      "time_per_iteration": 3.3958075046539307
    },
    {
      "auxiliary_loss_clip": 0.01196276,
      "auxiliary_loss_mlp": 0.01035758,
      "balance_loss_clip": 1.05713904,
      "balance_loss_mlp": 1.02485597,
      "epoch": 0.1596825587687128,
      "flos": 19710540950400.0,
      "grad_norm": 2.73269049048548,
      "language_loss": 0.73612595,
      "learning_rate": 3.826287374020798e-06,
      "loss": 0.75844634,
      "num_input_tokens_seen": 28002345,
      "step": 1328,
      "time_per_iteration": 2.5247321128845215
    },
    {
      "auxiliary_loss_clip": 0.01231704,
      "auxiliary_loss_mlp": 0.01036938,
      "balance_loss_clip": 1.06858444,
      "balance_loss_mlp": 1.02735353,
      "epoch": 0.1598028016593519,
      "flos": 22637727993600.0,
      "grad_norm": 2.068608124183978,
      "language_loss": 0.82837182,
      "learning_rate": 3.825969697187575e-06,
      "loss": 0.85105824,
      "num_input_tokens_seen": 28021675,
      "step": 1329,
      "time_per_iteration": 2.4613306522369385
    },
    {
      "auxiliary_loss_clip": 0.01181687,
      "auxiliary_loss_mlp": 0.01031509,
      "balance_loss_clip": 1.05808735,
      "balance_loss_mlp": 1.02122116,
      "epoch": 0.15992304454999098,
      "flos": 20482908122880.0,
      "grad_norm": 1.8308811204645985,
      "language_loss": 0.69524753,
      "learning_rate": 3.8256517433562015e-06,
      "loss": 0.71737945,
      "num_input_tokens_seen": 28039615,
      "step": 1330,
      "time_per_iteration": 2.5553975105285645
    },
    {
      "auxiliary_loss_clip": 0.01227855,
      "auxiliary_loss_mlp": 0.01036025,
      "balance_loss_clip": 1.06534481,
      "balance_loss_mlp": 1.02704883,
      "epoch": 0.16004328744063007,
      "flos": 17676094533120.0,
      "grad_norm": 2.7546571020653694,
      "language_loss": 0.91858697,
      "learning_rate": 3.82533351257491e-06,
      "loss": 0.94122577,
      "num_input_tokens_seen": 28057565,
      "step": 1331,
      "time_per_iteration": 2.4413905143737793
    },
    {
      "auxiliary_loss_clip": 0.0121539,
      "auxiliary_loss_mlp": 0.01036362,
      "balance_loss_clip": 1.06783652,
      "balance_loss_mlp": 1.02681303,
      "epoch": 0.16016353033126918,
      "flos": 24098717779200.0,
      "grad_norm": 1.7333625243184108,
      "language_loss": 0.88734305,
      "learning_rate": 3.825015004891975e-06,
      "loss": 0.90986061,
      "num_input_tokens_seen": 28076305,
      "step": 1332,
      "time_per_iteration": 2.5300896167755127
    },
    {
      "auxiliary_loss_clip": 0.01211026,
      "auxiliary_loss_mlp": 0.01030573,
      "balance_loss_clip": 1.06296003,
      "balance_loss_mlp": 1.02062535,
      "epoch": 0.16028377322190826,
      "flos": 27634841112960.0,
      "grad_norm": 1.8013514686793173,
      "language_loss": 0.7596792,
      "learning_rate": 3.824696220355716e-06,
      "loss": 0.78209519,
      "num_input_tokens_seen": 28097895,
      "step": 1333,
      "time_per_iteration": 2.5350000858306885
    },
    {
      "auxiliary_loss_clip": 0.01196898,
      "auxiliary_loss_mlp": 0.01044124,
      "balance_loss_clip": 1.06147027,
      "balance_loss_mlp": 1.03393137,
      "epoch": 0.16040401611254734,
      "flos": 20961202648320.0,
      "grad_norm": 1.5950696886137068,
      "language_loss": 0.78940523,
      "learning_rate": 3.824377159014491e-06,
      "loss": 0.81181544,
      "num_input_tokens_seen": 28118790,
      "step": 1334,
      "time_per_iteration": 3.5115861892700195
    },
    {
      "auxiliary_loss_clip": 0.01211728,
      "auxiliary_loss_mlp": 0.01036307,
      "balance_loss_clip": 1.06505656,
      "balance_loss_mlp": 1.02653742,
      "epoch": 0.16052425900318643,
      "flos": 21247051080960.0,
      "grad_norm": 2.7081115425188123,
      "language_loss": 0.85111851,
      "learning_rate": 3.824057820916702e-06,
      "loss": 0.87359881,
      "num_input_tokens_seen": 28135995,
      "step": 1335,
      "time_per_iteration": 2.549170970916748
    },
    {
      "auxiliary_loss_clip": 0.01201159,
      "auxiliary_loss_mlp": 0.01031602,
      "balance_loss_clip": 1.06250501,
      "balance_loss_mlp": 1.02070642,
      "epoch": 0.16064450189382554,
      "flos": 15524004096000.0,
      "grad_norm": 5.70837034613212,
      "language_loss": 0.71691126,
      "learning_rate": 3.8237382061107904e-06,
      "loss": 0.73923886,
      "num_input_tokens_seen": 28152715,
      "step": 1336,
      "time_per_iteration": 2.482983112335205
    },
    {
      "auxiliary_loss_clip": 0.01125025,
      "auxiliary_loss_mlp": 0.01045104,
      "balance_loss_clip": 1.04826617,
      "balance_loss_mlp": 1.03539991,
      "epoch": 0.16076474478446462,
      "flos": 21178497974400.0,
      "grad_norm": 1.9038784275590552,
      "language_loss": 0.78527892,
      "learning_rate": 3.823418314645243e-06,
      "loss": 0.80698025,
      "num_input_tokens_seen": 28171590,
      "step": 1337,
      "time_per_iteration": 2.6402690410614014
    },
    {
      "auxiliary_loss_clip": 0.01151947,
      "auxiliary_loss_mlp": 0.01040552,
      "balance_loss_clip": 1.05759811,
      "balance_loss_mlp": 1.03121805,
      "epoch": 0.1608849876751037,
      "flos": 18366476912640.0,
      "grad_norm": 2.2080051398346994,
      "language_loss": 0.75382489,
      "learning_rate": 3.823098146568588e-06,
      "loss": 0.77574986,
      "num_input_tokens_seen": 28191295,
      "step": 1338,
      "time_per_iteration": 2.5634398460388184
    },
    {
      "auxiliary_loss_clip": 0.01211793,
      "auxiliary_loss_mlp": 0.01037522,
      "balance_loss_clip": 1.06369948,
      "balance_loss_mlp": 1.02862263,
      "epoch": 0.1610052305657428,
      "flos": 29497024880640.0,
      "grad_norm": 1.6013405989881024,
      "language_loss": 0.71679163,
      "learning_rate": 3.822777701929394e-06,
      "loss": 0.73928481,
      "num_input_tokens_seen": 28213120,
      "step": 1339,
      "time_per_iteration": 2.5468263626098633
    },
    {
      "auxiliary_loss_clip": 0.01200914,
      "auxiliary_loss_mlp": 0.01041717,
      "balance_loss_clip": 1.05954421,
      "balance_loss_mlp": 1.03128624,
      "epoch": 0.1611254734563819,
      "flos": 26797871329920.0,
      "grad_norm": 1.8032768835029767,
      "language_loss": 0.73576581,
      "learning_rate": 3.8224569807762714e-06,
      "loss": 0.75819218,
      "num_input_tokens_seen": 28232440,
      "step": 1340,
      "time_per_iteration": 2.5195319652557373
    },
    {
      "auxiliary_loss_clip": 0.01147128,
      "auxiliary_loss_mlp": 0.01042173,
      "balance_loss_clip": 1.05072176,
      "balance_loss_mlp": 1.0315274,
      "epoch": 0.16124571634702098,
      "flos": 22419570741120.0,
      "grad_norm": 1.8950014598778249,
      "language_loss": 0.76562715,
      "learning_rate": 3.822135983157873e-06,
      "loss": 0.78752011,
      "num_input_tokens_seen": 28251715,
      "step": 1341,
      "time_per_iteration": 2.598649501800537
    },
    {
      "auxiliary_loss_clip": 0.01224813,
      "auxiliary_loss_mlp": 0.00765386,
      "balance_loss_clip": 1.06451225,
      "balance_loss_mlp": 1.00135827,
      "epoch": 0.16136595923766006,
      "flos": 10999116103680.0,
      "grad_norm": 2.945320386367759,
      "language_loss": 0.84666026,
      "learning_rate": 3.821814709122896e-06,
      "loss": 0.86656225,
      "num_input_tokens_seen": 28269765,
      "step": 1342,
      "time_per_iteration": 2.530895709991455
    },
    {
      "auxiliary_loss_clip": 0.01196426,
      "auxiliary_loss_mlp": 0.01036763,
      "balance_loss_clip": 1.06143856,
      "balance_loss_mlp": 1.02738142,
      "epoch": 0.16148620212829917,
      "flos": 21214983214080.0,
      "grad_norm": 2.1281212295278324,
      "language_loss": 0.84891212,
      "learning_rate": 3.821493158720076e-06,
      "loss": 0.87124407,
      "num_input_tokens_seen": 28288870,
      "step": 1343,
      "time_per_iteration": 2.579946517944336
    },
    {
      "auxiliary_loss_clip": 0.01181437,
      "auxiliary_loss_mlp": 0.01033738,
      "balance_loss_clip": 1.05649877,
      "balance_loss_mlp": 1.0229609,
      "epoch": 0.16160644501893826,
      "flos": 16758468760320.0,
      "grad_norm": 2.8835141669314845,
      "language_loss": 0.73465884,
      "learning_rate": 3.821171331998191e-06,
      "loss": 0.75681055,
      "num_input_tokens_seen": 28305400,
      "step": 1344,
      "time_per_iteration": 2.512822389602661
    },
    {
      "auxiliary_loss_clip": 0.01112276,
      "auxiliary_loss_mlp": 0.01009186,
      "balance_loss_clip": 1.04560614,
      "balance_loss_mlp": 1.0065279,
      "epoch": 0.16172668790957734,
      "flos": 64444967308800.0,
      "grad_norm": 0.9347021770854987,
      "language_loss": 0.5448283,
      "learning_rate": 3.820849229006064e-06,
      "loss": 0.5660429,
      "num_input_tokens_seen": 28373150,
      "step": 1345,
      "time_per_iteration": 3.2853260040283203
    },
    {
      "auxiliary_loss_clip": 0.01230609,
      "auxiliary_loss_mlp": 0.01033698,
      "balance_loss_clip": 1.0661428,
      "balance_loss_mlp": 1.0237323,
      "epoch": 0.16184693080021645,
      "flos": 23257689759360.0,
      "grad_norm": 2.5152504911007587,
      "language_loss": 0.70680386,
      "learning_rate": 3.8205268497925564e-06,
      "loss": 0.72944701,
      "num_input_tokens_seen": 28393620,
      "step": 1346,
      "time_per_iteration": 2.4712090492248535
    },
    {
      "auxiliary_loss_clip": 0.01229389,
      "auxiliary_loss_mlp": 0.01038249,
      "balance_loss_clip": 1.06678486,
      "balance_loss_mlp": 1.02845621,
      "epoch": 0.16196717369085553,
      "flos": 17451113696640.0,
      "grad_norm": 2.235659299754506,
      "language_loss": 0.78539979,
      "learning_rate": 3.8202041944065725e-06,
      "loss": 0.8080762,
      "num_input_tokens_seen": 28409440,
      "step": 1347,
      "time_per_iteration": 2.425034761428833
    },
    {
      "auxiliary_loss_clip": 0.01230157,
      "auxiliary_loss_mlp": 0.01036703,
      "balance_loss_clip": 1.06876183,
      "balance_loss_mlp": 1.02648044,
      "epoch": 0.16208741658149461,
      "flos": 23873377806720.0,
      "grad_norm": 1.7500583286342155,
      "language_loss": 0.73807669,
      "learning_rate": 3.819881262897061e-06,
      "loss": 0.76074529,
      "num_input_tokens_seen": 28427575,
      "step": 1348,
      "time_per_iteration": 2.4973952770233154
    },
    {
      "auxiliary_loss_clip": 0.01186739,
      "auxiliary_loss_mlp": 0.01034755,
      "balance_loss_clip": 1.06578946,
      "balance_loss_mlp": 1.02391839,
      "epoch": 0.1622076594721337,
      "flos": 25884806584320.0,
      "grad_norm": 1.884018408940885,
      "language_loss": 0.73562455,
      "learning_rate": 3.819558055313008e-06,
      "loss": 0.7578395,
      "num_input_tokens_seen": 28448260,
      "step": 1349,
      "time_per_iteration": 2.6526899337768555
    },
    {
      "auxiliary_loss_clip": 0.01218988,
      "auxiliary_loss_mlp": 0.01041079,
      "balance_loss_clip": 1.06480014,
      "balance_loss_mlp": 1.03117836,
      "epoch": 0.1623279023627728,
      "flos": 21539759011200.0,
      "grad_norm": 2.0984279190709736,
      "language_loss": 0.77566063,
      "learning_rate": 3.819234571703444e-06,
      "loss": 0.79826134,
      "num_input_tokens_seen": 28467085,
      "step": 1350,
      "time_per_iteration": 2.4966864585876465
    },
    {
      "auxiliary_loss_clip": 0.01204774,
      "auxiliary_loss_mlp": 0.01040655,
      "balance_loss_clip": 1.06003189,
      "balance_loss_mlp": 1.03030181,
      "epoch": 0.1624481452534119,
      "flos": 22085421494400.0,
      "grad_norm": 1.8600360865161258,
      "language_loss": 0.85695267,
      "learning_rate": 3.8189108121174435e-06,
      "loss": 0.87940693,
      "num_input_tokens_seen": 28486850,
      "step": 1351,
      "time_per_iteration": 2.5167431831359863
    },
    {
      "auxiliary_loss_clip": 0.01180566,
      "auxiliary_loss_mlp": 0.01039758,
      "balance_loss_clip": 1.06437624,
      "balance_loss_mlp": 1.0297029,
      "epoch": 0.16256838814405097,
      "flos": 27087490690560.0,
      "grad_norm": 1.8317684110019719,
      "language_loss": 0.83584058,
      "learning_rate": 3.818586776604118e-06,
      "loss": 0.85804385,
      "num_input_tokens_seen": 28507490,
      "step": 1352,
      "time_per_iteration": 2.5973947048187256
    },
    {
      "auxiliary_loss_clip": 0.01195985,
      "auxiliary_loss_mlp": 0.0104071,
      "balance_loss_clip": 1.06215477,
      "balance_loss_mlp": 1.03058279,
      "epoch": 0.16268863103469008,
      "flos": 20120354196480.0,
      "grad_norm": 1.8695855380481676,
      "language_loss": 0.61486769,
      "learning_rate": 3.818262465212625e-06,
      "loss": 0.63723469,
      "num_input_tokens_seen": 28527615,
      "step": 1353,
      "time_per_iteration": 2.514058828353882
    },
    {
      "auxiliary_loss_clip": 0.01205958,
      "auxiliary_loss_mlp": 0.01047808,
      "balance_loss_clip": 1.06521297,
      "balance_loss_mlp": 1.03678751,
      "epoch": 0.16280887392532917,
      "flos": 18332792933760.0,
      "grad_norm": 1.8389471092448109,
      "language_loss": 0.77340096,
      "learning_rate": 3.817937877992161e-06,
      "loss": 0.79593861,
      "num_input_tokens_seen": 28544910,
      "step": 1354,
      "time_per_iteration": 3.2517943382263184
    },
    {
      "auxiliary_loss_clip": 0.01183704,
      "auxiliary_loss_mlp": 0.00766408,
      "balance_loss_clip": 1.05734324,
      "balance_loss_mlp": 1.0012275,
      "epoch": 0.16292911681596825,
      "flos": 11874330892800.0,
      "grad_norm": 4.189629685814826,
      "language_loss": 0.85334384,
      "learning_rate": 3.817613014991967e-06,
      "loss": 0.87284499,
      "num_input_tokens_seen": 28561050,
      "step": 1355,
      "time_per_iteration": 2.5084424018859863
    },
    {
      "auxiliary_loss_clip": 0.01175051,
      "auxiliary_loss_mlp": 0.01035481,
      "balance_loss_clip": 1.05732155,
      "balance_loss_mlp": 1.02522874,
      "epoch": 0.16304935970660733,
      "flos": 26103466627200.0,
      "grad_norm": 2.079017450339228,
      "language_loss": 0.76704466,
      "learning_rate": 3.817287876261323e-06,
      "loss": 0.78915,
      "num_input_tokens_seen": 28581385,
      "step": 1356,
      "time_per_iteration": 2.577904462814331
    },
    {
      "auxiliary_loss_clip": 0.01194618,
      "auxiliary_loss_mlp": 0.01033754,
      "balance_loss_clip": 1.06390715,
      "balance_loss_mlp": 1.02274537,
      "epoch": 0.16316960259724644,
      "flos": 29351945848320.0,
      "grad_norm": 1.9203192633913142,
      "language_loss": 0.80038011,
      "learning_rate": 3.816962461849553e-06,
      "loss": 0.82266378,
      "num_input_tokens_seen": 28603255,
      "step": 1357,
      "time_per_iteration": 2.577784538269043
    },
    {
      "auxiliary_loss_clip": 0.01193545,
      "auxiliary_loss_mlp": 0.01038386,
      "balance_loss_clip": 1.06450117,
      "balance_loss_mlp": 1.02784228,
      "epoch": 0.16328984548788553,
      "flos": 20886759711360.0,
      "grad_norm": 1.9062865964148028,
      "language_loss": 0.84846306,
      "learning_rate": 3.8166367718060235e-06,
      "loss": 0.87078232,
      "num_input_tokens_seen": 28623145,
      "step": 1358,
      "time_per_iteration": 2.559354305267334
    },
    {
      "auxiliary_loss_clip": 0.01209237,
      "auxiliary_loss_mlp": 0.01034096,
      "balance_loss_clip": 1.06168008,
      "balance_loss_mlp": 1.02407002,
      "epoch": 0.1634100883785246,
      "flos": 18041090584320.0,
      "grad_norm": 2.463957080596287,
      "language_loss": 0.76440525,
      "learning_rate": 3.816310806180139e-06,
      "loss": 0.78683859,
      "num_input_tokens_seen": 28641555,
      "step": 1359,
      "time_per_iteration": 2.459226131439209
    },
    {
      "auxiliary_loss_clip": 0.01192513,
      "auxiliary_loss_mlp": 0.01042625,
      "balance_loss_clip": 1.0613699,
      "balance_loss_mlp": 1.03256989,
      "epoch": 0.16353033126916372,
      "flos": 24572128055040.0,
      "grad_norm": 1.6189070321330312,
      "language_loss": 0.8088994,
      "learning_rate": 3.81598456502135e-06,
      "loss": 0.83125079,
      "num_input_tokens_seen": 28661575,
      "step": 1360,
      "time_per_iteration": 3.468073844909668
    },
    {
      "auxiliary_loss_clip": 0.01196332,
      "auxiliary_loss_mlp": 0.01040531,
      "balance_loss_clip": 1.0652957,
      "balance_loss_mlp": 1.0299809,
      "epoch": 0.1636505741598028,
      "flos": 19892895321600.0,
      "grad_norm": 2.000039107262487,
      "language_loss": 0.87098432,
      "learning_rate": 3.8156580483791455e-06,
      "loss": 0.89335293,
      "num_input_tokens_seen": 28676765,
      "step": 1361,
      "time_per_iteration": 3.2954471111297607
    },
    {
      "auxiliary_loss_clip": 0.01230967,
      "auxiliary_loss_mlp": 0.01035631,
      "balance_loss_clip": 1.0666666,
      "balance_loss_mlp": 1.025653,
      "epoch": 0.16377081705044189,
      "flos": 28402611344640.0,
      "grad_norm": 2.5208244661310597,
      "language_loss": 0.7678265,
      "learning_rate": 3.815331256303059e-06,
      "loss": 0.79049248,
      "num_input_tokens_seen": 28696795,
      "step": 1362,
      "time_per_iteration": 2.5115602016448975
    },
    {
      "auxiliary_loss_clip": 0.01180133,
      "auxiliary_loss_mlp": 0.01035799,
      "balance_loss_clip": 1.06296825,
      "balance_loss_mlp": 1.02529657,
      "epoch": 0.163891059941081,
      "flos": 21908059113600.0,
      "grad_norm": 2.1770114400105514,
      "language_loss": 0.77557611,
      "learning_rate": 3.815004188842665e-06,
      "loss": 0.79773539,
      "num_input_tokens_seen": 28714835,
      "step": 1363,
      "time_per_iteration": 2.5333497524261475
    },
    {
      "auxiliary_loss_clip": 0.01193249,
      "auxiliary_loss_mlp": 0.01036788,
      "balance_loss_clip": 1.05873311,
      "balance_loss_mlp": 1.02604723,
      "epoch": 0.16401130283172008,
      "flos": 26797619934720.0,
      "grad_norm": 1.6382319331718556,
      "language_loss": 0.80046415,
      "learning_rate": 3.814676846047578e-06,
      "loss": 0.82276446,
      "num_input_tokens_seen": 28735710,
      "step": 1364,
      "time_per_iteration": 2.56845760345459
    },
    {
      "auxiliary_loss_clip": 0.01210519,
      "auxiliary_loss_mlp": 0.01043548,
      "balance_loss_clip": 1.06334269,
      "balance_loss_mlp": 1.03334308,
      "epoch": 0.16413154572235916,
      "flos": 32997417160320.0,
      "grad_norm": 1.6800424987292948,
      "language_loss": 0.70344496,
      "learning_rate": 3.8143492279674565e-06,
      "loss": 0.72598565,
      "num_input_tokens_seen": 28758405,
      "step": 1365,
      "time_per_iteration": 2.58172607421875
    },
    {
      "auxiliary_loss_clip": 0.01109381,
      "auxiliary_loss_mlp": 0.01006006,
      "balance_loss_clip": 1.04409742,
      "balance_loss_mlp": 1.0031327,
      "epoch": 0.16425178861299825,
      "flos": 40113622074240.0,
      "grad_norm": 0.8419780044993703,
      "language_loss": 0.58476913,
      "learning_rate": 3.8140213346519997e-06,
      "loss": 0.60592306,
      "num_input_tokens_seen": 28809000,
      "step": 1366,
      "time_per_iteration": 2.869281530380249
    },
    {
      "auxiliary_loss_clip": 0.01167376,
      "auxiliary_loss_mlp": 0.01036153,
      "balance_loss_clip": 1.05608976,
      "balance_loss_mlp": 1.02591872,
      "epoch": 0.16437203150363736,
      "flos": 25447486498560.0,
      "grad_norm": 1.733083969531301,
      "language_loss": 0.77403069,
      "learning_rate": 3.813693166150948e-06,
      "loss": 0.79606593,
      "num_input_tokens_seen": 28829210,
      "step": 1367,
      "time_per_iteration": 2.5883116722106934
    },
    {
      "auxiliary_loss_clip": 0.01174026,
      "auxiliary_loss_mlp": 0.01035768,
      "balance_loss_clip": 1.0583725,
      "balance_loss_mlp": 1.02519977,
      "epoch": 0.16449227439427644,
      "flos": 23476888506240.0,
      "grad_norm": 2.2346238575865276,
      "language_loss": 0.85482037,
      "learning_rate": 3.813364722514086e-06,
      "loss": 0.87691832,
      "num_input_tokens_seen": 28847545,
      "step": 1368,
      "time_per_iteration": 2.583721160888672
    },
    {
      "auxiliary_loss_clip": 0.01210375,
      "auxiliary_loss_mlp": 0.01036431,
      "balance_loss_clip": 1.06160176,
      "balance_loss_mlp": 1.02639985,
      "epoch": 0.16461251728491552,
      "flos": 13545217802880.0,
      "grad_norm": 2.0230408962172817,
      "language_loss": 0.80816555,
      "learning_rate": 3.8130360037912368e-06,
      "loss": 0.83063364,
      "num_input_tokens_seen": 28863990,
      "step": 1369,
      "time_per_iteration": 2.4576709270477295
    },
    {
      "auxiliary_loss_clip": 0.01210056,
      "auxiliary_loss_mlp": 0.01041436,
      "balance_loss_clip": 1.06145298,
      "balance_loss_mlp": 1.03005731,
      "epoch": 0.16473276017555463,
      "flos": 23003298662400.0,
      "grad_norm": 1.9511306278999785,
      "language_loss": 0.81501842,
      "learning_rate": 3.812707010032268e-06,
      "loss": 0.83753335,
      "num_input_tokens_seen": 28883045,
      "step": 1370,
      "time_per_iteration": 2.518853187561035
    },
    {
      "auxiliary_loss_clip": 0.01219745,
      "auxiliary_loss_mlp": 0.01040782,
      "balance_loss_clip": 1.06846142,
      "balance_loss_mlp": 1.03033924,
      "epoch": 0.16485300306619372,
      "flos": 24790680357120.0,
      "grad_norm": 1.7125036783675724,
      "language_loss": 0.79440933,
      "learning_rate": 3.8123777412870863e-06,
      "loss": 0.81701458,
      "num_input_tokens_seen": 28902545,
      "step": 1371,
      "time_per_iteration": 2.55684232711792
    },
    {
      "auxiliary_loss_clip": 0.01203217,
      "auxiliary_loss_mlp": 0.01041595,
      "balance_loss_clip": 1.06136775,
      "balance_loss_mlp": 1.03152823,
      "epoch": 0.1649732459568328,
      "flos": 21106497162240.0,
      "grad_norm": 2.1325425615849722,
      "language_loss": 0.78685129,
      "learning_rate": 3.812048197605643e-06,
      "loss": 0.80929935,
      "num_input_tokens_seen": 28921440,
      "step": 1372,
      "time_per_iteration": 2.525838613510132
    },
    {
      "auxiliary_loss_clip": 0.01212257,
      "auxiliary_loss_mlp": 0.01029226,
      "balance_loss_clip": 1.06244302,
      "balance_loss_mlp": 1.01880085,
      "epoch": 0.16509348884747188,
      "flos": 20266726118400.0,
      "grad_norm": 1.8115963360665694,
      "language_loss": 0.81618851,
      "learning_rate": 3.8117183790379277e-06,
      "loss": 0.83860332,
      "num_input_tokens_seen": 28939890,
      "step": 1373,
      "time_per_iteration": 2.5070483684539795
    },
    {
      "auxiliary_loss_clip": 0.0122795,
      "auxiliary_loss_mlp": 0.01034944,
      "balance_loss_clip": 1.06444144,
      "balance_loss_mlp": 1.02419734,
      "epoch": 0.165213731738111,
      "flos": 11035493602560.0,
      "grad_norm": 2.714810433806094,
      "language_loss": 0.93467635,
      "learning_rate": 3.811388285633976e-06,
      "loss": 0.95730531,
      "num_input_tokens_seen": 28955875,
      "step": 1374,
      "time_per_iteration": 2.4512100219726562
    },
    {
      "auxiliary_loss_clip": 0.01171078,
      "auxiliary_loss_mlp": 0.01046444,
      "balance_loss_clip": 1.0563519,
      "balance_loss_mlp": 1.03607297,
      "epoch": 0.16533397462875007,
      "flos": 29972051268480.0,
      "grad_norm": 1.8133684746255796,
      "language_loss": 0.61846185,
      "learning_rate": 3.811057917443861e-06,
      "loss": 0.6406371,
      "num_input_tokens_seen": 28975140,
      "step": 1375,
      "time_per_iteration": 2.6408092975616455
    },
    {
      "auxiliary_loss_clip": 0.01125922,
      "auxiliary_loss_mlp": 0.0100883,
      "balance_loss_clip": 1.04679835,
      "balance_loss_mlp": 1.00600457,
      "epoch": 0.16545421751938916,
      "flos": 65556763027200.0,
      "grad_norm": 0.8520100086531267,
      "language_loss": 0.68370509,
      "learning_rate": 3.8107272745177e-06,
      "loss": 0.70505261,
      "num_input_tokens_seen": 29047470,
      "step": 1376,
      "time_per_iteration": 3.237907648086548
    },
    {
      "auxiliary_loss_clip": 0.01183851,
      "auxiliary_loss_mlp": 0.01037385,
      "balance_loss_clip": 1.0614059,
      "balance_loss_mlp": 1.02759159,
      "epoch": 0.16557446041002827,
      "flos": 22492361652480.0,
      "grad_norm": 1.81254315151487,
      "language_loss": 0.78804159,
      "learning_rate": 3.8103963569056513e-06,
      "loss": 0.81025398,
      "num_input_tokens_seen": 29066605,
      "step": 1377,
      "time_per_iteration": 2.5636985301971436
    },
    {
      "auxiliary_loss_clip": 0.01190261,
      "auxiliary_loss_mlp": 0.01039603,
      "balance_loss_clip": 1.05754375,
      "balance_loss_mlp": 1.02918434,
      "epoch": 0.16569470330066735,
      "flos": 24602723464320.0,
      "grad_norm": 1.8081945780223116,
      "language_loss": 0.88220656,
      "learning_rate": 3.8100651646579146e-06,
      "loss": 0.90450519,
      "num_input_tokens_seen": 29085815,
      "step": 1378,
      "time_per_iteration": 2.5551843643188477
    },
    {
      "auxiliary_loss_clip": 0.0119106,
      "auxiliary_loss_mlp": 0.01040802,
      "balance_loss_clip": 1.0567534,
      "balance_loss_mlp": 1.03037679,
      "epoch": 0.16581494619130643,
      "flos": 15006207588480.0,
      "grad_norm": 2.2579932200894492,
      "language_loss": 0.92520642,
      "learning_rate": 3.8097336978247317e-06,
      "loss": 0.94752502,
      "num_input_tokens_seen": 29102520,
      "step": 1379,
      "time_per_iteration": 2.5022709369659424
    },
    {
      "auxiliary_loss_clip": 0.01180913,
      "auxiliary_loss_mlp": 0.01033551,
      "balance_loss_clip": 1.05739605,
      "balance_loss_mlp": 1.02250028,
      "epoch": 0.16593518908194552,
      "flos": 17420338719360.0,
      "grad_norm": 8.971629474181267,
      "language_loss": 0.89236295,
      "learning_rate": 3.8094019564563854e-06,
      "loss": 0.91450757,
      "num_input_tokens_seen": 29119450,
      "step": 1380,
      "time_per_iteration": 2.4834752082824707
    },
    {
      "auxiliary_loss_clip": 0.01224668,
      "auxiliary_loss_mlp": 0.00765699,
      "balance_loss_clip": 1.06271076,
      "balance_loss_mlp": 1.00107872,
      "epoch": 0.16605543197258463,
      "flos": 20412631163520.0,
      "grad_norm": 2.052760973060971,
      "language_loss": 0.75581467,
      "learning_rate": 3.809069940603201e-06,
      "loss": 0.77571833,
      "num_input_tokens_seen": 29137405,
      "step": 1381,
      "time_per_iteration": 3.274097442626953
    },
    {
      "auxiliary_loss_clip": 0.01183993,
      "auxiliary_loss_mlp": 0.01036738,
      "balance_loss_clip": 1.05729604,
      "balance_loss_mlp": 1.02640867,
      "epoch": 0.1661756748632237,
      "flos": 14209745368320.0,
      "grad_norm": 2.0132959896732205,
      "language_loss": 0.78133857,
      "learning_rate": 3.8087376503155452e-06,
      "loss": 0.80354583,
      "num_input_tokens_seen": 29154890,
      "step": 1382,
      "time_per_iteration": 2.5003859996795654
    },
    {
      "auxiliary_loss_clip": 0.01113596,
      "auxiliary_loss_mlp": 0.01005909,
      "balance_loss_clip": 1.0397017,
      "balance_loss_mlp": 1.00326288,
      "epoch": 0.1662959177538628,
      "flos": 66080877350400.0,
      "grad_norm": 0.8994570687113425,
      "language_loss": 0.56258345,
      "learning_rate": 3.808405085643826e-06,
      "loss": 0.5837785,
      "num_input_tokens_seen": 29219770,
      "step": 1383,
      "time_per_iteration": 3.1710710525512695
    },
    {
      "auxiliary_loss_clip": 0.01229323,
      "auxiliary_loss_mlp": 0.00764991,
      "balance_loss_clip": 1.06563878,
      "balance_loss_mlp": 1.00100219,
      "epoch": 0.1664161606445019,
      "flos": 20740567357440.0,
      "grad_norm": 1.982860553189857,
      "language_loss": 0.89035559,
      "learning_rate": 3.8080722466384925e-06,
      "loss": 0.91029871,
      "num_input_tokens_seen": 29237620,
      "step": 1384,
      "time_per_iteration": 2.564457654953003
    },
    {
      "auxiliary_loss_clip": 0.0122652,
      "auxiliary_loss_mlp": 0.0103542,
      "balance_loss_clip": 1.06080079,
      "balance_loss_mlp": 1.02411914,
      "epoch": 0.166536403535141,
      "flos": 25260930236160.0,
      "grad_norm": 2.4412553749917314,
      "language_loss": 0.70852494,
      "learning_rate": 3.8077391333500376e-06,
      "loss": 0.73114431,
      "num_input_tokens_seen": 29256760,
      "step": 1385,
      "time_per_iteration": 2.5124237537384033
    },
    {
      "auxiliary_loss_clip": 0.01197055,
      "auxiliary_loss_mlp": 0.01034695,
      "balance_loss_clip": 1.06316423,
      "balance_loss_mlp": 1.02525413,
      "epoch": 0.16665664642578007,
      "flos": 25447450584960.0,
      "grad_norm": 1.6219565142174277,
      "language_loss": 0.76914406,
      "learning_rate": 3.8074057458289934e-06,
      "loss": 0.79146153,
      "num_input_tokens_seen": 29277450,
      "step": 1386,
      "time_per_iteration": 2.571053981781006
    },
    {
      "auxiliary_loss_clip": 0.01197183,
      "auxiliary_loss_mlp": 0.01032856,
      "balance_loss_clip": 1.05859804,
      "balance_loss_mlp": 1.022717,
      "epoch": 0.16677688931641918,
      "flos": 22200767043840.0,
      "grad_norm": 2.1283436875111574,
      "language_loss": 0.82579291,
      "learning_rate": 3.807072084125934e-06,
      "loss": 0.84809327,
      "num_input_tokens_seen": 29299300,
      "step": 1387,
      "time_per_iteration": 4.2407801151275635
    },
    {
      "auxiliary_loss_clip": 0.0119133,
      "auxiliary_loss_mlp": 0.01036626,
      "balance_loss_clip": 1.06075597,
      "balance_loss_mlp": 1.02601099,
      "epoch": 0.16689713220705826,
      "flos": 16945958776320.0,
      "grad_norm": 2.206606544761664,
      "language_loss": 0.80379069,
      "learning_rate": 3.806738148291477e-06,
      "loss": 0.82607031,
      "num_input_tokens_seen": 29316125,
      "step": 1388,
      "time_per_iteration": 3.274813413619995
    },
    {
      "auxiliary_loss_clip": 0.01153644,
      "auxiliary_loss_mlp": 0.01036511,
      "balance_loss_clip": 1.05329514,
      "balance_loss_mlp": 1.02517402,
      "epoch": 0.16701737509769735,
      "flos": 36244423923840.0,
      "grad_norm": 2.0105182950254163,
      "language_loss": 0.7117635,
      "learning_rate": 3.8064039383762793e-06,
      "loss": 0.73366505,
      "num_input_tokens_seen": 29338490,
      "step": 1389,
      "time_per_iteration": 2.7404568195343018
    },
    {
      "auxiliary_loss_clip": 0.01211217,
      "auxiliary_loss_mlp": 0.01035724,
      "balance_loss_clip": 1.0650568,
      "balance_loss_mlp": 1.02581751,
      "epoch": 0.16713761798833643,
      "flos": 23258659426560.0,
      "grad_norm": 2.0744782597917077,
      "language_loss": 0.77264869,
      "learning_rate": 3.8060694544310396e-06,
      "loss": 0.79511809,
      "num_input_tokens_seen": 29357000,
      "step": 1390,
      "time_per_iteration": 2.5060887336730957
    },
    {
      "auxiliary_loss_clip": 0.01227837,
      "auxiliary_loss_mlp": 0.01046129,
      "balance_loss_clip": 1.0635972,
      "balance_loss_mlp": 1.03505468,
      "epoch": 0.16725786087897554,
      "flos": 25302515207040.0,
      "grad_norm": 2.8604823169462423,
      "language_loss": 0.78473896,
      "learning_rate": 3.8057346965065006e-06,
      "loss": 0.80747861,
      "num_input_tokens_seen": 29378230,
      "step": 1391,
      "time_per_iteration": 2.5625696182250977
    },
    {
      "auxiliary_loss_clip": 0.01194707,
      "auxiliary_loss_mlp": 0.01039949,
      "balance_loss_clip": 1.06337333,
      "balance_loss_mlp": 1.02995348,
      "epoch": 0.16737810376961462,
      "flos": 31831541516160.0,
      "grad_norm": 1.7428483930149752,
      "language_loss": 0.84235871,
      "learning_rate": 3.805399664653443e-06,
      "loss": 0.86470526,
      "num_input_tokens_seen": 29400370,
      "step": 1392,
      "time_per_iteration": 2.5940401554107666
    },
    {
      "auxiliary_loss_clip": 0.01228854,
      "auxiliary_loss_mlp": 0.01032973,
      "balance_loss_clip": 1.06441784,
      "balance_loss_mlp": 1.02242899,
      "epoch": 0.1674983466602537,
      "flos": 27961843553280.0,
      "grad_norm": 2.5359832136223766,
      "language_loss": 0.73990822,
      "learning_rate": 3.805064358922692e-06,
      "loss": 0.76252645,
      "num_input_tokens_seen": 29418660,
      "step": 1393,
      "time_per_iteration": 2.4924235343933105
    },
    {
      "auxiliary_loss_clip": 0.01215762,
      "auxiliary_loss_mlp": 0.01034291,
      "balance_loss_clip": 1.0626502,
      "balance_loss_mlp": 1.02334189,
      "epoch": 0.16761858955089282,
      "flos": 21762656858880.0,
      "grad_norm": 1.6605558146107995,
      "language_loss": 0.80892026,
      "learning_rate": 3.8047287793651136e-06,
      "loss": 0.83142078,
      "num_input_tokens_seen": 29440105,
      "step": 1394,
      "time_per_iteration": 2.5098814964294434
    },
    {
      "auxiliary_loss_clip": 0.01182484,
      "auxiliary_loss_mlp": 0.01040966,
      "balance_loss_clip": 1.05948305,
      "balance_loss_mlp": 1.03104222,
      "epoch": 0.1677388324415319,
      "flos": 23805507058560.0,
      "grad_norm": 1.9090367690346433,
      "language_loss": 0.89084631,
      "learning_rate": 3.8043929260316137e-06,
      "loss": 0.91308081,
      "num_input_tokens_seen": 29458260,
      "step": 1395,
      "time_per_iteration": 2.5944623947143555
    },
    {
      "auxiliary_loss_clip": 0.01200802,
      "auxiliary_loss_mlp": 0.01038303,
      "balance_loss_clip": 1.06691086,
      "balance_loss_mlp": 1.02762794,
      "epoch": 0.16785907533217098,
      "flos": 20558859431040.0,
      "grad_norm": 2.6490667294790984,
      "language_loss": 0.83648103,
      "learning_rate": 3.8040567989731417e-06,
      "loss": 0.85887212,
      "num_input_tokens_seen": 29476205,
      "step": 1396,
      "time_per_iteration": 2.5344557762145996
    },
    {
      "auxiliary_loss_clip": 0.01206737,
      "auxiliary_loss_mlp": 0.01033,
      "balance_loss_clip": 1.06304944,
      "balance_loss_mlp": 1.02339196,
      "epoch": 0.16797931822281006,
      "flos": 15669657745920.0,
      "grad_norm": 2.0475327141272888,
      "language_loss": 0.79928887,
      "learning_rate": 3.8037203982406876e-06,
      "loss": 0.82168621,
      "num_input_tokens_seen": 29494370,
      "step": 1397,
      "time_per_iteration": 2.488276481628418
    },
    {
      "auxiliary_loss_clip": 0.01226939,
      "auxiliary_loss_mlp": 0.01035158,
      "balance_loss_clip": 1.06578791,
      "balance_loss_mlp": 1.02456009,
      "epoch": 0.16809956111344918,
      "flos": 16541101607040.0,
      "grad_norm": 1.8410107359287795,
      "language_loss": 0.73271787,
      "learning_rate": 3.8033837238852835e-06,
      "loss": 0.75533879,
      "num_input_tokens_seen": 29511070,
      "step": 1398,
      "time_per_iteration": 2.436167001724243
    },
    {
      "auxiliary_loss_clip": 0.01186933,
      "auxiliary_loss_mlp": 0.01036228,
      "balance_loss_clip": 1.05714464,
      "balance_loss_mlp": 1.02669716,
      "epoch": 0.16821980400408826,
      "flos": 23258084808960.0,
      "grad_norm": 1.7759487014664332,
      "language_loss": 0.69822156,
      "learning_rate": 3.8030467759580017e-06,
      "loss": 0.72045314,
      "num_input_tokens_seen": 29531990,
      "step": 1399,
      "time_per_iteration": 2.5321736335754395
    },
    {
      "auxiliary_loss_clip": 0.01215549,
      "auxiliary_loss_mlp": 0.01040832,
      "balance_loss_clip": 1.06260562,
      "balance_loss_mlp": 1.03009772,
      "epoch": 0.16834004689472734,
      "flos": 20774754126720.0,
      "grad_norm": 1.8760981690256469,
      "language_loss": 0.87315619,
      "learning_rate": 3.802709554509958e-06,
      "loss": 0.89572001,
      "num_input_tokens_seen": 29549790,
      "step": 1400,
      "time_per_iteration": 2.4656083583831787
    },
    {
      "auxiliary_loss_clip": 0.01194745,
      "auxiliary_loss_mlp": 0.01030626,
      "balance_loss_clip": 1.0597322,
      "balance_loss_mlp": 1.02144051,
      "epoch": 0.16846028978536645,
      "flos": 26687302289280.0,
      "grad_norm": 2.18740556275661,
      "language_loss": 0.79255962,
      "learning_rate": 3.8023720595923083e-06,
      "loss": 0.81481344,
      "num_input_tokens_seen": 29569045,
      "step": 1401,
      "time_per_iteration": 2.5857348442077637
    },
    {
      "auxiliary_loss_clip": 0.01161763,
      "auxiliary_loss_mlp": 0.01036317,
      "balance_loss_clip": 1.05569124,
      "balance_loss_mlp": 1.0259341,
      "epoch": 0.16858053267600553,
      "flos": 18843298980480.0,
      "grad_norm": 1.984343811345821,
      "language_loss": 0.87713242,
      "learning_rate": 3.80203429125625e-06,
      "loss": 0.89911318,
      "num_input_tokens_seen": 29587220,
      "step": 1402,
      "time_per_iteration": 2.5734236240386963
    },
    {
      "auxiliary_loss_clip": 0.01141583,
      "auxiliary_loss_mlp": 0.01034267,
      "balance_loss_clip": 1.05420411,
      "balance_loss_mlp": 1.02434897,
      "epoch": 0.16870077556664462,
      "flos": 27744548227200.0,
      "grad_norm": 1.7811897472461624,
      "language_loss": 0.70326531,
      "learning_rate": 3.8016962495530225e-06,
      "loss": 0.72502381,
      "num_input_tokens_seen": 29606410,
      "step": 1403,
      "time_per_iteration": 2.6648831367492676
    },
    {
      "auxiliary_loss_clip": 0.01229687,
      "auxiliary_loss_mlp": 0.01039471,
      "balance_loss_clip": 1.0662477,
      "balance_loss_mlp": 1.02994609,
      "epoch": 0.1688210184572837,
      "flos": 13730768484480.0,
      "grad_norm": 2.2473037856654776,
      "language_loss": 0.77412438,
      "learning_rate": 3.8013579345339063e-06,
      "loss": 0.79681587,
      "num_input_tokens_seen": 29621275,
      "step": 1404,
      "time_per_iteration": 2.428229570388794
    },
    {
      "auxiliary_loss_clip": 0.01185222,
      "auxiliary_loss_mlp": 0.01032288,
      "balance_loss_clip": 1.05883288,
      "balance_loss_mlp": 1.02195835,
      "epoch": 0.1689412613479228,
      "flos": 26468785900800.0,
      "grad_norm": 2.128282175240056,
      "language_loss": 0.69736111,
      "learning_rate": 3.801019346250224e-06,
      "loss": 0.7195363,
      "num_input_tokens_seen": 29641420,
      "step": 1405,
      "time_per_iteration": 2.6064810752868652
    },
    {
      "auxiliary_loss_clip": 0.01208163,
      "auxiliary_loss_mlp": 0.01032851,
      "balance_loss_clip": 1.0626018,
      "balance_loss_mlp": 1.02249122,
      "epoch": 0.1690615042385619,
      "flos": 21138852337920.0,
      "grad_norm": 4.432058064474053,
      "language_loss": 0.84003842,
      "learning_rate": 3.8006804847533395e-06,
      "loss": 0.86244857,
      "num_input_tokens_seen": 29660935,
      "step": 1406,
      "time_per_iteration": 2.5351383686065674
    },
    {
      "auxiliary_loss_clip": 0.01230538,
      "auxiliary_loss_mlp": 0.01039557,
      "balance_loss_clip": 1.06615186,
      "balance_loss_mlp": 1.03051496,
      "epoch": 0.16918174712920098,
      "flos": 20849340718080.0,
      "grad_norm": 2.1460214244826257,
      "language_loss": 0.8576861,
      "learning_rate": 3.8003413500946556e-06,
      "loss": 0.88038707,
      "num_input_tokens_seen": 29681045,
      "step": 1407,
      "time_per_iteration": 3.298356056213379
    },
    {
      "auxiliary_loss_clip": 0.01198509,
      "auxiliary_loss_mlp": 0.01041213,
      "balance_loss_clip": 1.06286693,
      "balance_loss_mlp": 1.03038335,
      "epoch": 0.1693019900198401,
      "flos": 16983270028800.0,
      "grad_norm": 2.373936961580943,
      "language_loss": 0.82466704,
      "learning_rate": 3.8000019423256216e-06,
      "loss": 0.84706426,
      "num_input_tokens_seen": 29698810,
      "step": 1408,
      "time_per_iteration": 2.5213544368743896
    },
    {
      "auxiliary_loss_clip": 0.01187343,
      "auxiliary_loss_mlp": 0.01045447,
      "balance_loss_clip": 1.06205058,
      "balance_loss_mlp": 1.03530288,
      "epoch": 0.16942223291047917,
      "flos": 26796901662720.0,
      "grad_norm": 1.5668950939483715,
      "language_loss": 0.87990034,
      "learning_rate": 3.7996622614977234e-06,
      "loss": 0.90222824,
      "num_input_tokens_seen": 29720000,
      "step": 1409,
      "time_per_iteration": 2.570082664489746
    },
    {
      "auxiliary_loss_clip": 0.01194845,
      "auxiliary_loss_mlp": 0.01035001,
      "balance_loss_clip": 1.06289458,
      "balance_loss_mlp": 1.02547002,
      "epoch": 0.16954247580111825,
      "flos": 18583700411520.0,
      "grad_norm": 1.7840986055052674,
      "language_loss": 0.79178345,
      "learning_rate": 3.799322307662492e-06,
      "loss": 0.81408191,
      "num_input_tokens_seen": 29737820,
      "step": 1410,
      "time_per_iteration": 2.5340983867645264
    },
    {
      "auxiliary_loss_clip": 0.01170895,
      "auxiliary_loss_mlp": 0.01031126,
      "balance_loss_clip": 1.05707288,
      "balance_loss_mlp": 1.02070141,
      "epoch": 0.16966271869175734,
      "flos": 13983651210240.0,
      "grad_norm": 2.1602495070158882,
      "language_loss": 0.83495837,
      "learning_rate": 3.798982080871496e-06,
      "loss": 0.8569786,
      "num_input_tokens_seen": 29752960,
      "step": 1411,
      "time_per_iteration": 2.6597843170166016
    },
    {
      "auxiliary_loss_clip": 0.01230922,
      "auxiliary_loss_mlp": 0.01039459,
      "balance_loss_clip": 1.06725669,
      "balance_loss_mlp": 1.0287838,
      "epoch": 0.16978296158239645,
      "flos": 37487328284160.0,
      "grad_norm": 2.186501420728542,
      "language_loss": 0.67638087,
      "learning_rate": 3.798641581176349e-06,
      "loss": 0.69908476,
      "num_input_tokens_seen": 29775240,
      "step": 1412,
      "time_per_iteration": 2.607224702835083
    },
    {
      "auxiliary_loss_clip": 0.0119717,
      "auxiliary_loss_mlp": 0.01043152,
      "balance_loss_clip": 1.06025338,
      "balance_loss_mlp": 1.03231621,
      "epoch": 0.16990320447303553,
      "flos": 28328958506880.0,
      "grad_norm": 1.8804893321233875,
      "language_loss": 0.74920118,
      "learning_rate": 3.7983008086287044e-06,
      "loss": 0.77160436,
      "num_input_tokens_seen": 29796560,
      "step": 1413,
      "time_per_iteration": 2.6074204444885254
    },
    {
      "auxiliary_loss_clip": 0.01192482,
      "auxiliary_loss_mlp": 0.01039117,
      "balance_loss_clip": 1.05804729,
      "balance_loss_mlp": 1.02831626,
      "epoch": 0.1700234473636746,
      "flos": 20188189031040.0,
      "grad_norm": 3.1100469396977504,
      "language_loss": 0.79611427,
      "learning_rate": 3.797959763280257e-06,
      "loss": 0.81843024,
      "num_input_tokens_seen": 29815245,
      "step": 1414,
      "time_per_iteration": 4.217747211456299
    },
    {
      "auxiliary_loss_clip": 0.01216257,
      "auxiliary_loss_mlp": 0.01046714,
      "balance_loss_clip": 1.06441629,
      "balance_loss_mlp": 1.03708172,
      "epoch": 0.17014369025431372,
      "flos": 24858658846080.0,
      "grad_norm": 1.8866636825495138,
      "language_loss": 0.79411358,
      "learning_rate": 3.797618445182743e-06,
      "loss": 0.81674325,
      "num_input_tokens_seen": 29836640,
      "step": 1415,
      "time_per_iteration": 3.3129775524139404
    },
    {
      "auxiliary_loss_clip": 0.01162699,
      "auxiliary_loss_mlp": 0.01035782,
      "balance_loss_clip": 1.05588186,
      "balance_loss_mlp": 1.02528572,
      "epoch": 0.1702639331449528,
      "flos": 16467233287680.0,
      "grad_norm": 1.943683991808472,
      "language_loss": 0.84756851,
      "learning_rate": 3.79727685438794e-06,
      "loss": 0.86955333,
      "num_input_tokens_seen": 29850830,
      "step": 1416,
      "time_per_iteration": 2.5465750694274902
    },
    {
      "auxiliary_loss_clip": 0.01134974,
      "auxiliary_loss_mlp": 0.01003247,
      "balance_loss_clip": 1.04604411,
      "balance_loss_mlp": 1.00050533,
      "epoch": 0.1703841760355919,
      "flos": 52508870979840.0,
      "grad_norm": 0.8388428360397058,
      "language_loss": 0.61706597,
      "learning_rate": 3.796934990947667e-06,
      "loss": 0.63844818,
      "num_input_tokens_seen": 29912515,
      "step": 1417,
      "time_per_iteration": 3.1363086700439453
    },
    {
      "auxiliary_loss_clip": 0.01133715,
      "auxiliary_loss_mlp": 0.0100463,
      "balance_loss_clip": 1.04555082,
      "balance_loss_mlp": 1.00185287,
      "epoch": 0.170504418926231,
      "flos": 49370637576960.0,
      "grad_norm": 0.878300632858872,
      "language_loss": 0.62479937,
      "learning_rate": 3.7965928549137854e-06,
      "loss": 0.64618289,
      "num_input_tokens_seen": 29969330,
      "step": 1418,
      "time_per_iteration": 2.9846107959747314
    },
    {
      "auxiliary_loss_clip": 0.01185632,
      "auxiliary_loss_mlp": 0.01040592,
      "balance_loss_clip": 1.0556097,
      "balance_loss_mlp": 1.02974439,
      "epoch": 0.17062466181687008,
      "flos": 25849219184640.0,
      "grad_norm": 2.0942533726877905,
      "language_loss": 0.77478254,
      "learning_rate": 3.7962504463381953e-06,
      "loss": 0.79704475,
      "num_input_tokens_seen": 29990820,
      "step": 1419,
      "time_per_iteration": 2.6042940616607666
    },
    {
      "auxiliary_loss_clip": 0.0119339,
      "auxiliary_loss_mlp": 0.00766508,
      "balance_loss_clip": 1.06383729,
      "balance_loss_mlp": 1.00090909,
      "epoch": 0.17074490470750917,
      "flos": 20960412549120.0,
      "grad_norm": 1.7503120276866715,
      "language_loss": 0.7898913,
      "learning_rate": 3.7959077652728412e-06,
      "loss": 0.80949026,
      "num_input_tokens_seen": 30009275,
      "step": 1420,
      "time_per_iteration": 2.5444934368133545
    },
    {
      "auxiliary_loss_clip": 0.01196036,
      "auxiliary_loss_mlp": 0.01039795,
      "balance_loss_clip": 1.05955565,
      "balance_loss_mlp": 1.02935195,
      "epoch": 0.17086514759814825,
      "flos": 20959766104320.0,
      "grad_norm": 2.6277076389224443,
      "language_loss": 0.77596462,
      "learning_rate": 3.795564811769707e-06,
      "loss": 0.79832292,
      "num_input_tokens_seen": 30027630,
      "step": 1421,
      "time_per_iteration": 2.5280802249908447
    },
    {
      "auxiliary_loss_clip": 0.0119714,
      "auxiliary_loss_mlp": 0.01037743,
      "balance_loss_clip": 1.06417918,
      "balance_loss_mlp": 1.02626896,
      "epoch": 0.17098539048878736,
      "flos": 28474073452800.0,
      "grad_norm": 1.9606844455074244,
      "language_loss": 0.78223407,
      "learning_rate": 3.795221585880818e-06,
      "loss": 0.80458289,
      "num_input_tokens_seen": 30048310,
      "step": 1422,
      "time_per_iteration": 2.607741117477417
    },
    {
      "auxiliary_loss_clip": 0.01182946,
      "auxiliary_loss_mlp": 0.01037483,
      "balance_loss_clip": 1.06351352,
      "balance_loss_mlp": 1.0280118,
      "epoch": 0.17110563337942644,
      "flos": 16290014561280.0,
      "grad_norm": 1.8635863668084915,
      "language_loss": 0.91161311,
      "learning_rate": 3.794878087658242e-06,
      "loss": 0.93381739,
      "num_input_tokens_seen": 30066080,
      "step": 1423,
      "time_per_iteration": 2.5263290405273438
    },
    {
      "auxiliary_loss_clip": 0.01212412,
      "auxiliary_loss_mlp": 0.01036035,
      "balance_loss_clip": 1.06093574,
      "balance_loss_mlp": 1.02628374,
      "epoch": 0.17122587627006552,
      "flos": 29674207693440.0,
      "grad_norm": 1.972740840965769,
      "language_loss": 0.78389752,
      "learning_rate": 3.7945343171540873e-06,
      "loss": 0.80638194,
      "num_input_tokens_seen": 30086955,
      "step": 1424,
      "time_per_iteration": 2.557713747024536
    },
    {
      "auxiliary_loss_clip": 0.01229543,
      "auxiliary_loss_mlp": 0.01038775,
      "balance_loss_clip": 1.06529009,
      "balance_loss_mlp": 1.02756333,
      "epoch": 0.17134611916070464,
      "flos": 25338389915520.0,
      "grad_norm": 1.7429233529747776,
      "language_loss": 0.7850129,
      "learning_rate": 3.7941902744205033e-06,
      "loss": 0.8076961,
      "num_input_tokens_seen": 30107990,
      "step": 1425,
      "time_per_iteration": 2.517632246017456
    },
    {
      "auxiliary_loss_clip": 0.01200463,
      "auxiliary_loss_mlp": 0.01035194,
      "balance_loss_clip": 1.06053579,
      "balance_loss_mlp": 1.02386379,
      "epoch": 0.17146636205134372,
      "flos": 13953845900160.0,
      "grad_norm": 1.8725603799353174,
      "language_loss": 0.83362359,
      "learning_rate": 3.7938459595096817e-06,
      "loss": 0.85598022,
      "num_input_tokens_seen": 30126535,
      "step": 1426,
      "time_per_iteration": 2.5453715324401855
    },
    {
      "auxiliary_loss_clip": 0.01218655,
      "auxiliary_loss_mlp": 0.01037703,
      "balance_loss_clip": 1.06246901,
      "balance_loss_mlp": 1.02697444,
      "epoch": 0.1715866049419828,
      "flos": 23915214172800.0,
      "grad_norm": 1.749569964461334,
      "language_loss": 0.86320484,
      "learning_rate": 3.7935013724738545e-06,
      "loss": 0.88576841,
      "num_input_tokens_seen": 30147035,
      "step": 1427,
      "time_per_iteration": 2.524198055267334
    },
    {
      "auxiliary_loss_clip": 0.01206981,
      "auxiliary_loss_mlp": 0.01040626,
      "balance_loss_clip": 1.06174147,
      "balance_loss_mlp": 1.03084469,
      "epoch": 0.17170684783262188,
      "flos": 22709369669760.0,
      "grad_norm": 1.7177835555974934,
      "language_loss": 0.77919137,
      "learning_rate": 3.7931565133652945e-06,
      "loss": 0.80166739,
      "num_input_tokens_seen": 30167110,
      "step": 1428,
      "time_per_iteration": 2.503652572631836
    },
    {
      "auxiliary_loss_clip": 0.01226606,
      "auxiliary_loss_mlp": 0.01036469,
      "balance_loss_clip": 1.06393242,
      "balance_loss_mlp": 1.02600861,
      "epoch": 0.171827090723261,
      "flos": 26613290315520.0,
      "grad_norm": 2.3727193250489607,
      "language_loss": 0.68187082,
      "learning_rate": 3.792811382236317e-06,
      "loss": 0.70450157,
      "num_input_tokens_seen": 30185620,
      "step": 1429,
      "time_per_iteration": 2.523554801940918
    },
    {
      "auxiliary_loss_clip": 0.01216202,
      "auxiliary_loss_mlp": 0.0103349,
      "balance_loss_clip": 1.06277788,
      "balance_loss_mlp": 1.02280283,
      "epoch": 0.17194733361390008,
      "flos": 28148507556480.0,
      "grad_norm": 2.075552834756765,
      "language_loss": 0.78314114,
      "learning_rate": 3.792465979139279e-06,
      "loss": 0.80563807,
      "num_input_tokens_seen": 30208225,
      "step": 1430,
      "time_per_iteration": 2.573277473449707
    },
    {
      "auxiliary_loss_clip": 0.01100972,
      "auxiliary_loss_mlp": 0.0102745,
      "balance_loss_clip": 1.03785551,
      "balance_loss_mlp": 1.02462447,
      "epoch": 0.17206757650453916,
      "flos": 65530689753600.0,
      "grad_norm": 0.9244882717881197,
      "language_loss": 0.65645969,
      "learning_rate": 3.792120304126576e-06,
      "loss": 0.67774385,
      "num_input_tokens_seen": 30271600,
      "step": 1431,
      "time_per_iteration": 3.1704819202423096
    },
    {
      "auxiliary_loss_clip": 0.01138726,
      "auxiliary_loss_mlp": 0.01027081,
      "balance_loss_clip": 1.05465043,
      "balance_loss_mlp": 1.01774073,
      "epoch": 0.17218781939517827,
      "flos": 22273486128000.0,
      "grad_norm": 1.878238450222419,
      "language_loss": 0.83875877,
      "learning_rate": 3.791774357250649e-06,
      "loss": 0.86041683,
      "num_input_tokens_seen": 30290430,
      "step": 1432,
      "time_per_iteration": 2.774661064147949
    },
    {
      "auxiliary_loss_clip": 0.01192632,
      "auxiliary_loss_mlp": 0.01048282,
      "balance_loss_clip": 1.05880046,
      "balance_loss_mlp": 1.03708231,
      "epoch": 0.17230806228581735,
      "flos": 14137313592960.0,
      "grad_norm": 2.074017535049121,
      "language_loss": 0.79137564,
      "learning_rate": 3.7914281385639757e-06,
      "loss": 0.81378484,
      "num_input_tokens_seen": 30308305,
      "step": 1433,
      "time_per_iteration": 2.5613033771514893
    },
    {
      "auxiliary_loss_clip": 0.01209972,
      "auxiliary_loss_mlp": 0.01034717,
      "balance_loss_clip": 1.05939877,
      "balance_loss_mlp": 1.02413106,
      "epoch": 0.17242830517645644,
      "flos": 20704836303360.0,
      "grad_norm": 2.020455169326723,
      "language_loss": 0.79531109,
      "learning_rate": 3.7910816481190784e-06,
      "loss": 0.81775796,
      "num_input_tokens_seen": 30328120,
      "step": 1434,
      "time_per_iteration": 3.3162264823913574
    },
    {
      "auxiliary_loss_clip": 0.01182037,
      "auxiliary_loss_mlp": 0.01037166,
      "balance_loss_clip": 1.05602324,
      "balance_loss_mlp": 1.02668166,
      "epoch": 0.17254854806709552,
      "flos": 30774582887040.0,
      "grad_norm": 2.0229952467878256,
      "language_loss": 0.74950767,
      "learning_rate": 3.7907348859685193e-06,
      "loss": 0.77169973,
      "num_input_tokens_seen": 30349825,
      "step": 1435,
      "time_per_iteration": 2.602051258087158
    },
    {
      "auxiliary_loss_clip": 0.01202795,
      "auxiliary_loss_mlp": 0.01034527,
      "balance_loss_clip": 1.06140232,
      "balance_loss_mlp": 1.02376199,
      "epoch": 0.17266879095773463,
      "flos": 26614726859520.0,
      "grad_norm": 1.92522314709842,
      "language_loss": 0.80151463,
      "learning_rate": 3.790387852164902e-06,
      "loss": 0.82388783,
      "num_input_tokens_seen": 30370555,
      "step": 1436,
      "time_per_iteration": 2.533043384552002
    },
    {
      "auxiliary_loss_clip": 0.01210455,
      "auxiliary_loss_mlp": 0.01037869,
      "balance_loss_clip": 1.06172287,
      "balance_loss_mlp": 1.02795076,
      "epoch": 0.1727890338483737,
      "flos": 20266295155200.0,
      "grad_norm": 4.618461891576178,
      "language_loss": 0.77113122,
      "learning_rate": 3.7900405467608707e-06,
      "loss": 0.79361439,
      "num_input_tokens_seen": 30390100,
      "step": 1437,
      "time_per_iteration": 2.4939024448394775
    },
    {
      "auxiliary_loss_clip": 0.01149515,
      "auxiliary_loss_mlp": 0.01033304,
      "balance_loss_clip": 1.05029559,
      "balance_loss_mlp": 1.02269423,
      "epoch": 0.1729092767390128,
      "flos": 18179812909440.0,
      "grad_norm": 3.2653280561745,
      "language_loss": 0.78959757,
      "learning_rate": 3.7896929698091114e-06,
      "loss": 0.81142581,
      "num_input_tokens_seen": 30402915,
      "step": 1438,
      "time_per_iteration": 2.5417556762695312
    },
    {
      "auxiliary_loss_clip": 0.01228201,
      "auxiliary_loss_mlp": 0.01042917,
      "balance_loss_clip": 1.06677377,
      "balance_loss_mlp": 1.03250384,
      "epoch": 0.1730295196296519,
      "flos": 26759518583040.0,
      "grad_norm": 3.5712285430426873,
      "language_loss": 0.68447709,
      "learning_rate": 3.7893451213623518e-06,
      "loss": 0.70718831,
      "num_input_tokens_seen": 30420145,
      "step": 1439,
      "time_per_iteration": 2.4793314933776855
    },
    {
      "auxiliary_loss_clip": 0.01210656,
      "auxiliary_loss_mlp": 0.00765794,
      "balance_loss_clip": 1.06375563,
      "balance_loss_mlp": 1.00084448,
      "epoch": 0.173149762520291,
      "flos": 23842531002240.0,
      "grad_norm": 2.2291654417466815,
      "language_loss": 0.82187635,
      "learning_rate": 3.7889970014733606e-06,
      "loss": 0.84164083,
      "num_input_tokens_seen": 30439250,
      "step": 1440,
      "time_per_iteration": 3.323448896408081
    },
    {
      "auxiliary_loss_clip": 0.01146114,
      "auxiliary_loss_mlp": 0.01036768,
      "balance_loss_clip": 1.05067933,
      "balance_loss_mlp": 1.02589035,
      "epoch": 0.17327000541093007,
      "flos": 23368186972800.0,
      "grad_norm": 1.7709030682981897,
      "language_loss": 0.78132027,
      "learning_rate": 3.7886486101949463e-06,
      "loss": 0.8031491,
      "num_input_tokens_seen": 30460430,
      "step": 1441,
      "time_per_iteration": 4.266057252883911
    },
    {
      "auxiliary_loss_clip": 0.01154186,
      "auxiliary_loss_mlp": 0.0104612,
      "balance_loss_clip": 1.05389822,
      "balance_loss_mlp": 1.0355165,
      "epoch": 0.17339024830156918,
      "flos": 18221290139520.0,
      "grad_norm": 1.9061362269322848,
      "language_loss": 0.88305372,
      "learning_rate": 3.7882999475799594e-06,
      "loss": 0.90505677,
      "num_input_tokens_seen": 30478465,
      "step": 1442,
      "time_per_iteration": 2.5881457328796387
    },
    {
      "auxiliary_loss_clip": 0.01148973,
      "auxiliary_loss_mlp": 0.01039488,
      "balance_loss_clip": 1.05625772,
      "balance_loss_mlp": 1.02933741,
      "epoch": 0.17351049119220827,
      "flos": 23332024955520.0,
      "grad_norm": 1.8436320015938201,
      "language_loss": 0.81345373,
      "learning_rate": 3.787951013681293e-06,
      "loss": 0.83533835,
      "num_input_tokens_seen": 30496510,
      "step": 1443,
      "time_per_iteration": 2.612109422683716
    },
    {
      "auxiliary_loss_clip": 0.0120618,
      "auxiliary_loss_mlp": 0.01043906,
      "balance_loss_clip": 1.05773818,
      "balance_loss_mlp": 1.03287363,
      "epoch": 0.17363073408284735,
      "flos": 23803495896960.0,
      "grad_norm": 1.9162495103262853,
      "language_loss": 0.77306521,
      "learning_rate": 3.787601808551879e-06,
      "loss": 0.79556608,
      "num_input_tokens_seen": 30516325,
      "step": 1444,
      "time_per_iteration": 2.516946792602539
    },
    {
      "auxiliary_loss_clip": 0.0118117,
      "auxiliary_loss_mlp": 0.01045223,
      "balance_loss_clip": 1.05812764,
      "balance_loss_mlp": 1.03485739,
      "epoch": 0.17375097697348643,
      "flos": 18515290959360.0,
      "grad_norm": 2.439937634319878,
      "language_loss": 0.83674777,
      "learning_rate": 3.7872523322446926e-06,
      "loss": 0.85901171,
      "num_input_tokens_seen": 30535210,
      "step": 1445,
      "time_per_iteration": 2.5380184650421143
    },
    {
      "auxiliary_loss_clip": 0.01169404,
      "auxiliary_loss_mlp": 0.01030467,
      "balance_loss_clip": 1.05203438,
      "balance_loss_mlp": 1.02079868,
      "epoch": 0.17387121986412554,
      "flos": 38877897456000.0,
      "grad_norm": 1.7687062258596444,
      "language_loss": 0.60204852,
      "learning_rate": 3.7869025848127478e-06,
      "loss": 0.62404728,
      "num_input_tokens_seen": 30559405,
      "step": 1446,
      "time_per_iteration": 2.8109099864959717
    },
    {
      "auxiliary_loss_clip": 0.01208658,
      "auxiliary_loss_mlp": 0.01038238,
      "balance_loss_clip": 1.05843425,
      "balance_loss_mlp": 1.02814138,
      "epoch": 0.17399146275476463,
      "flos": 20375714960640.0,
      "grad_norm": 4.609653296830032,
      "language_loss": 0.80659467,
      "learning_rate": 3.786552566309102e-06,
      "loss": 0.82906365,
      "num_input_tokens_seen": 30577615,
      "step": 1447,
      "time_per_iteration": 2.563925266265869
    },
    {
      "auxiliary_loss_clip": 0.01191322,
      "auxiliary_loss_mlp": 0.00765469,
      "balance_loss_clip": 1.06150126,
      "balance_loss_mlp": 1.00075054,
      "epoch": 0.1741117056454037,
      "flos": 19164339763200.0,
      "grad_norm": 2.155001230410101,
      "language_loss": 0.86027318,
      "learning_rate": 3.7862022767868517e-06,
      "loss": 0.87984109,
      "num_input_tokens_seen": 30595205,
      "step": 1448,
      "time_per_iteration": 2.5667619705200195
    },
    {
      "auxiliary_loss_clip": 0.01177304,
      "auxiliary_loss_mlp": 0.01042081,
      "balance_loss_clip": 1.06241715,
      "balance_loss_mlp": 1.0320673,
      "epoch": 0.17423194853604282,
      "flos": 25374300537600.0,
      "grad_norm": 2.703550903405167,
      "language_loss": 0.84045398,
      "learning_rate": 3.7858517162991367e-06,
      "loss": 0.86264777,
      "num_input_tokens_seen": 30615280,
      "step": 1449,
      "time_per_iteration": 2.5943922996520996
    },
    {
      "auxiliary_loss_clip": 0.01180372,
      "auxiliary_loss_mlp": 0.01037063,
      "balance_loss_clip": 1.05726457,
      "balance_loss_mlp": 1.02626264,
      "epoch": 0.1743521914266819,
      "flos": 25191874339200.0,
      "grad_norm": 2.35581001849865,
      "language_loss": 0.60471541,
      "learning_rate": 3.7855008848991363e-06,
      "loss": 0.62688971,
      "num_input_tokens_seen": 30633485,
      "step": 1450,
      "time_per_iteration": 2.662813901901245
    },
    {
      "auxiliary_loss_clip": 0.01190967,
      "auxiliary_loss_mlp": 0.0103882,
      "balance_loss_clip": 1.06117392,
      "balance_loss_mlp": 1.02910447,
      "epoch": 0.17447243431732098,
      "flos": 25666577504640.0,
      "grad_norm": 2.1520722266441887,
      "language_loss": 0.77669948,
      "learning_rate": 3.7851497826400714e-06,
      "loss": 0.7989974,
      "num_input_tokens_seen": 30653625,
      "step": 1451,
      "time_per_iteration": 2.6153042316436768
    },
    {
      "auxiliary_loss_clip": 0.01226882,
      "auxiliary_loss_mlp": 0.01039907,
      "balance_loss_clip": 1.06416225,
      "balance_loss_mlp": 1.02949405,
      "epoch": 0.17459267720796007,
      "flos": 36281950657920.0,
      "grad_norm": 2.3020210001543644,
      "language_loss": 0.76027685,
      "learning_rate": 3.7847984095752034e-06,
      "loss": 0.78294468,
      "num_input_tokens_seen": 30677080,
      "step": 1452,
      "time_per_iteration": 2.599299907684326
    },
    {
      "auxiliary_loss_clip": 0.01225495,
      "auxiliary_loss_mlp": 0.01031859,
      "balance_loss_clip": 1.06297827,
      "balance_loss_mlp": 1.02205372,
      "epoch": 0.17471292009859918,
      "flos": 20011113959040.0,
      "grad_norm": 1.9298031505748081,
      "language_loss": 0.80195272,
      "learning_rate": 3.784446765757836e-06,
      "loss": 0.82452631,
      "num_input_tokens_seen": 30695725,
      "step": 1453,
      "time_per_iteration": 2.4770569801330566
    },
    {
      "auxiliary_loss_clip": 0.01160946,
      "auxiliary_loss_mlp": 0.01034635,
      "balance_loss_clip": 1.05506372,
      "balance_loss_mlp": 1.02446628,
      "epoch": 0.17483316298923826,
      "flos": 27819242559360.0,
      "grad_norm": 2.3354274889735556,
      "language_loss": 0.776748,
      "learning_rate": 3.7840948512413133e-06,
      "loss": 0.79870385,
      "num_input_tokens_seen": 30713310,
      "step": 1454,
      "time_per_iteration": 2.60199236869812
    },
    {
      "auxiliary_loss_clip": 0.01177273,
      "auxiliary_loss_mlp": 0.01037327,
      "balance_loss_clip": 1.06002808,
      "balance_loss_mlp": 1.02616858,
      "epoch": 0.17495340587987734,
      "flos": 44017934791680.0,
      "grad_norm": 1.89224206983156,
      "language_loss": 0.78945887,
      "learning_rate": 3.7837426660790196e-06,
      "loss": 0.81160486,
      "num_input_tokens_seen": 30734725,
      "step": 1455,
      "time_per_iteration": 2.738637924194336
    },
    {
      "auxiliary_loss_clip": 0.01222324,
      "auxiliary_loss_mlp": 0.01042938,
      "balance_loss_clip": 1.0622673,
      "balance_loss_mlp": 1.03334212,
      "epoch": 0.17507364877051645,
      "flos": 20885825957760.0,
      "grad_norm": 2.4297032226263107,
      "language_loss": 0.81725085,
      "learning_rate": 3.783390210324382e-06,
      "loss": 0.83990347,
      "num_input_tokens_seen": 30754450,
      "step": 1456,
      "time_per_iteration": 2.4665238857269287
    },
    {
      "auxiliary_loss_clip": 0.01177982,
      "auxiliary_loss_mlp": 0.0103402,
      "balance_loss_clip": 1.05898142,
      "balance_loss_mlp": 1.02416682,
      "epoch": 0.17519389166115554,
      "flos": 24717602136960.0,
      "grad_norm": 1.973457794068408,
      "language_loss": 0.73072994,
      "learning_rate": 3.7830374840308676e-06,
      "loss": 0.75284994,
      "num_input_tokens_seen": 30774605,
      "step": 1457,
      "time_per_iteration": 2.5830042362213135
    },
    {
      "auxiliary_loss_clip": 0.01213608,
      "auxiliary_loss_mlp": 0.01038405,
      "balance_loss_clip": 1.06407142,
      "balance_loss_mlp": 1.02771223,
      "epoch": 0.17531413455179462,
      "flos": 23798144770560.0,
      "grad_norm": 2.555148700001538,
      "language_loss": 0.82423353,
      "learning_rate": 3.7826844872519842e-06,
      "loss": 0.84675366,
      "num_input_tokens_seen": 30792460,
      "step": 1458,
      "time_per_iteration": 2.5104057788848877
    },
    {
      "auxiliary_loss_clip": 0.0119335,
      "auxiliary_loss_mlp": 0.01035738,
      "balance_loss_clip": 1.06257606,
      "balance_loss_mlp": 1.0265826,
      "epoch": 0.1754343774424337,
      "flos": 24572379450240.0,
      "grad_norm": 1.8569162347459516,
      "language_loss": 0.72393346,
      "learning_rate": 3.782331220041282e-06,
      "loss": 0.74622434,
      "num_input_tokens_seen": 30812525,
      "step": 1459,
      "time_per_iteration": 2.5745351314544678
    },
    {
      "auxiliary_loss_clip": 0.01187546,
      "auxiliary_loss_mlp": 0.01035122,
      "balance_loss_clip": 1.05800128,
      "balance_loss_mlp": 1.02525711,
      "epoch": 0.17555462033307281,
      "flos": 18114599767680.0,
      "grad_norm": 2.042219647492752,
      "language_loss": 0.82882649,
      "learning_rate": 3.7819776824523504e-06,
      "loss": 0.85105312,
      "num_input_tokens_seen": 30830390,
      "step": 1460,
      "time_per_iteration": 3.3179755210876465
    },
    {
      "auxiliary_loss_clip": 0.01202059,
      "auxiliary_loss_mlp": 0.01039125,
      "balance_loss_clip": 1.06060779,
      "balance_loss_mlp": 1.02904034,
      "epoch": 0.1756748632237119,
      "flos": 28366018364160.0,
      "grad_norm": 1.8475730294686807,
      "language_loss": 0.84003615,
      "learning_rate": 3.7816238745388213e-06,
      "loss": 0.8624481,
      "num_input_tokens_seen": 30849935,
      "step": 1461,
      "time_per_iteration": 2.5845611095428467
    },
    {
      "auxiliary_loss_clip": 0.01200645,
      "auxiliary_loss_mlp": 0.0103381,
      "balance_loss_clip": 1.05853176,
      "balance_loss_mlp": 1.02432132,
      "epoch": 0.17579510611435098,
      "flos": 25732939881600.0,
      "grad_norm": 1.8750138804707794,
      "language_loss": 0.87164271,
      "learning_rate": 3.781269796354367e-06,
      "loss": 0.89398724,
      "num_input_tokens_seen": 30869555,
      "step": 1462,
      "time_per_iteration": 2.5806846618652344
    },
    {
      "auxiliary_loss_clip": 0.01195185,
      "auxiliary_loss_mlp": 0.01037909,
      "balance_loss_clip": 1.06106198,
      "balance_loss_mlp": 1.02833033,
      "epoch": 0.1759153490049901,
      "flos": 18588081870720.0,
      "grad_norm": 1.947386897657048,
      "language_loss": 0.86154115,
      "learning_rate": 3.7809154479527006e-06,
      "loss": 0.88387203,
      "num_input_tokens_seen": 30888760,
      "step": 1463,
      "time_per_iteration": 2.5226638317108154
    },
    {
      "auxiliary_loss_clip": 0.01169522,
      "auxiliary_loss_mlp": 0.01028578,
      "balance_loss_clip": 1.05673873,
      "balance_loss_mlp": 1.01911306,
      "epoch": 0.17603559189562917,
      "flos": 18619323724800.0,
      "grad_norm": 2.001769956131492,
      "language_loss": 0.84472024,
      "learning_rate": 3.780560829387577e-06,
      "loss": 0.86670125,
      "num_input_tokens_seen": 30907260,
      "step": 1464,
      "time_per_iteration": 2.554675579071045
    },
    {
      "auxiliary_loss_clip": 0.01128691,
      "auxiliary_loss_mlp": 0.01005733,
      "balance_loss_clip": 1.0426625,
      "balance_loss_mlp": 1.00286007,
      "epoch": 0.17615583478626826,
      "flos": 60530775373440.0,
      "grad_norm": 0.8495114524484088,
      "language_loss": 0.57915521,
      "learning_rate": 3.7802059407127915e-06,
      "loss": 0.60049939,
      "num_input_tokens_seen": 30965810,
      "step": 1465,
      "time_per_iteration": 3.0502445697784424
    },
    {
      "auxiliary_loss_clip": 0.01186942,
      "auxiliary_loss_mlp": 0.01040673,
      "balance_loss_clip": 1.05628407,
      "balance_loss_mlp": 1.03075504,
      "epoch": 0.17627607767690734,
      "flos": 23616221362560.0,
      "grad_norm": 2.2109287551385486,
      "language_loss": 0.85917604,
      "learning_rate": 3.7798507819821797e-06,
      "loss": 0.8814522,
      "num_input_tokens_seen": 30982935,
      "step": 1466,
      "time_per_iteration": 2.5486092567443848
    },
    {
      "auxiliary_loss_clip": 0.01173108,
      "auxiliary_loss_mlp": 0.0104459,
      "balance_loss_clip": 1.05823112,
      "balance_loss_mlp": 1.03389132,
      "epoch": 0.17639632056754645,
      "flos": 17639070589440.0,
      "grad_norm": 2.0832754892882273,
      "language_loss": 0.78466392,
      "learning_rate": 3.7794953532496197e-06,
      "loss": 0.8068409,
      "num_input_tokens_seen": 30998840,
      "step": 1467,
      "time_per_iteration": 4.221733093261719
    },
    {
      "auxiliary_loss_clip": 0.01071842,
      "auxiliary_loss_mlp": 0.00754799,
      "balance_loss_clip": 1.03470135,
      "balance_loss_mlp": 0.99977905,
      "epoch": 0.17651656345818553,
      "flos": 57932604910080.0,
      "grad_norm": 0.855341136007038,
      "language_loss": 0.57919025,
      "learning_rate": 3.7791396545690295e-06,
      "loss": 0.59745669,
      "num_input_tokens_seen": 31060075,
      "step": 1468,
      "time_per_iteration": 3.8507862091064453
    },
    {
      "auxiliary_loss_clip": 0.01210613,
      "auxiliary_loss_mlp": 0.01038874,
      "balance_loss_clip": 1.06518435,
      "balance_loss_mlp": 1.02934897,
      "epoch": 0.17663680634882462,
      "flos": 22929502170240.0,
      "grad_norm": 2.158932158853507,
      "language_loss": 0.81077164,
      "learning_rate": 3.7787836859943685e-06,
      "loss": 0.8332665,
      "num_input_tokens_seen": 31078800,
      "step": 1469,
      "time_per_iteration": 2.529515504837036
    },
    {
      "auxiliary_loss_clip": 0.01209728,
      "auxiliary_loss_mlp": 0.01038,
      "balance_loss_clip": 1.06318355,
      "balance_loss_mlp": 1.02758694,
      "epoch": 0.17675704923946373,
      "flos": 22637979388800.0,
      "grad_norm": 2.278284758983468,
      "language_loss": 0.78876251,
      "learning_rate": 3.7784274475796363e-06,
      "loss": 0.81123972,
      "num_input_tokens_seen": 31097430,
      "step": 1470,
      "time_per_iteration": 2.5154569149017334
    },
    {
      "auxiliary_loss_clip": 0.01178776,
      "auxiliary_loss_mlp": 0.010363,
      "balance_loss_clip": 1.05563188,
      "balance_loss_mlp": 1.02638745,
      "epoch": 0.1768772921301028,
      "flos": 27126525795840.0,
      "grad_norm": 1.9951574275723658,
      "language_loss": 0.7625432,
      "learning_rate": 3.7780709393788745e-06,
      "loss": 0.78469396,
      "num_input_tokens_seen": 31117905,
      "step": 1471,
      "time_per_iteration": 2.7140955924987793
    },
    {
      "auxiliary_loss_clip": 0.01222892,
      "auxiliary_loss_mlp": 0.01033481,
      "balance_loss_clip": 1.06321156,
      "balance_loss_mlp": 1.02297306,
      "epoch": 0.1769975350207419,
      "flos": 19172133014400.0,
      "grad_norm": 4.366009816233025,
      "language_loss": 0.75224197,
      "learning_rate": 3.777714161446165e-06,
      "loss": 0.77480567,
      "num_input_tokens_seen": 31137610,
      "step": 1472,
      "time_per_iteration": 2.4866795539855957
    },
    {
      "auxiliary_loss_clip": 0.01208347,
      "auxiliary_loss_mlp": 0.01030992,
      "balance_loss_clip": 1.06228113,
      "balance_loss_mlp": 1.02123463,
      "epoch": 0.177117777911381,
      "flos": 36134932291200.0,
      "grad_norm": 2.1389740925599288,
      "language_loss": 0.69545299,
      "learning_rate": 3.7773571138356304e-06,
      "loss": 0.71784639,
      "num_input_tokens_seen": 31157780,
      "step": 1473,
      "time_per_iteration": 2.6275858879089355
    },
    {
      "auxiliary_loss_clip": 0.01150146,
      "auxiliary_loss_mlp": 0.01028438,
      "balance_loss_clip": 1.0546205,
      "balance_loss_mlp": 1.01954544,
      "epoch": 0.17723802080202009,
      "flos": 22090593052800.0,
      "grad_norm": 2.5795402931985296,
      "language_loss": 0.89260077,
      "learning_rate": 3.776999796601435e-06,
      "loss": 0.91438657,
      "num_input_tokens_seen": 31176540,
      "step": 1474,
      "time_per_iteration": 2.667428493499756
    },
    {
      "auxiliary_loss_clip": 0.01214343,
      "auxiliary_loss_mlp": 0.01036738,
      "balance_loss_clip": 1.06284189,
      "balance_loss_mlp": 1.02686131,
      "epoch": 0.17735826369265917,
      "flos": 30222671437440.0,
      "grad_norm": 2.1946095154893928,
      "language_loss": 0.7291131,
      "learning_rate": 3.776642209797783e-06,
      "loss": 0.75162393,
      "num_input_tokens_seen": 31198370,
      "step": 1475,
      "time_per_iteration": 2.5917141437530518
    },
    {
      "auxiliary_loss_clip": 0.01204982,
      "auxiliary_loss_mlp": 0.01032309,
      "balance_loss_clip": 1.05961418,
      "balance_loss_mlp": 1.02152681,
      "epoch": 0.17747850658329825,
      "flos": 21397588980480.0,
      "grad_norm": 2.429136957104276,
      "language_loss": 0.78360379,
      "learning_rate": 3.7762843534789205e-06,
      "loss": 0.80597675,
      "num_input_tokens_seen": 31217120,
      "step": 1476,
      "time_per_iteration": 2.5454788208007812
    },
    {
      "auxiliary_loss_clip": 0.01198183,
      "auxiliary_loss_mlp": 0.01035271,
      "balance_loss_clip": 1.05950713,
      "balance_loss_mlp": 1.02564454,
      "epoch": 0.17759874947393736,
      "flos": 16983341856000.0,
      "grad_norm": 2.0581964809595625,
      "language_loss": 0.88305044,
      "learning_rate": 3.7759262276991343e-06,
      "loss": 0.90538502,
      "num_input_tokens_seen": 31234730,
      "step": 1477,
      "time_per_iteration": 2.535046339035034
    },
    {
      "auxiliary_loss_clip": 0.01200286,
      "auxiliary_loss_mlp": 0.01032532,
      "balance_loss_clip": 1.06151652,
      "balance_loss_mlp": 1.02220833,
      "epoch": 0.17771899236457644,
      "flos": 11546107390080.0,
      "grad_norm": 2.3361819428161628,
      "language_loss": 0.80585402,
      "learning_rate": 3.7755678325127506e-06,
      "loss": 0.82818222,
      "num_input_tokens_seen": 31252410,
      "step": 1478,
      "time_per_iteration": 2.5008137226104736
    },
    {
      "auxiliary_loss_clip": 0.011603,
      "auxiliary_loss_mlp": 0.01032603,
      "balance_loss_clip": 1.05909979,
      "balance_loss_mlp": 1.02286983,
      "epoch": 0.17783923525521553,
      "flos": 18807747494400.0,
      "grad_norm": 1.713092054486288,
      "language_loss": 0.75895846,
      "learning_rate": 3.7752091679741393e-06,
      "loss": 0.78088742,
      "num_input_tokens_seen": 31270200,
      "step": 1479,
      "time_per_iteration": 2.5827689170837402
    },
    {
      "auxiliary_loss_clip": 0.01207346,
      "auxiliary_loss_mlp": 0.01033358,
      "balance_loss_clip": 1.06191838,
      "balance_loss_mlp": 1.02300429,
      "epoch": 0.17795947814585464,
      "flos": 30408365773440.0,
      "grad_norm": 4.302458204812137,
      "language_loss": 0.77533615,
      "learning_rate": 3.774850234137708e-06,
      "loss": 0.7977432,
      "num_input_tokens_seen": 31287495,
      "step": 1480,
      "time_per_iteration": 2.7176499366760254
    },
    {
      "auxiliary_loss_clip": 0.01206287,
      "auxiliary_loss_mlp": 0.01037276,
      "balance_loss_clip": 1.06163311,
      "balance_loss_mlp": 1.02706623,
      "epoch": 0.17807972103649372,
      "flos": 24389055411840.0,
      "grad_norm": 2.197802986589133,
      "language_loss": 0.82569242,
      "learning_rate": 3.7744910310579076e-06,
      "loss": 0.84812808,
      "num_input_tokens_seen": 31306420,
      "step": 1481,
      "time_per_iteration": 2.55305552482605
    },
    {
      "auxiliary_loss_clip": 0.01225942,
      "auxiliary_loss_mlp": 0.01031003,
      "balance_loss_clip": 1.06754196,
      "balance_loss_mlp": 1.02240181,
      "epoch": 0.1781999639271328,
      "flos": 20301559332480.0,
      "grad_norm": 2.1169771261957395,
      "language_loss": 0.85378468,
      "learning_rate": 3.774131558789229e-06,
      "loss": 0.87635422,
      "num_input_tokens_seen": 31325750,
      "step": 1482,
      "time_per_iteration": 2.4669079780578613
    },
    {
      "auxiliary_loss_clip": 0.01224661,
      "auxiliary_loss_mlp": 0.00764944,
      "balance_loss_clip": 1.06543303,
      "balance_loss_mlp": 1.00091124,
      "epoch": 0.1783202068177719,
      "flos": 15924479806080.0,
      "grad_norm": 2.5119747375358408,
      "language_loss": 0.69803882,
      "learning_rate": 3.773771817386203e-06,
      "loss": 0.71793485,
      "num_input_tokens_seen": 31343080,
      "step": 1483,
      "time_per_iteration": 2.470374822616577
    },
    {
      "auxiliary_loss_clip": 0.01192895,
      "auxiliary_loss_mlp": 0.01032248,
      "balance_loss_clip": 1.06067371,
      "balance_loss_mlp": 1.02302694,
      "epoch": 0.178440449708411,
      "flos": 20631758083200.0,
      "grad_norm": 1.4890868257939718,
      "language_loss": 0.79691088,
      "learning_rate": 3.773411806903403e-06,
      "loss": 0.81916225,
      "num_input_tokens_seen": 31362160,
      "step": 1484,
      "time_per_iteration": 2.5353639125823975
    },
    {
      "auxiliary_loss_clip": 0.01152099,
      "auxiliary_loss_mlp": 0.01036826,
      "balance_loss_clip": 1.05368328,
      "balance_loss_mlp": 1.02650261,
      "epoch": 0.17856069259905008,
      "flos": 21686059105920.0,
      "grad_norm": 1.6847964208177657,
      "language_loss": 0.94846934,
      "learning_rate": 3.7730515273954415e-06,
      "loss": 0.97035861,
      "num_input_tokens_seen": 31380770,
      "step": 1485,
      "time_per_iteration": 2.637413263320923
    },
    {
      "auxiliary_loss_clip": 0.01224606,
      "auxiliary_loss_mlp": 0.01034421,
      "balance_loss_clip": 1.06681061,
      "balance_loss_mlp": 1.0253253,
      "epoch": 0.17868093548968916,
      "flos": 26572962320640.0,
      "grad_norm": 1.8637740361500366,
      "language_loss": 0.8491075,
      "learning_rate": 3.772690978916973e-06,
      "loss": 0.87169778,
      "num_input_tokens_seen": 31400525,
      "step": 1486,
      "time_per_iteration": 2.5730369091033936
    },
    {
      "auxiliary_loss_clip": 0.01209227,
      "auxiliary_loss_mlp": 0.01038043,
      "balance_loss_clip": 1.06364071,
      "balance_loss_mlp": 1.02794051,
      "epoch": 0.17880117838032827,
      "flos": 18581006891520.0,
      "grad_norm": 1.9737619373313253,
      "language_loss": 0.86346591,
      "learning_rate": 3.772330161522693e-06,
      "loss": 0.88593858,
      "num_input_tokens_seen": 31418435,
      "step": 1487,
      "time_per_iteration": 3.3004579544067383
    },
    {
      "auxiliary_loss_clip": 0.01193917,
      "auxiliary_loss_mlp": 0.01037744,
      "balance_loss_clip": 1.06570292,
      "balance_loss_mlp": 1.0276525,
      "epoch": 0.17892142127096736,
      "flos": 26541217676160.0,
      "grad_norm": 1.9298624089719283,
      "language_loss": 0.80043507,
      "learning_rate": 3.7719690752673365e-06,
      "loss": 0.82275164,
      "num_input_tokens_seen": 31439230,
      "step": 1488,
      "time_per_iteration": 2.592899799346924
    },
    {
      "auxiliary_loss_clip": 0.01183723,
      "auxiliary_loss_mlp": 0.01036004,
      "balance_loss_clip": 1.06366706,
      "balance_loss_mlp": 1.02635431,
      "epoch": 0.17904166416160644,
      "flos": 23872623621120.0,
      "grad_norm": 2.047057719401131,
      "language_loss": 0.7816599,
      "learning_rate": 3.7716077202056796e-06,
      "loss": 0.80385721,
      "num_input_tokens_seen": 31457705,
      "step": 1489,
      "time_per_iteration": 2.610059976577759
    },
    {
      "auxiliary_loss_clip": 0.01179867,
      "auxiliary_loss_mlp": 0.01033833,
      "balance_loss_clip": 1.05894876,
      "balance_loss_mlp": 1.02436733,
      "epoch": 0.17916190705224552,
      "flos": 19134426712320.0,
      "grad_norm": 2.740640488183298,
      "language_loss": 0.93789661,
      "learning_rate": 3.7712460963925404e-06,
      "loss": 0.9600336,
      "num_input_tokens_seen": 31473645,
      "step": 1490,
      "time_per_iteration": 2.505293846130371
    },
    {
      "auxiliary_loss_clip": 0.01185243,
      "auxiliary_loss_mlp": 0.01031464,
      "balance_loss_clip": 1.05834007,
      "balance_loss_mlp": 1.02176023,
      "epoch": 0.17928214994288463,
      "flos": 25152120961920.0,
      "grad_norm": 2.267990210001713,
      "language_loss": 0.75221699,
      "learning_rate": 3.7708842038827775e-06,
      "loss": 0.77438414,
      "num_input_tokens_seen": 31492605,
      "step": 1491,
      "time_per_iteration": 2.5933423042297363
    },
    {
      "auxiliary_loss_clip": 0.01207974,
      "auxiliary_loss_mlp": 0.01035046,
      "balance_loss_clip": 1.06115985,
      "balance_loss_mlp": 1.02586687,
      "epoch": 0.17940239283352372,
      "flos": 22384629786240.0,
      "grad_norm": 1.8536466907740017,
      "language_loss": 0.85825455,
      "learning_rate": 3.770522042731288e-06,
      "loss": 0.88068473,
      "num_input_tokens_seen": 31514500,
      "step": 1492,
      "time_per_iteration": 2.542797803878784
    },
    {
      "auxiliary_loss_clip": 0.01156358,
      "auxiliary_loss_mlp": 0.01046,
      "balance_loss_clip": 1.05834937,
      "balance_loss_mlp": 1.03580189,
      "epoch": 0.1795226357241628,
      "flos": 23178685795200.0,
      "grad_norm": 1.7988507727588825,
      "language_loss": 0.87881899,
      "learning_rate": 3.7701596129930122e-06,
      "loss": 0.90084255,
      "num_input_tokens_seen": 31533225,
      "step": 1493,
      "time_per_iteration": 4.280670404434204
    },
    {
      "auxiliary_loss_clip": 0.01187977,
      "auxiliary_loss_mlp": 0.01031104,
      "balance_loss_clip": 1.06107438,
      "balance_loss_mlp": 1.02035737,
      "epoch": 0.1796428786148019,
      "flos": 22090413484800.0,
      "grad_norm": 2.064594396199253,
      "language_loss": 0.73572993,
      "learning_rate": 3.7697969147229315e-06,
      "loss": 0.75792074,
      "num_input_tokens_seen": 31551385,
      "step": 1494,
      "time_per_iteration": 2.570464611053467
    },
    {
      "auxiliary_loss_clip": 0.01205212,
      "auxiliary_loss_mlp": 0.01035446,
      "balance_loss_clip": 1.06118536,
      "balance_loss_mlp": 1.02566493,
      "epoch": 0.179763121505441,
      "flos": 21324618501120.0,
      "grad_norm": 2.103074518000646,
      "language_loss": 0.85422015,
      "learning_rate": 3.7694339479760647e-06,
      "loss": 0.87662673,
      "num_input_tokens_seen": 31570415,
      "step": 1495,
      "time_per_iteration": 3.258523464202881
    },
    {
      "auxiliary_loss_clip": 0.01113256,
      "auxiliary_loss_mlp": 0.01003051,
      "balance_loss_clip": 1.03997636,
      "balance_loss_mlp": 1.00044,
      "epoch": 0.17988336439608008,
      "flos": 68161864815360.0,
      "grad_norm": 4.514920265727813,
      "language_loss": 0.57378691,
      "learning_rate": 3.769070712807476e-06,
      "loss": 0.5949499,
      "num_input_tokens_seen": 31632445,
      "step": 1496,
      "time_per_iteration": 3.197551965713501
    },
    {
      "auxiliary_loss_clip": 0.01137732,
      "auxiliary_loss_mlp": 0.01038205,
      "balance_loss_clip": 1.05715156,
      "balance_loss_mlp": 1.02822709,
      "epoch": 0.18000360728671919,
      "flos": 21945047143680.0,
      "grad_norm": 1.764423469370373,
      "language_loss": 0.7895304,
      "learning_rate": 3.768707209272266e-06,
      "loss": 0.81128979,
      "num_input_tokens_seen": 31652575,
      "step": 1497,
      "time_per_iteration": 2.6191747188568115
    },
    {
      "auxiliary_loss_clip": 0.01190722,
      "auxiliary_loss_mlp": 0.01037318,
      "balance_loss_clip": 1.06033218,
      "balance_loss_mlp": 1.02738142,
      "epoch": 0.18012385017735827,
      "flos": 18986330937600.0,
      "grad_norm": 2.0468522117321974,
      "language_loss": 0.76846707,
      "learning_rate": 3.768343437425579e-06,
      "loss": 0.79074752,
      "num_input_tokens_seen": 31671145,
      "step": 1498,
      "time_per_iteration": 2.5120108127593994
    },
    {
      "auxiliary_loss_clip": 0.01126844,
      "auxiliary_loss_mlp": 0.010336,
      "balance_loss_clip": 1.05371094,
      "balance_loss_mlp": 1.0236342,
      "epoch": 0.18024409306799735,
      "flos": 19748103598080.0,
      "grad_norm": 2.3269925917180316,
      "language_loss": 0.8611154,
      "learning_rate": 3.7679793973225987e-06,
      "loss": 0.88271976,
      "num_input_tokens_seen": 31686955,
      "step": 1499,
      "time_per_iteration": 2.649259090423584
    },
    {
      "auxiliary_loss_clip": 0.0107695,
      "auxiliary_loss_mlp": 0.01003823,
      "balance_loss_clip": 1.03430009,
      "balance_loss_mlp": 1.00114083,
      "epoch": 0.18036433595863643,
      "flos": 67227183060480.0,
      "grad_norm": 0.8480529418827248,
      "language_loss": 0.61628759,
      "learning_rate": 3.767615089018549e-06,
      "loss": 0.63709533,
      "num_input_tokens_seen": 31749300,
      "step": 1500,
      "time_per_iteration": 3.1692075729370117
    },
    {
      "auxiliary_loss_clip": 0.01188465,
      "auxiliary_loss_mlp": 0.01036555,
      "balance_loss_clip": 1.05985451,
      "balance_loss_mlp": 1.02629113,
      "epoch": 0.18048457884927555,
      "flos": 18181464935040.0,
      "grad_norm": 2.2498618680409166,
      "language_loss": 0.86419702,
      "learning_rate": 3.7672505125686966e-06,
      "loss": 0.88644719,
      "num_input_tokens_seen": 31765665,
      "step": 1501,
      "time_per_iteration": 2.505552053451538
    },
    {
      "auxiliary_loss_clip": 0.01164955,
      "auxiliary_loss_mlp": 0.01037045,
      "balance_loss_clip": 1.05516553,
      "balance_loss_mlp": 1.02716851,
      "epoch": 0.18060482173991463,
      "flos": 15813767111040.0,
      "grad_norm": 3.0057506041281044,
      "language_loss": 0.84138823,
      "learning_rate": 3.7668856680283455e-06,
      "loss": 0.86340821,
      "num_input_tokens_seen": 31782690,
      "step": 1502,
      "time_per_iteration": 2.561558246612549
    },
    {
      "auxiliary_loss_clip": 0.01199503,
      "auxiliary_loss_mlp": 0.01037237,
      "balance_loss_clip": 1.06205773,
      "balance_loss_mlp": 1.02741385,
      "epoch": 0.1807250646305537,
      "flos": 18587399512320.0,
      "grad_norm": 1.7969620063242147,
      "language_loss": 0.82497483,
      "learning_rate": 3.7665205554528437e-06,
      "loss": 0.84734225,
      "num_input_tokens_seen": 31802045,
      "step": 1503,
      "time_per_iteration": 2.5025320053100586
    },
    {
      "auxiliary_loss_clip": 0.01199605,
      "auxiliary_loss_mlp": 0.0103255,
      "balance_loss_clip": 1.06707644,
      "balance_loss_mlp": 1.02281022,
      "epoch": 0.18084530752119282,
      "flos": 23149131880320.0,
      "grad_norm": 1.7578252908519156,
      "language_loss": 0.74376667,
      "learning_rate": 3.7661551748975782e-06,
      "loss": 0.76608825,
      "num_input_tokens_seen": 31820220,
      "step": 1504,
      "time_per_iteration": 2.5450692176818848
    },
    {
      "auxiliary_loss_clip": 0.01104328,
      "auxiliary_loss_mlp": 0.01003541,
      "balance_loss_clip": 1.03164744,
      "balance_loss_mlp": 1.00079918,
      "epoch": 0.1809655504118319,
      "flos": 59803153568640.0,
      "grad_norm": 0.9104568629095467,
      "language_loss": 0.60551387,
      "learning_rate": 3.7657895264179772e-06,
      "loss": 0.62659252,
      "num_input_tokens_seen": 31876195,
      "step": 1505,
      "time_per_iteration": 3.0797336101531982
    },
    {
      "auxiliary_loss_clip": 0.01184072,
      "auxiliary_loss_mlp": 0.01033478,
      "balance_loss_clip": 1.05677164,
      "balance_loss_mlp": 1.02409005,
      "epoch": 0.181085793302471,
      "flos": 44201941188480.0,
      "grad_norm": 2.9132779773083035,
      "language_loss": 0.74252313,
      "learning_rate": 3.765423610069509e-06,
      "loss": 0.76469862,
      "num_input_tokens_seen": 31901585,
      "step": 1506,
      "time_per_iteration": 2.7813596725463867
    },
    {
      "auxiliary_loss_clip": 0.01196176,
      "auxiliary_loss_mlp": 0.01037243,
      "balance_loss_clip": 1.06472647,
      "balance_loss_mlp": 1.02733707,
      "epoch": 0.18120603619311007,
      "flos": 34898384638080.0,
      "grad_norm": 1.830509107320071,
      "language_loss": 0.72516954,
      "learning_rate": 3.765057425907683e-06,
      "loss": 0.74750376,
      "num_input_tokens_seen": 31923045,
      "step": 1507,
      "time_per_iteration": 2.6557724475860596
    },
    {
      "auxiliary_loss_clip": 0.01210801,
      "auxiliary_loss_mlp": 0.01038067,
      "balance_loss_clip": 1.0616945,
      "balance_loss_mlp": 1.02778554,
      "epoch": 0.18132627908374918,
      "flos": 21506757390720.0,
      "grad_norm": 1.8188248935205904,
      "language_loss": 0.78347516,
      "learning_rate": 3.764690973988048e-06,
      "loss": 0.80596387,
      "num_input_tokens_seen": 31943385,
      "step": 1508,
      "time_per_iteration": 2.527029037475586
    },
    {
      "auxiliary_loss_clip": 0.01181735,
      "auxiliary_loss_mlp": 0.01031418,
      "balance_loss_clip": 1.05971861,
      "balance_loss_mlp": 1.02176225,
      "epoch": 0.18144652197438826,
      "flos": 29057693633280.0,
      "grad_norm": 2.188100403863643,
      "language_loss": 0.74156475,
      "learning_rate": 3.7643242543661967e-06,
      "loss": 0.76369631,
      "num_input_tokens_seen": 31966045,
      "step": 1509,
      "time_per_iteration": 2.610428810119629
    },
    {
      "auxiliary_loss_clip": 0.0109427,
      "auxiliary_loss_mlp": 0.01010746,
      "balance_loss_clip": 1.02890992,
      "balance_loss_mlp": 1.00825465,
      "epoch": 0.18156676486502735,
      "flos": 68675064382080.0,
      "grad_norm": 0.8151927676223004,
      "language_loss": 0.60522521,
      "learning_rate": 3.7639572670977573e-06,
      "loss": 0.6262753,
      "num_input_tokens_seen": 32021540,
      "step": 1510,
      "time_per_iteration": 3.000666618347168
    },
    {
      "auxiliary_loss_clip": 0.01179659,
      "auxiliary_loss_mlp": 0.01038989,
      "balance_loss_clip": 1.05718935,
      "balance_loss_mlp": 1.02893996,
      "epoch": 0.18168700775566646,
      "flos": 26471515334400.0,
      "grad_norm": 1.5609653937695833,
      "language_loss": 0.76758742,
      "learning_rate": 3.7635900122384042e-06,
      "loss": 0.78977394,
      "num_input_tokens_seen": 32044535,
      "step": 1511,
      "time_per_iteration": 2.663465738296509
    },
    {
      "auxiliary_loss_clip": 0.01195938,
      "auxiliary_loss_mlp": 0.0104001,
      "balance_loss_clip": 1.05869639,
      "balance_loss_mlp": 1.0290246,
      "epoch": 0.18180725064630554,
      "flos": 15005668884480.0,
      "grad_norm": 3.4339188296995973,
      "language_loss": 0.8700766,
      "learning_rate": 3.7632224898438477e-06,
      "loss": 0.89243615,
      "num_input_tokens_seen": 32061010,
      "step": 1512,
      "time_per_iteration": 2.4977450370788574
    },
    {
      "auxiliary_loss_clip": 0.0118461,
      "auxiliary_loss_mlp": 0.0103333,
      "balance_loss_clip": 1.05840421,
      "balance_loss_mlp": 1.02357912,
      "epoch": 0.18192749353694462,
      "flos": 19682387665920.0,
      "grad_norm": 1.4877012113122992,
      "language_loss": 0.79553455,
      "learning_rate": 3.762854699969842e-06,
      "loss": 0.81771398,
      "num_input_tokens_seen": 32081520,
      "step": 1513,
      "time_per_iteration": 3.386384963989258
    },
    {
      "auxiliary_loss_clip": 0.01206675,
      "auxiliary_loss_mlp": 0.01041821,
      "balance_loss_clip": 1.0639956,
      "balance_loss_mlp": 1.0308243,
      "epoch": 0.1820477364275837,
      "flos": 20702717400960.0,
      "grad_norm": 1.8971641765236287,
      "language_loss": 0.73319352,
      "learning_rate": 3.762486642672179e-06,
      "loss": 0.75567847,
      "num_input_tokens_seen": 32098460,
      "step": 1514,
      "time_per_iteration": 2.507464647293091
    },
    {
      "auxiliary_loss_clip": 0.01190944,
      "auxiliary_loss_mlp": 0.01038044,
      "balance_loss_clip": 1.05835521,
      "balance_loss_mlp": 1.0280962,
      "epoch": 0.18216797931822282,
      "flos": 17128708197120.0,
      "grad_norm": 2.0553824890871986,
      "language_loss": 0.87007093,
      "learning_rate": 3.7621183180066946e-06,
      "loss": 0.89236081,
      "num_input_tokens_seen": 32116420,
      "step": 1515,
      "time_per_iteration": 2.5627338886260986
    },
    {
      "auxiliary_loss_clip": 0.01191499,
      "auxiliary_loss_mlp": 0.01035312,
      "balance_loss_clip": 1.05834126,
      "balance_loss_mlp": 1.02504253,
      "epoch": 0.1822882222088619,
      "flos": 29242561956480.0,
      "grad_norm": 1.5215141883411794,
      "language_loss": 0.7374227,
      "learning_rate": 3.7617497260292625e-06,
      "loss": 0.75969082,
      "num_input_tokens_seen": 32138475,
      "step": 1516,
      "time_per_iteration": 2.6125621795654297
    },
    {
      "auxiliary_loss_clip": 0.01187473,
      "auxiliary_loss_mlp": 0.01034012,
      "balance_loss_clip": 1.06143785,
      "balance_loss_mlp": 1.02330732,
      "epoch": 0.18240846509950098,
      "flos": 17702739446400.0,
      "grad_norm": 2.605499572940373,
      "language_loss": 0.78701055,
      "learning_rate": 3.7613808667957967e-06,
      "loss": 0.80922544,
      "num_input_tokens_seen": 32151165,
      "step": 1517,
      "time_per_iteration": 2.476233720779419
    },
    {
      "auxiliary_loss_clip": 0.01194993,
      "auxiliary_loss_mlp": 0.01044304,
      "balance_loss_clip": 1.06038189,
      "balance_loss_mlp": 1.03433812,
      "epoch": 0.1825287079901401,
      "flos": 14790025584000.0,
      "grad_norm": 2.1357697738897183,
      "language_loss": 0.90924817,
      "learning_rate": 3.7610117403622547e-06,
      "loss": 0.93164116,
      "num_input_tokens_seen": 32167725,
      "step": 1518,
      "time_per_iteration": 2.5070536136627197
    },
    {
      "auxiliary_loss_clip": 0.01169235,
      "auxiliary_loss_mlp": 0.01040558,
      "balance_loss_clip": 1.05352306,
      "balance_loss_mlp": 1.0300144,
      "epoch": 0.18264895088077918,
      "flos": 21946232292480.0,
      "grad_norm": 1.681642278010146,
      "language_loss": 0.89922565,
      "learning_rate": 3.7606423467846313e-06,
      "loss": 0.92132354,
      "num_input_tokens_seen": 32187330,
      "step": 1519,
      "time_per_iteration": 3.410222291946411
    },
    {
      "auxiliary_loss_clip": 0.01184287,
      "auxiliary_loss_mlp": 0.01040141,
      "balance_loss_clip": 1.06102538,
      "balance_loss_mlp": 1.03022909,
      "epoch": 0.18276919377141826,
      "flos": 20886759711360.0,
      "grad_norm": 1.434769108150616,
      "language_loss": 0.7946586,
      "learning_rate": 3.760272686118964e-06,
      "loss": 0.81690288,
      "num_input_tokens_seen": 32205550,
      "step": 1520,
      "time_per_iteration": 3.417752981185913
    },
    {
      "auxiliary_loss_clip": 0.01193656,
      "auxiliary_loss_mlp": 0.01038381,
      "balance_loss_clip": 1.05914736,
      "balance_loss_mlp": 1.02868962,
      "epoch": 0.18288943666205737,
      "flos": 21469877101440.0,
      "grad_norm": 2.016186871273352,
      "language_loss": 0.92624295,
      "learning_rate": 3.7599027584213297e-06,
      "loss": 0.94856334,
      "num_input_tokens_seen": 32224430,
      "step": 1521,
      "time_per_iteration": 3.2623822689056396
    },
    {
      "auxiliary_loss_clip": 0.01211489,
      "auxiliary_loss_mlp": 0.01038563,
      "balance_loss_clip": 1.06056035,
      "balance_loss_mlp": 1.02835262,
      "epoch": 0.18300967955269645,
      "flos": 21539363961600.0,
      "grad_norm": 1.9658953811751863,
      "language_loss": 0.7793622,
      "learning_rate": 3.7595325637478465e-06,
      "loss": 0.80186266,
      "num_input_tokens_seen": 32242455,
      "step": 1522,
      "time_per_iteration": 2.48207950592041
    },
    {
      "auxiliary_loss_clip": 0.01183828,
      "auxiliary_loss_mlp": 0.01044729,
      "balance_loss_clip": 1.06006742,
      "balance_loss_mlp": 1.03318357,
      "epoch": 0.18312992244333554,
      "flos": 28876237102080.0,
      "grad_norm": 2.0363307076798165,
      "language_loss": 0.81768596,
      "learning_rate": 3.7591621021546723e-06,
      "loss": 0.83997154,
      "num_input_tokens_seen": 32264450,
      "step": 1523,
      "time_per_iteration": 2.598876714706421
    },
    {
      "auxiliary_loss_clip": 0.01198881,
      "auxiliary_loss_mlp": 0.01036027,
      "balance_loss_clip": 1.05824828,
      "balance_loss_mlp": 1.02474356,
      "epoch": 0.18325016533397462,
      "flos": 20120102801280.0,
      "grad_norm": 1.7008505788703276,
      "language_loss": 0.81537443,
      "learning_rate": 3.7587913736980062e-06,
      "loss": 0.83772361,
      "num_input_tokens_seen": 32284090,
      "step": 1524,
      "time_per_iteration": 2.50065016746521
    },
    {
      "auxiliary_loss_clip": 0.01130617,
      "auxiliary_loss_mlp": 0.0103698,
      "balance_loss_clip": 1.04904747,
      "balance_loss_mlp": 1.02687693,
      "epoch": 0.18337040822461373,
      "flos": 23329187781120.0,
      "grad_norm": 1.6063378258604248,
      "language_loss": 0.84336019,
      "learning_rate": 3.7584203784340865e-06,
      "loss": 0.86503613,
      "num_input_tokens_seen": 32303260,
      "step": 1525,
      "time_per_iteration": 2.6171212196350098
    },
    {
      "auxiliary_loss_clip": 0.01187262,
      "auxiliary_loss_mlp": 0.01037775,
      "balance_loss_clip": 1.05623412,
      "balance_loss_mlp": 1.02760017,
      "epoch": 0.1834906511152528,
      "flos": 25009555881600.0,
      "grad_norm": 2.1812027389513156,
      "language_loss": 0.85733509,
      "learning_rate": 3.7580491164191938e-06,
      "loss": 0.8795855,
      "num_input_tokens_seen": 32321570,
      "step": 1526,
      "time_per_iteration": 2.5753986835479736
    },
    {
      "auxiliary_loss_clip": 0.01109682,
      "auxiliary_loss_mlp": 0.01004203,
      "balance_loss_clip": 1.02729988,
      "balance_loss_mlp": 1.00154424,
      "epoch": 0.1836108940058919,
      "flos": 67251493589760.0,
      "grad_norm": 0.8076315705857933,
      "language_loss": 0.61306846,
      "learning_rate": 3.757677587709648e-06,
      "loss": 0.63420737,
      "num_input_tokens_seen": 32384835,
      "step": 1527,
      "time_per_iteration": 3.209385871887207
    },
    {
      "auxiliary_loss_clip": 0.01172461,
      "auxiliary_loss_mlp": 0.01036698,
      "balance_loss_clip": 1.05861878,
      "balance_loss_mlp": 1.02675009,
      "epoch": 0.183731136896531,
      "flos": 25738721971200.0,
      "grad_norm": 2.1933410267974964,
      "language_loss": 0.75526786,
      "learning_rate": 3.7573057923618095e-06,
      "loss": 0.77735949,
      "num_input_tokens_seen": 32404930,
      "step": 1528,
      "time_per_iteration": 2.6907362937927246
    },
    {
      "auxiliary_loss_clip": 0.01158788,
      "auxiliary_loss_mlp": 0.01034745,
      "balance_loss_clip": 1.05195332,
      "balance_loss_mlp": 1.02396226,
      "epoch": 0.1838513797871701,
      "flos": 20449403712000.0,
      "grad_norm": 2.6770094243595057,
      "language_loss": 0.74216211,
      "learning_rate": 3.7569337304320793e-06,
      "loss": 0.76409739,
      "num_input_tokens_seen": 32424515,
      "step": 1529,
      "time_per_iteration": 2.6035349369049072
    },
    {
      "auxiliary_loss_clip": 0.01091968,
      "auxiliary_loss_mlp": 0.01002657,
      "balance_loss_clip": 1.02373588,
      "balance_loss_mlp": 1.00010586,
      "epoch": 0.18397162267780917,
      "flos": 68565141786240.0,
      "grad_norm": 2.002893437752971,
      "language_loss": 0.64455068,
      "learning_rate": 3.756561401976899e-06,
      "loss": 0.66549695,
      "num_input_tokens_seen": 32484220,
      "step": 1530,
      "time_per_iteration": 2.99403715133667
    },
    {
      "auxiliary_loss_clip": 0.01224167,
      "auxiliary_loss_mlp": 0.0103464,
      "balance_loss_clip": 1.06396341,
      "balance_loss_mlp": 1.02453113,
      "epoch": 0.18409186556844825,
      "flos": 31941104976000.0,
      "grad_norm": 2.902315012913277,
      "language_loss": 0.82708448,
      "learning_rate": 3.7561888070527514e-06,
      "loss": 0.84967256,
      "num_input_tokens_seen": 32506260,
      "step": 1531,
      "time_per_iteration": 2.581179618835449
    },
    {
      "auxiliary_loss_clip": 0.01159704,
      "auxiliary_loss_mlp": 0.00764749,
      "balance_loss_clip": 1.05558276,
      "balance_loss_mlp": 1.00165319,
      "epoch": 0.18421210845908736,
      "flos": 20120533764480.0,
      "grad_norm": 2.304512789812378,
      "language_loss": 0.79688239,
      "learning_rate": 3.7558159457161577e-06,
      "loss": 0.81612694,
      "num_input_tokens_seen": 32524225,
      "step": 1532,
      "time_per_iteration": 2.5616469383239746
    },
    {
      "auxiliary_loss_clip": 0.01195696,
      "auxiliary_loss_mlp": 0.00765474,
      "balance_loss_clip": 1.06339979,
      "balance_loss_mlp": 1.00163865,
      "epoch": 0.18433235134972645,
      "flos": 23110491824640.0,
      "grad_norm": 2.171485483059511,
      "language_loss": 0.78154618,
      "learning_rate": 3.755442818023681e-06,
      "loss": 0.80115783,
      "num_input_tokens_seen": 32543850,
      "step": 1533,
      "time_per_iteration": 2.5394325256347656
    },
    {
      "auxiliary_loss_clip": 0.01180838,
      "auxiliary_loss_mlp": 0.01032199,
      "balance_loss_clip": 1.0600965,
      "balance_loss_mlp": 1.02269816,
      "epoch": 0.18445259424036553,
      "flos": 18291351617280.0,
      "grad_norm": 2.746214856149023,
      "language_loss": 0.75996596,
      "learning_rate": 3.7550694240319246e-06,
      "loss": 0.78209633,
      "num_input_tokens_seen": 32561725,
      "step": 1534,
      "time_per_iteration": 2.5386745929718018
    },
    {
      "auxiliary_loss_clip": 0.01209616,
      "auxiliary_loss_mlp": 0.01032439,
      "balance_loss_clip": 1.06069005,
      "balance_loss_mlp": 1.02284217,
      "epoch": 0.18457283713100464,
      "flos": 21324079797120.0,
      "grad_norm": 2.347459196984291,
      "language_loss": 0.76074064,
      "learning_rate": 3.7546957637975326e-06,
      "loss": 0.78316116,
      "num_input_tokens_seen": 32579135,
      "step": 1535,
      "time_per_iteration": 2.4987564086914062
    },
    {
      "auxiliary_loss_clip": 0.01133956,
      "auxiliary_loss_mlp": 0.01035106,
      "balance_loss_clip": 1.04652584,
      "balance_loss_mlp": 1.02533722,
      "epoch": 0.18469308002164372,
      "flos": 20375679047040.0,
      "grad_norm": 1.5648573358390365,
      "language_loss": 0.73890042,
      "learning_rate": 3.7543218373771873e-06,
      "loss": 0.76059109,
      "num_input_tokens_seen": 32598460,
      "step": 1536,
      "time_per_iteration": 2.6410598754882812
    },
    {
      "auxiliary_loss_clip": 0.01138834,
      "auxiliary_loss_mlp": 0.00764842,
      "balance_loss_clip": 1.05341995,
      "balance_loss_mlp": 1.00161314,
      "epoch": 0.1848133229122828,
      "flos": 26435892021120.0,
      "grad_norm": 1.4305178161697758,
      "language_loss": 0.78245389,
      "learning_rate": 3.753947644827615e-06,
      "loss": 0.8014906,
      "num_input_tokens_seen": 32621920,
      "step": 1537,
      "time_per_iteration": 2.7086331844329834
    },
    {
      "auxiliary_loss_clip": 0.01097863,
      "auxiliary_loss_mlp": 0.01007098,
      "balance_loss_clip": 1.02693629,
      "balance_loss_mlp": 1.00420141,
      "epoch": 0.1849335658029219,
      "flos": 70547447612160.0,
      "grad_norm": 0.9466289929534218,
      "language_loss": 0.57233226,
      "learning_rate": 3.753573186205579e-06,
      "loss": 0.59338188,
      "num_input_tokens_seen": 32690040,
      "step": 1538,
      "time_per_iteration": 3.2292301654815674
    },
    {
      "auxiliary_loss_clip": 0.01178675,
      "auxiliary_loss_mlp": 0.00764978,
      "balance_loss_clip": 1.05461168,
      "balance_loss_mlp": 1.00152361,
      "epoch": 0.185053808693561,
      "flos": 17384140788480.0,
      "grad_norm": 2.3997511599309775,
      "language_loss": 0.77776992,
      "learning_rate": 3.753198461567885e-06,
      "loss": 0.79720652,
      "num_input_tokens_seen": 32707285,
      "step": 1539,
      "time_per_iteration": 2.5214316844940186
    },
    {
      "auxiliary_loss_clip": 0.01172123,
      "auxiliary_loss_mlp": 0.01039535,
      "balance_loss_clip": 1.06048667,
      "balance_loss_mlp": 1.0299505,
      "epoch": 0.18517405158420008,
      "flos": 28986159697920.0,
      "grad_norm": 1.7380681267525369,
      "language_loss": 0.91797101,
      "learning_rate": 3.7528234709713783e-06,
      "loss": 0.94008756,
      "num_input_tokens_seen": 32730030,
      "step": 1540,
      "time_per_iteration": 3.4087469577789307
    },
    {
      "auxiliary_loss_clip": 0.01207114,
      "auxiliary_loss_mlp": 0.01037142,
      "balance_loss_clip": 1.06225419,
      "balance_loss_mlp": 1.02755165,
      "epoch": 0.18529429447483917,
      "flos": 26794962328320.0,
      "grad_norm": 2.1009670548790442,
      "language_loss": 0.84451687,
      "learning_rate": 3.7524482144729447e-06,
      "loss": 0.86695945,
      "num_input_tokens_seen": 32749485,
      "step": 1541,
      "time_per_iteration": 2.5261754989624023
    },
    {
      "auxiliary_loss_clip": 0.01169729,
      "auxiliary_loss_mlp": 0.01043812,
      "balance_loss_clip": 1.05321252,
      "balance_loss_mlp": 1.03370941,
      "epoch": 0.18541453736547828,
      "flos": 13581595301760.0,
      "grad_norm": 2.0002675676265924,
      "language_loss": 0.83764476,
      "learning_rate": 3.7520726921295106e-06,
      "loss": 0.85978013,
      "num_input_tokens_seen": 32766205,
      "step": 1542,
      "time_per_iteration": 2.537696123123169
    },
    {
      "auxiliary_loss_clip": 0.01200029,
      "auxiliary_loss_mlp": 0.01037473,
      "balance_loss_clip": 1.05551124,
      "balance_loss_mlp": 1.02770996,
      "epoch": 0.18553478025611736,
      "flos": 24025424077440.0,
      "grad_norm": 2.1540854720962472,
      "language_loss": 0.72559702,
      "learning_rate": 3.751696903998042e-06,
      "loss": 0.74797213,
      "num_input_tokens_seen": 32784840,
      "step": 1543,
      "time_per_iteration": 2.530993938446045
    },
    {
      "auxiliary_loss_clip": 0.01202769,
      "auxiliary_loss_mlp": 0.01035386,
      "balance_loss_clip": 1.06119204,
      "balance_loss_mlp": 1.02571845,
      "epoch": 0.18565502314675644,
      "flos": 25885165720320.0,
      "grad_norm": 1.8648186820523522,
      "language_loss": 0.70149606,
      "learning_rate": 3.7513208501355456e-06,
      "loss": 0.72387761,
      "num_input_tokens_seen": 32805945,
      "step": 1544,
      "time_per_iteration": 2.5216071605682373
    },
    {
      "auxiliary_loss_clip": 0.01184486,
      "auxiliary_loss_mlp": 0.01036935,
      "balance_loss_clip": 1.05612588,
      "balance_loss_mlp": 1.027601,
      "epoch": 0.18577526603739553,
      "flos": 19610063631360.0,
      "grad_norm": 1.9026032211099624,
      "language_loss": 0.83359313,
      "learning_rate": 3.750944530599069e-06,
      "loss": 0.85580736,
      "num_input_tokens_seen": 32825515,
      "step": 1545,
      "time_per_iteration": 2.5397531986236572
    },
    {
      "auxiliary_loss_clip": 0.01212013,
      "auxiliary_loss_mlp": 0.00765541,
      "balance_loss_clip": 1.06248677,
      "balance_loss_mlp": 1.00149095,
      "epoch": 0.18589550892803464,
      "flos": 18474891137280.0,
      "grad_norm": 2.4093407187983225,
      "language_loss": 0.80668128,
      "learning_rate": 3.7505679454456992e-06,
      "loss": 0.82645679,
      "num_input_tokens_seen": 32842125,
      "step": 1546,
      "time_per_iteration": 3.226943254470825
    },
    {
      "auxiliary_loss_clip": 0.01123606,
      "auxiliary_loss_mlp": 0.01035045,
      "balance_loss_clip": 1.04883432,
      "balance_loss_mlp": 1.02495944,
      "epoch": 0.18601575181867372,
      "flos": 23549966726400.0,
      "grad_norm": 1.9983446963873355,
      "language_loss": 0.69847667,
      "learning_rate": 3.750191094732564e-06,
      "loss": 0.72006321,
      "num_input_tokens_seen": 32862990,
      "step": 1547,
      "time_per_iteration": 3.718635320663452
    },
    {
      "auxiliary_loss_clip": 0.01124866,
      "auxiliary_loss_mlp": 0.0103661,
      "balance_loss_clip": 1.04918528,
      "balance_loss_mlp": 1.02711463,
      "epoch": 0.1861359947093128,
      "flos": 26360192108160.0,
      "grad_norm": 1.8191058941031168,
      "language_loss": 0.75083649,
      "learning_rate": 3.7498139785168313e-06,
      "loss": 0.77245122,
      "num_input_tokens_seen": 32883595,
      "step": 1548,
      "time_per_iteration": 3.481144428253174
    },
    {
      "auxiliary_loss_clip": 0.01202968,
      "auxiliary_loss_mlp": 0.01040031,
      "balance_loss_clip": 1.06250334,
      "balance_loss_mlp": 1.02975488,
      "epoch": 0.1862562375999519,
      "flos": 23331198942720.0,
      "grad_norm": 1.8334972983197446,
      "language_loss": 0.77460778,
      "learning_rate": 3.749436596855709e-06,
      "loss": 0.79703778,
      "num_input_tokens_seen": 32902895,
      "step": 1549,
      "time_per_iteration": 2.5085291862487793
    },
    {
      "auxiliary_loss_clip": 0.0119827,
      "auxiliary_loss_mlp": 0.01032316,
      "balance_loss_clip": 1.05699182,
      "balance_loss_mlp": 1.0219748,
      "epoch": 0.186376480490591,
      "flos": 16648222942080.0,
      "grad_norm": 2.4494272208097,
      "language_loss": 0.90518767,
      "learning_rate": 3.749058949806446e-06,
      "loss": 0.92749357,
      "num_input_tokens_seen": 32919620,
      "step": 1550,
      "time_per_iteration": 2.480605363845825
    },
    {
      "auxiliary_loss_clip": 0.0120421,
      "auxiliary_loss_mlp": 0.01031887,
      "balance_loss_clip": 1.05871773,
      "balance_loss_mlp": 1.02238619,
      "epoch": 0.18649672338123008,
      "flos": 21468656039040.0,
      "grad_norm": 1.570947577975064,
      "language_loss": 0.84367442,
      "learning_rate": 3.748681037426331e-06,
      "loss": 0.8660354,
      "num_input_tokens_seen": 32938830,
      "step": 1551,
      "time_per_iteration": 2.526944398880005
    },
    {
      "auxiliary_loss_clip": 0.01221474,
      "auxiliary_loss_mlp": 0.01039448,
      "balance_loss_clip": 1.06297278,
      "balance_loss_mlp": 1.03006029,
      "epoch": 0.1866169662718692,
      "flos": 12312728386560.0,
      "grad_norm": 2.2582317905945386,
      "language_loss": 0.91802633,
      "learning_rate": 3.7483028597726936e-06,
      "loss": 0.94063556,
      "num_input_tokens_seen": 32955600,
      "step": 1552,
      "time_per_iteration": 2.4482104778289795
    },
    {
      "auxiliary_loss_clip": 0.01172441,
      "auxiliary_loss_mlp": 0.01039889,
      "balance_loss_clip": 1.05642521,
      "balance_loss_mlp": 1.02983928,
      "epoch": 0.18673720916250827,
      "flos": 23581280407680.0,
      "grad_norm": 2.5686354009574552,
      "language_loss": 0.62850809,
      "learning_rate": 3.7479244169029017e-06,
      "loss": 0.65063143,
      "num_input_tokens_seen": 32975390,
      "step": 1553,
      "time_per_iteration": 2.60425066947937
    },
    {
      "auxiliary_loss_clip": 0.01205993,
      "auxiliary_loss_mlp": 0.01028614,
      "balance_loss_clip": 1.05682731,
      "balance_loss_mlp": 1.01936948,
      "epoch": 0.18685745205314735,
      "flos": 19718370115200.0,
      "grad_norm": 3.81416848843719,
      "language_loss": 0.73412538,
      "learning_rate": 3.7475457088743658e-06,
      "loss": 0.75647146,
      "num_input_tokens_seen": 32992640,
      "step": 1554,
      "time_per_iteration": 2.513031244277954
    },
    {
      "auxiliary_loss_clip": 0.01181296,
      "auxiliary_loss_mlp": 0.01039496,
      "balance_loss_clip": 1.05780029,
      "balance_loss_mlp": 1.02874887,
      "epoch": 0.18697769494378644,
      "flos": 34204123589760.0,
      "grad_norm": 2.08819388970875,
      "language_loss": 0.74445719,
      "learning_rate": 3.7471667357445348e-06,
      "loss": 0.76666504,
      "num_input_tokens_seen": 33012470,
      "step": 1555,
      "time_per_iteration": 2.6156959533691406
    },
    {
      "auxiliary_loss_clip": 0.01147034,
      "auxiliary_loss_mlp": 0.0102712,
      "balance_loss_clip": 1.05498934,
      "balance_loss_mlp": 1.01796448,
      "epoch": 0.18709793783442555,
      "flos": 34241327101440.0,
      "grad_norm": 1.8715351199422372,
      "language_loss": 0.72317046,
      "learning_rate": 3.7467874975709e-06,
      "loss": 0.74491203,
      "num_input_tokens_seen": 33033275,
      "step": 1556,
      "time_per_iteration": 2.774231433868408
    },
    {
      "auxiliary_loss_clip": 0.01210674,
      "auxiliary_loss_mlp": 0.01044213,
      "balance_loss_clip": 1.06219125,
      "balance_loss_mlp": 1.03426504,
      "epoch": 0.18721818072506463,
      "flos": 40734550529280.0,
      "grad_norm": 1.9946060261181777,
      "language_loss": 0.77806765,
      "learning_rate": 3.7464079944109904e-06,
      "loss": 0.8006165,
      "num_input_tokens_seen": 33055135,
      "step": 1557,
      "time_per_iteration": 2.656710624694824
    },
    {
      "auxiliary_loss_clip": 0.01176867,
      "auxiliary_loss_mlp": 0.01032369,
      "balance_loss_clip": 1.05627906,
      "balance_loss_mlp": 1.02304721,
      "epoch": 0.18733842361570371,
      "flos": 22157386392960.0,
      "grad_norm": 1.8940492322244606,
      "language_loss": 0.77750742,
      "learning_rate": 3.746028226322376e-06,
      "loss": 0.79959977,
      "num_input_tokens_seen": 33071015,
      "step": 1558,
      "time_per_iteration": 2.5436840057373047
    },
    {
      "auxiliary_loss_clip": 0.01186655,
      "auxiliary_loss_mlp": 0.01032227,
      "balance_loss_clip": 1.05809438,
      "balance_loss_mlp": 1.02285719,
      "epoch": 0.18745866650634282,
      "flos": 18914940656640.0,
      "grad_norm": 1.691101638614279,
      "language_loss": 0.75481844,
      "learning_rate": 3.745648193362669e-06,
      "loss": 0.77700734,
      "num_input_tokens_seen": 33090370,
      "step": 1559,
      "time_per_iteration": 2.5269317626953125
    },
    {
      "auxiliary_loss_clip": 0.01191393,
      "auxiliary_loss_mlp": 0.01034652,
      "balance_loss_clip": 1.05902386,
      "balance_loss_mlp": 1.02544856,
      "epoch": 0.1875789093969819,
      "flos": 19314626267520.0,
      "grad_norm": 1.990582355898355,
      "language_loss": 0.72553068,
      "learning_rate": 3.745267895589518e-06,
      "loss": 0.74779111,
      "num_input_tokens_seen": 33108910,
      "step": 1560,
      "time_per_iteration": 2.5093231201171875
    },
    {
      "auxiliary_loss_clip": 0.0119095,
      "auxiliary_loss_mlp": 0.01035706,
      "balance_loss_clip": 1.06006432,
      "balance_loss_mlp": 1.02597272,
      "epoch": 0.187699152287621,
      "flos": 17018965169280.0,
      "grad_norm": 1.9053093804656958,
      "language_loss": 0.82285178,
      "learning_rate": 3.7448873330606154e-06,
      "loss": 0.84511834,
      "num_input_tokens_seen": 33126680,
      "step": 1561,
      "time_per_iteration": 2.5056815147399902
    },
    {
      "auxiliary_loss_clip": 0.01169884,
      "auxiliary_loss_mlp": 0.01035417,
      "balance_loss_clip": 1.05939007,
      "balance_loss_mlp": 1.02529597,
      "epoch": 0.18781939517826007,
      "flos": 22346384780160.0,
      "grad_norm": 2.055608567219313,
      "language_loss": 0.87481058,
      "learning_rate": 3.7445065058336914e-06,
      "loss": 0.89686364,
      "num_input_tokens_seen": 33145550,
      "step": 1562,
      "time_per_iteration": 2.571474313735962
    },
    {
      "auxiliary_loss_clip": 0.01146731,
      "auxiliary_loss_mlp": 0.01030448,
      "balance_loss_clip": 1.05012631,
      "balance_loss_mlp": 1.02114391,
      "epoch": 0.18793963806889918,
      "flos": 14611478054400.0,
      "grad_norm": 1.7780072356699181,
      "language_loss": 0.86268795,
      "learning_rate": 3.7441254139665176e-06,
      "loss": 0.88445973,
      "num_input_tokens_seen": 33161735,
      "step": 1563,
      "time_per_iteration": 2.5689122676849365
    },
    {
      "auxiliary_loss_clip": 0.01220112,
      "auxiliary_loss_mlp": 0.01038636,
      "balance_loss_clip": 1.06428432,
      "balance_loss_mlp": 1.02958822,
      "epoch": 0.18805988095953827,
      "flos": 17457075354240.0,
      "grad_norm": 1.889981719556016,
      "language_loss": 0.82513249,
      "learning_rate": 3.743744057516905e-06,
      "loss": 0.84771991,
      "num_input_tokens_seen": 33179795,
      "step": 1564,
      "time_per_iteration": 2.4595906734466553
    },
    {
      "auxiliary_loss_clip": 0.01160012,
      "auxiliary_loss_mlp": 0.01040223,
      "balance_loss_clip": 1.05430913,
      "balance_loss_mlp": 1.02976871,
      "epoch": 0.18818012385017735,
      "flos": 15043877976960.0,
      "grad_norm": 2.6625675410773963,
      "language_loss": 0.87648696,
      "learning_rate": 3.743362436542706e-06,
      "loss": 0.89848924,
      "num_input_tokens_seen": 33194485,
      "step": 1565,
      "time_per_iteration": 2.594287633895874
    },
    {
      "auxiliary_loss_clip": 0.01216345,
      "auxiliary_loss_mlp": 0.01031719,
      "balance_loss_clip": 1.06055367,
      "balance_loss_mlp": 1.02247405,
      "epoch": 0.18830036674081646,
      "flos": 47551975136640.0,
      "grad_norm": 1.816273244239361,
      "language_loss": 0.76638985,
      "learning_rate": 3.7429805511018115e-06,
      "loss": 0.78887045,
      "num_input_tokens_seen": 33216145,
      "step": 1566,
      "time_per_iteration": 2.6811671257019043
    },
    {
      "auxiliary_loss_clip": 0.01170264,
      "auxiliary_loss_mlp": 0.00765555,
      "balance_loss_clip": 1.05768275,
      "balance_loss_mlp": 1.00140524,
      "epoch": 0.18842060963145554,
      "flos": 30044626698240.0,
      "grad_norm": 1.7407027727474496,
      "language_loss": 0.77963156,
      "learning_rate": 3.7425984012521524e-06,
      "loss": 0.79898977,
      "num_input_tokens_seen": 33236345,
      "step": 1567,
      "time_per_iteration": 3.412440061569214
    },
    {
      "auxiliary_loss_clip": 0.01082143,
      "auxiliary_loss_mlp": 0.00755266,
      "balance_loss_clip": 1.02781129,
      "balance_loss_mlp": 0.99990445,
      "epoch": 0.18854085252209463,
      "flos": 70318372625280.0,
      "grad_norm": 0.7348696941141475,
      "language_loss": 0.60379803,
      "learning_rate": 3.7422159870517025e-06,
      "loss": 0.62217212,
      "num_input_tokens_seen": 33301600,
      "step": 1568,
      "time_per_iteration": 3.1371662616729736
    },
    {
      "auxiliary_loss_clip": 0.01185823,
      "auxiliary_loss_mlp": 0.01033086,
      "balance_loss_clip": 1.05699897,
      "balance_loss_mlp": 1.02343059,
      "epoch": 0.1886610954127337,
      "flos": 21289318410240.0,
      "grad_norm": 1.5667916484034317,
      "language_loss": 0.7874403,
      "learning_rate": 3.7418333085584717e-06,
      "loss": 0.80962938,
      "num_input_tokens_seen": 33322785,
      "step": 1569,
      "time_per_iteration": 2.583754539489746
    },
    {
      "auxiliary_loss_clip": 0.01176572,
      "auxiliary_loss_mlp": 0.01034576,
      "balance_loss_clip": 1.05920696,
      "balance_loss_mlp": 1.02469969,
      "epoch": 0.18878133830337282,
      "flos": 17266819991040.0,
      "grad_norm": 2.183867635388492,
      "language_loss": 0.90675151,
      "learning_rate": 3.7414503658305128e-06,
      "loss": 0.92886305,
      "num_input_tokens_seen": 33340020,
      "step": 1570,
      "time_per_iteration": 2.550875186920166
    },
    {
      "auxiliary_loss_clip": 0.01162559,
      "auxiliary_loss_mlp": 0.01034235,
      "balance_loss_clip": 1.05030751,
      "balance_loss_mlp": 1.02482903,
      "epoch": 0.1889015811940119,
      "flos": 25775207210880.0,
      "grad_norm": 2.513363555119363,
      "language_loss": 0.77377975,
      "learning_rate": 3.7410671589259185e-06,
      "loss": 0.7957477,
      "num_input_tokens_seen": 33358620,
      "step": 1571,
      "time_per_iteration": 2.6687686443328857
    },
    {
      "auxiliary_loss_clip": 0.01221091,
      "auxiliary_loss_mlp": 0.01035802,
      "balance_loss_clip": 1.0634793,
      "balance_loss_mlp": 1.02565145,
      "epoch": 0.18902182408465099,
      "flos": 21032197879680.0,
      "grad_norm": 2.0974020198640164,
      "language_loss": 0.79677409,
      "learning_rate": 3.7406836879028205e-06,
      "loss": 0.81934303,
      "num_input_tokens_seen": 33378845,
      "step": 1572,
      "time_per_iteration": 3.241908073425293
    },
    {
      "auxiliary_loss_clip": 0.01204745,
      "auxiliary_loss_mlp": 0.01033645,
      "balance_loss_clip": 1.06179833,
      "balance_loss_mlp": 1.02382195,
      "epoch": 0.1891420669752901,
      "flos": 22272121411200.0,
      "grad_norm": 2.877257024225958,
      "language_loss": 0.76615435,
      "learning_rate": 3.7402999528193907e-06,
      "loss": 0.78853822,
      "num_input_tokens_seen": 33398345,
      "step": 1573,
      "time_per_iteration": 2.522653818130493
    },
    {
      "auxiliary_loss_clip": 0.01159945,
      "auxiliary_loss_mlp": 0.00765082,
      "balance_loss_clip": 1.05522072,
      "balance_loss_mlp": 1.00125611,
      "epoch": 0.18926230986592918,
      "flos": 22017802141440.0,
      "grad_norm": 2.5088768662122134,
      "language_loss": 0.85467756,
      "learning_rate": 3.739915953733842e-06,
      "loss": 0.87392783,
      "num_input_tokens_seen": 33416390,
      "step": 1574,
      "time_per_iteration": 4.194178819656372
    },
    {
      "auxiliary_loss_clip": 0.01218979,
      "auxiliary_loss_mlp": 0.01032959,
      "balance_loss_clip": 1.06255531,
      "balance_loss_mlp": 1.0234108,
      "epoch": 0.18938255275656826,
      "flos": 24462672336000.0,
      "grad_norm": 1.5902777594996012,
      "language_loss": 0.82046533,
      "learning_rate": 3.7395316907044264e-06,
      "loss": 0.84298474,
      "num_input_tokens_seen": 33437175,
      "step": 1575,
      "time_per_iteration": 2.5510268211364746
    },
    {
      "auxiliary_loss_clip": 0.01203771,
      "auxiliary_loss_mlp": 0.01036116,
      "balance_loss_clip": 1.05937088,
      "balance_loss_mlp": 1.0265733,
      "epoch": 0.18950279564720737,
      "flos": 24427049022720.0,
      "grad_norm": 1.5638724528012768,
      "language_loss": 0.79413855,
      "learning_rate": 3.7391471637894364e-06,
      "loss": 0.81653738,
      "num_input_tokens_seen": 33459440,
      "step": 1576,
      "time_per_iteration": 2.5726404190063477
    },
    {
      "auxiliary_loss_clip": 0.01177521,
      "auxiliary_loss_mlp": 0.01036276,
      "balance_loss_clip": 1.05484235,
      "balance_loss_mlp": 1.02683473,
      "epoch": 0.18962303853784646,
      "flos": 19756291898880.0,
      "grad_norm": 1.8238264378690268,
      "language_loss": 0.85006297,
      "learning_rate": 3.738762373047205e-06,
      "loss": 0.87220097,
      "num_input_tokens_seen": 33479360,
      "step": 1577,
      "time_per_iteration": 2.580717086791992
    },
    {
      "auxiliary_loss_clip": 0.01176726,
      "auxiliary_loss_mlp": 0.01036154,
      "balance_loss_clip": 1.05898416,
      "balance_loss_mlp": 1.02662933,
      "epoch": 0.18974328142848554,
      "flos": 21032054225280.0,
      "grad_norm": 1.6618809774010064,
      "language_loss": 0.83225596,
      "learning_rate": 3.738377318536103e-06,
      "loss": 0.85438478,
      "num_input_tokens_seen": 33499245,
      "step": 1578,
      "time_per_iteration": 2.584155559539795
    },
    {
      "auxiliary_loss_clip": 0.01216222,
      "auxiliary_loss_mlp": 0.01035089,
      "balance_loss_clip": 1.06367838,
      "balance_loss_mlp": 1.02635121,
      "epoch": 0.18986352431912462,
      "flos": 12966122736000.0,
      "grad_norm": 1.9758848388073804,
      "language_loss": 0.70866024,
      "learning_rate": 3.7379920003145447e-06,
      "loss": 0.73117328,
      "num_input_tokens_seen": 33513520,
      "step": 1579,
      "time_per_iteration": 2.4178686141967773
    },
    {
      "auxiliary_loss_clip": 0.01182925,
      "auxiliary_loss_mlp": 0.01037389,
      "balance_loss_clip": 1.05954337,
      "balance_loss_mlp": 1.02701235,
      "epoch": 0.18998376720976373,
      "flos": 23767908497280.0,
      "grad_norm": 2.258022130184419,
      "language_loss": 0.837309,
      "learning_rate": 3.7376064184409817e-06,
      "loss": 0.85951215,
      "num_input_tokens_seen": 33533100,
      "step": 1580,
      "time_per_iteration": 2.534109592437744
    },
    {
      "auxiliary_loss_clip": 0.01188387,
      "auxiliary_loss_mlp": 0.01033981,
      "balance_loss_clip": 1.05996442,
      "balance_loss_mlp": 1.02377629,
      "epoch": 0.19010401010040281,
      "flos": 22966023323520.0,
      "grad_norm": 1.4164268897440808,
      "language_loss": 0.87074792,
      "learning_rate": 3.7372205729739063e-06,
      "loss": 0.89297158,
      "num_input_tokens_seen": 33554915,
      "step": 1581,
      "time_per_iteration": 2.535048723220825
    },
    {
      "auxiliary_loss_clip": 0.01206932,
      "auxiliary_loss_mlp": 0.01030462,
      "balance_loss_clip": 1.06021559,
      "balance_loss_mlp": 1.02000725,
      "epoch": 0.1902242529910419,
      "flos": 19135647774720.0,
      "grad_norm": 2.822840903334306,
      "language_loss": 0.72001266,
      "learning_rate": 3.7368344639718514e-06,
      "loss": 0.74238664,
      "num_input_tokens_seen": 33572850,
      "step": 1582,
      "time_per_iteration": 2.4754326343536377
    },
    {
      "auxiliary_loss_clip": 0.01205246,
      "auxiliary_loss_mlp": 0.01041084,
      "balance_loss_clip": 1.06047106,
      "balance_loss_mlp": 1.03226805,
      "epoch": 0.190344495881681,
      "flos": 25483935824640.0,
      "grad_norm": 1.6753643999543122,
      "language_loss": 0.80642939,
      "learning_rate": 3.7364480914933895e-06,
      "loss": 0.82889271,
      "num_input_tokens_seen": 33593090,
      "step": 1583,
      "time_per_iteration": 2.531172513961792
    },
    {
      "auxiliary_loss_clip": 0.01156561,
      "auxiliary_loss_mlp": 0.00765206,
      "balance_loss_clip": 1.05529881,
      "balance_loss_mlp": 1.00116158,
      "epoch": 0.1904647387723201,
      "flos": 26792843425920.0,
      "grad_norm": 1.7764484770584368,
      "language_loss": 0.80815411,
      "learning_rate": 3.7360614555971325e-06,
      "loss": 0.82737184,
      "num_input_tokens_seen": 33612745,
      "step": 1584,
      "time_per_iteration": 2.633847236633301
    },
    {
      "auxiliary_loss_clip": 0.01202965,
      "auxiliary_loss_mlp": 0.00764723,
      "balance_loss_clip": 1.06017041,
      "balance_loss_mlp": 1.00108933,
      "epoch": 0.19058498166295917,
      "flos": 23987753688960.0,
      "grad_norm": 1.9150613475059173,
      "language_loss": 0.84991652,
      "learning_rate": 3.735674556341733e-06,
      "loss": 0.86959338,
      "num_input_tokens_seen": 33632360,
      "step": 1585,
      "time_per_iteration": 2.5255250930786133
    },
    {
      "auxiliary_loss_clip": 0.01188029,
      "auxiliary_loss_mlp": 0.01038093,
      "balance_loss_clip": 1.06202781,
      "balance_loss_mlp": 1.02861571,
      "epoch": 0.19070522455359826,
      "flos": 28293299280000.0,
      "grad_norm": 2.195488658542415,
      "language_loss": 0.82918012,
      "learning_rate": 3.7352873937858835e-06,
      "loss": 0.85144138,
      "num_input_tokens_seen": 33653895,
      "step": 1586,
      "time_per_iteration": 2.5863850116729736
    },
    {
      "auxiliary_loss_clip": 0.01168593,
      "auxiliary_loss_mlp": 0.00765267,
      "balance_loss_clip": 1.05731654,
      "balance_loss_mlp": 1.00098729,
      "epoch": 0.19082546744423737,
      "flos": 25660220797440.0,
      "grad_norm": 1.8382884467899543,
      "language_loss": 0.71705234,
      "learning_rate": 3.734899967988316e-06,
      "loss": 0.73639095,
      "num_input_tokens_seen": 33672075,
      "step": 1587,
      "time_per_iteration": 2.5889408588409424
    },
    {
      "auxiliary_loss_clip": 0.0116527,
      "auxiliary_loss_mlp": 0.01031333,
      "balance_loss_clip": 1.05372965,
      "balance_loss_mlp": 1.02186811,
      "epoch": 0.19094571033487645,
      "flos": 19719483436800.0,
      "grad_norm": 1.7617970397877933,
      "language_loss": 0.84209192,
      "learning_rate": 3.7345122790078026e-06,
      "loss": 0.86405796,
      "num_input_tokens_seen": 33689640,
      "step": 1588,
      "time_per_iteration": 2.548868417739868
    },
    {
      "auxiliary_loss_clip": 0.0120359,
      "auxiliary_loss_mlp": 0.01034373,
      "balance_loss_clip": 1.06087148,
      "balance_loss_mlp": 1.0238055,
      "epoch": 0.19106595322551553,
      "flos": 21616320850560.0,
      "grad_norm": 2.9013520651165643,
      "language_loss": 0.93263906,
      "learning_rate": 3.7341243269031556e-06,
      "loss": 0.95501864,
      "num_input_tokens_seen": 33708630,
      "step": 1589,
      "time_per_iteration": 2.487123966217041
    },
    {
      "auxiliary_loss_clip": 0.01179851,
      "auxiliary_loss_mlp": 0.01033735,
      "balance_loss_clip": 1.05791235,
      "balance_loss_mlp": 1.02462196,
      "epoch": 0.19118619611615464,
      "flos": 29896890059520.0,
      "grad_norm": 1.742265582404153,
      "language_loss": 0.77464372,
      "learning_rate": 3.7337361117332275e-06,
      "loss": 0.79677957,
      "num_input_tokens_seen": 33730370,
      "step": 1590,
      "time_per_iteration": 2.59507155418396
    },
    {
      "auxiliary_loss_clip": 0.01172771,
      "auxiliary_loss_mlp": 0.01029884,
      "balance_loss_clip": 1.05422902,
      "balance_loss_mlp": 1.02107489,
      "epoch": 0.19130643900679373,
      "flos": 17273428093440.0,
      "grad_norm": 1.987391806197037,
      "language_loss": 0.77362764,
      "learning_rate": 3.7333476335569087e-06,
      "loss": 0.79565418,
      "num_input_tokens_seen": 33748370,
      "step": 1591,
      "time_per_iteration": 2.5393762588500977
    },
    {
      "auxiliary_loss_clip": 0.01188525,
      "auxiliary_loss_mlp": 0.01035647,
      "balance_loss_clip": 1.05983901,
      "balance_loss_mlp": 1.02528167,
      "epoch": 0.1914266818974328,
      "flos": 24826339584000.0,
      "grad_norm": 2.221837379579082,
      "language_loss": 0.67002207,
      "learning_rate": 3.7329588924331325e-06,
      "loss": 0.69226372,
      "num_input_tokens_seen": 33769575,
      "step": 1592,
      "time_per_iteration": 2.57541561126709
    },
    {
      "auxiliary_loss_clip": 0.01164584,
      "auxiliary_loss_mlp": 0.01035212,
      "balance_loss_clip": 1.05271208,
      "balance_loss_mlp": 1.02580631,
      "epoch": 0.1915469247880719,
      "flos": 18952467390720.0,
      "grad_norm": 1.6975999779184505,
      "language_loss": 0.82252264,
      "learning_rate": 3.732569888420871e-06,
      "loss": 0.84452057,
      "num_input_tokens_seen": 33789110,
      "step": 1593,
      "time_per_iteration": 3.3779399394989014
    },
    {
      "auxiliary_loss_clip": 0.01219186,
      "auxiliary_loss_mlp": 0.01034044,
      "balance_loss_clip": 1.06030118,
      "balance_loss_mlp": 1.02360141,
      "epoch": 0.191667167678711,
      "flos": 21032952065280.0,
      "grad_norm": 3.3382731143799758,
      "language_loss": 0.82523894,
      "learning_rate": 3.732180621579134e-06,
      "loss": 0.84777123,
      "num_input_tokens_seen": 33808325,
      "step": 1594,
      "time_per_iteration": 2.4570164680480957
    },
    {
      "auxiliary_loss_clip": 0.01184581,
      "auxiliary_loss_mlp": 0.01035344,
      "balance_loss_clip": 1.05961871,
      "balance_loss_mlp": 1.02532482,
      "epoch": 0.1917874105693501,
      "flos": 34237663914240.0,
      "grad_norm": 4.589277036258129,
      "language_loss": 0.81210053,
      "learning_rate": 3.7317910919669745e-06,
      "loss": 0.8342998,
      "num_input_tokens_seen": 33829520,
      "step": 1595,
      "time_per_iteration": 2.6728694438934326
    },
    {
      "auxiliary_loss_clip": 0.01202828,
      "auxiliary_loss_mlp": 0.01040124,
      "balance_loss_clip": 1.06096947,
      "balance_loss_mlp": 1.03000343,
      "epoch": 0.19190765345998917,
      "flos": 23550613171200.0,
      "grad_norm": 2.118972675721372,
      "language_loss": 0.76637793,
      "learning_rate": 3.7314012996434826e-06,
      "loss": 0.78880751,
      "num_input_tokens_seen": 33848250,
      "step": 1596,
      "time_per_iteration": 2.498216390609741
    },
    {
      "auxiliary_loss_clip": 0.01190154,
      "auxiliary_loss_mlp": 0.01030894,
      "balance_loss_clip": 1.05953336,
      "balance_loss_mlp": 1.02100563,
      "epoch": 0.19202789635062828,
      "flos": 19861330245120.0,
      "grad_norm": 1.9762185883592378,
      "language_loss": 0.81549913,
      "learning_rate": 3.7310112446677907e-06,
      "loss": 0.83770967,
      "num_input_tokens_seen": 33866160,
      "step": 1597,
      "time_per_iteration": 2.504812479019165
    },
    {
      "auxiliary_loss_clip": 0.01222459,
      "auxiliary_loss_mlp": 0.01030574,
      "balance_loss_clip": 1.06431508,
      "balance_loss_mlp": 1.020805,
      "epoch": 0.19214813924126736,
      "flos": 20922957642240.0,
      "grad_norm": 3.6050880450292677,
      "language_loss": 0.69361144,
      "learning_rate": 3.7306209270990695e-06,
      "loss": 0.71614176,
      "num_input_tokens_seen": 33884165,
      "step": 1598,
      "time_per_iteration": 2.488412857055664
    },
    {
      "auxiliary_loss_clip": 0.01188485,
      "auxiliary_loss_mlp": 0.01040707,
      "balance_loss_clip": 1.05908775,
      "balance_loss_mlp": 1.03145051,
      "epoch": 0.19226838213190645,
      "flos": 26359725231360.0,
      "grad_norm": 1.9539046762315475,
      "language_loss": 0.86761081,
      "learning_rate": 3.7302303469965292e-06,
      "loss": 0.88990283,
      "num_input_tokens_seen": 33903705,
      "step": 1599,
      "time_per_iteration": 3.3601696491241455
    },
    {
      "auxiliary_loss_clip": 0.0120324,
      "auxiliary_loss_mlp": 0.01042113,
      "balance_loss_clip": 1.06108558,
      "balance_loss_mlp": 1.0325942,
      "epoch": 0.19238862502254553,
      "flos": 20850525866880.0,
      "grad_norm": 5.035027889308814,
      "language_loss": 0.709975,
      "learning_rate": 3.7298395044194206e-06,
      "loss": 0.73242855,
      "num_input_tokens_seen": 33922515,
      "step": 1600,
      "time_per_iteration": 3.328348159790039
    },
    {
      "auxiliary_loss_clip": 0.01222932,
      "auxiliary_loss_mlp": 0.01034773,
      "balance_loss_clip": 1.0658021,
      "balance_loss_mlp": 1.0252068,
      "epoch": 0.19250886791318464,
      "flos": 21726063878400.0,
      "grad_norm": 2.5307715227204817,
      "language_loss": 0.94272757,
      "learning_rate": 3.7294483994270356e-06,
      "loss": 0.96530461,
      "num_input_tokens_seen": 33940840,
      "step": 1601,
      "time_per_iteration": 3.2153122425079346
    },
    {
      "auxiliary_loss_clip": 0.01146556,
      "auxiliary_loss_mlp": 0.0103168,
      "balance_loss_clip": 1.05223823,
      "balance_loss_mlp": 1.02334726,
      "epoch": 0.19262911080382372,
      "flos": 23367827836800.0,
      "grad_norm": 2.1602791193047346,
      "language_loss": 0.78748274,
      "learning_rate": 3.7290570320787033e-06,
      "loss": 0.80926508,
      "num_input_tokens_seen": 33960420,
      "step": 1602,
      "time_per_iteration": 2.5850577354431152
    },
    {
      "auxiliary_loss_clip": 0.01203137,
      "auxiliary_loss_mlp": 0.01031909,
      "balance_loss_clip": 1.06198585,
      "balance_loss_mlp": 1.02238417,
      "epoch": 0.1927493536944628,
      "flos": 21943502858880.0,
      "grad_norm": 2.260678450619943,
      "language_loss": 0.71356058,
      "learning_rate": 3.728665402433793e-06,
      "loss": 0.73591101,
      "num_input_tokens_seen": 33978990,
      "step": 1603,
      "time_per_iteration": 2.5086870193481445
    },
    {
      "auxiliary_loss_clip": 0.01192163,
      "auxiliary_loss_mlp": 0.01034638,
      "balance_loss_clip": 1.06267428,
      "balance_loss_mlp": 1.02552414,
      "epoch": 0.19286959658510192,
      "flos": 16545590807040.0,
      "grad_norm": 2.3352052621092008,
      "language_loss": 0.86060452,
      "learning_rate": 3.7282735105517164e-06,
      "loss": 0.88287258,
      "num_input_tokens_seen": 33997115,
      "step": 1604,
      "time_per_iteration": 2.497164249420166
    },
    {
      "auxiliary_loss_clip": 0.01165361,
      "auxiliary_loss_mlp": 0.01036685,
      "balance_loss_clip": 1.05348873,
      "balance_loss_mlp": 1.02702296,
      "epoch": 0.192989839475741,
      "flos": 21616967295360.0,
      "grad_norm": 1.9861077245212593,
      "language_loss": 0.67265725,
      "learning_rate": 3.727881356491922e-06,
      "loss": 0.69467771,
      "num_input_tokens_seen": 34015525,
      "step": 1605,
      "time_per_iteration": 2.606354236602783
    },
    {
      "auxiliary_loss_clip": 0.01219973,
      "auxiliary_loss_mlp": 0.01036542,
      "balance_loss_clip": 1.06457198,
      "balance_loss_mlp": 1.02789354,
      "epoch": 0.19311008236638008,
      "flos": 19281516906240.0,
      "grad_norm": 1.8079207738179592,
      "language_loss": 0.75772434,
      "learning_rate": 3.7274889403139002e-06,
      "loss": 0.78028947,
      "num_input_tokens_seen": 34033150,
      "step": 1606,
      "time_per_iteration": 2.4447648525238037
    },
    {
      "auxiliary_loss_clip": 0.01157561,
      "auxiliary_loss_mlp": 0.01032947,
      "balance_loss_clip": 1.05782604,
      "balance_loss_mlp": 1.0236131,
      "epoch": 0.1932303252570192,
      "flos": 28652369587200.0,
      "grad_norm": 2.4854466326576645,
      "language_loss": 0.78287339,
      "learning_rate": 3.727096262077179e-06,
      "loss": 0.80477846,
      "num_input_tokens_seen": 34052145,
      "step": 1607,
      "time_per_iteration": 2.6555755138397217
    },
    {
      "auxiliary_loss_clip": 0.01205283,
      "auxiliary_loss_mlp": 0.01031442,
      "balance_loss_clip": 1.06171179,
      "balance_loss_mlp": 1.02213192,
      "epoch": 0.19335056814765827,
      "flos": 18368990864640.0,
      "grad_norm": 1.712899263160362,
      "language_loss": 0.85260642,
      "learning_rate": 3.7267033218413285e-06,
      "loss": 0.87497365,
      "num_input_tokens_seen": 34069940,
      "step": 1608,
      "time_per_iteration": 2.485337734222412
    },
    {
      "auxiliary_loss_clip": 0.01144435,
      "auxiliary_loss_mlp": 0.01037978,
      "balance_loss_clip": 1.04949069,
      "balance_loss_mlp": 1.02715993,
      "epoch": 0.19347081103829736,
      "flos": 13260877741440.0,
      "grad_norm": 2.3274551957348333,
      "language_loss": 0.81559098,
      "learning_rate": 3.726310119665957e-06,
      "loss": 0.83741516,
      "num_input_tokens_seen": 34086275,
      "step": 1609,
      "time_per_iteration": 2.5958425998687744
    },
    {
      "auxiliary_loss_clip": 0.01203492,
      "auxiliary_loss_mlp": 0.01032228,
      "balance_loss_clip": 1.05945301,
      "balance_loss_mlp": 1.02264357,
      "epoch": 0.19359105392893644,
      "flos": 20300122788480.0,
      "grad_norm": 6.797871910514251,
      "language_loss": 0.8529979,
      "learning_rate": 3.725916655610713e-06,
      "loss": 0.87535512,
      "num_input_tokens_seen": 34105605,
      "step": 1610,
      "time_per_iteration": 2.50962233543396
    },
    {
      "auxiliary_loss_clip": 0.01179336,
      "auxiliary_loss_mlp": 0.0103085,
      "balance_loss_clip": 1.05387163,
      "balance_loss_mlp": 1.02072346,
      "epoch": 0.19371129681957555,
      "flos": 20484596062080.0,
      "grad_norm": 2.271893400566545,
      "language_loss": 0.756423,
      "learning_rate": 3.725522929735284e-06,
      "loss": 0.77852488,
      "num_input_tokens_seen": 34122540,
      "step": 1611,
      "time_per_iteration": 2.4995856285095215
    },
    {
      "auxiliary_loss_clip": 0.01195033,
      "auxiliary_loss_mlp": 0.0103125,
      "balance_loss_clip": 1.05754519,
      "balance_loss_mlp": 1.02136147,
      "epoch": 0.19383153971021463,
      "flos": 30445497457920.0,
      "grad_norm": 3.9646823815996592,
      "language_loss": 0.74336874,
      "learning_rate": 3.725128942099399e-06,
      "loss": 0.76563156,
      "num_input_tokens_seen": 34142940,
      "step": 1612,
      "time_per_iteration": 2.5952932834625244
    },
    {
      "auxiliary_loss_clip": 0.01176847,
      "auxiliary_loss_mlp": 0.01033384,
      "balance_loss_clip": 1.0547688,
      "balance_loss_mlp": 1.02373433,
      "epoch": 0.19395178260085372,
      "flos": 24569937325440.0,
      "grad_norm": 3.5108690578685433,
      "language_loss": 0.79512978,
      "learning_rate": 3.7247346927628245e-06,
      "loss": 0.81723213,
      "num_input_tokens_seen": 34162875,
      "step": 1613,
      "time_per_iteration": 2.5632119178771973
    },
    {
      "auxiliary_loss_clip": 0.01183374,
      "auxiliary_loss_mlp": 0.00765144,
      "balance_loss_clip": 1.05575013,
      "balance_loss_mlp": 1.00093484,
      "epoch": 0.19407202549149283,
      "flos": 28950608211840.0,
      "grad_norm": 1.8605008041969282,
      "language_loss": 0.79384148,
      "learning_rate": 3.7243401817853694e-06,
      "loss": 0.81332666,
      "num_input_tokens_seen": 34183565,
      "step": 1614,
      "time_per_iteration": 2.5685200691223145
    },
    {
      "auxiliary_loss_clip": 0.01195809,
      "auxiliary_loss_mlp": 0.01033267,
      "balance_loss_clip": 1.05703545,
      "balance_loss_mlp": 1.02399826,
      "epoch": 0.1941922683821319,
      "flos": 18004497603840.0,
      "grad_norm": 1.9572675795824275,
      "language_loss": 0.71726632,
      "learning_rate": 3.723945409226879e-06,
      "loss": 0.73955715,
      "num_input_tokens_seen": 34202055,
      "step": 1615,
      "time_per_iteration": 2.445333242416382
    },
    {
      "auxiliary_loss_clip": 0.01202986,
      "auxiliary_loss_mlp": 0.01042162,
      "balance_loss_clip": 1.05983257,
      "balance_loss_mlp": 1.03220236,
      "epoch": 0.194312511272771,
      "flos": 9720337034880.0,
      "grad_norm": 2.569418678549781,
      "language_loss": 0.80127764,
      "learning_rate": 3.723550375147241e-06,
      "loss": 0.82372916,
      "num_input_tokens_seen": 34216830,
      "step": 1616,
      "time_per_iteration": 2.473280668258667
    },
    {
      "auxiliary_loss_clip": 0.01159501,
      "auxiliary_loss_mlp": 0.0103576,
      "balance_loss_clip": 1.05108607,
      "balance_loss_mlp": 1.02555561,
      "epoch": 0.19443275416341008,
      "flos": 27016208150400.0,
      "grad_norm": 1.6555038259686654,
      "language_loss": 0.79878938,
      "learning_rate": 3.7231550796063816e-06,
      "loss": 0.82074201,
      "num_input_tokens_seen": 34236840,
      "step": 1617,
      "time_per_iteration": 2.594310760498047
    },
    {
      "auxiliary_loss_clip": 0.01195926,
      "auxiliary_loss_mlp": 0.01038931,
      "balance_loss_clip": 1.06091058,
      "balance_loss_mlp": 1.02860737,
      "epoch": 0.1945529970540492,
      "flos": 15846625077120.0,
      "grad_norm": 2.2155233497930267,
      "language_loss": 0.64881492,
      "learning_rate": 3.722759522664266e-06,
      "loss": 0.6711635,
      "num_input_tokens_seen": 34254140,
      "step": 1618,
      "time_per_iteration": 2.5026543140411377
    },
    {
      "auxiliary_loss_clip": 0.01159288,
      "auxiliary_loss_mlp": 0.01029697,
      "balance_loss_clip": 1.05464172,
      "balance_loss_mlp": 1.01980817,
      "epoch": 0.19467323994468827,
      "flos": 19314985403520.0,
      "grad_norm": 2.094113840125367,
      "language_loss": 0.81364161,
      "learning_rate": 3.7223637043809016e-06,
      "loss": 0.83553147,
      "num_input_tokens_seen": 34273120,
      "step": 1619,
      "time_per_iteration": 3.3568055629730225
    },
    {
      "auxiliary_loss_clip": 0.01177042,
      "auxiliary_loss_mlp": 0.01039793,
      "balance_loss_clip": 1.05788827,
      "balance_loss_mlp": 1.03059053,
      "epoch": 0.19479348283532735,
      "flos": 24133227770880.0,
      "grad_norm": 1.7939700519857602,
      "language_loss": 0.86837673,
      "learning_rate": 3.7219676248163322e-06,
      "loss": 0.89054513,
      "num_input_tokens_seen": 34290285,
      "step": 1620,
      "time_per_iteration": 2.550503730773926
    },
    {
      "auxiliary_loss_clip": 0.01209746,
      "auxiliary_loss_mlp": 0.01033836,
      "balance_loss_clip": 1.06290007,
      "balance_loss_mlp": 1.02362609,
      "epoch": 0.19491372572596646,
      "flos": 25775638174080.0,
      "grad_norm": 1.9266301918596251,
      "language_loss": 0.93155599,
      "learning_rate": 3.721571284030643e-06,
      "loss": 0.95399183,
      "num_input_tokens_seen": 34310095,
      "step": 1621,
      "time_per_iteration": 2.527374744415283
    },
    {
      "auxiliary_loss_clip": 0.01207716,
      "auxiliary_loss_mlp": 0.01026182,
      "balance_loss_clip": 1.06077635,
      "balance_loss_mlp": 1.01655555,
      "epoch": 0.19503396861660555,
      "flos": 19645220067840.0,
      "grad_norm": 2.611281306459111,
      "language_loss": 0.79294729,
      "learning_rate": 3.7211746820839587e-06,
      "loss": 0.81528628,
      "num_input_tokens_seen": 34327190,
      "step": 1622,
      "time_per_iteration": 2.482006788253784
    },
    {
      "auxiliary_loss_clip": 0.01109937,
      "auxiliary_loss_mlp": 0.01030952,
      "balance_loss_clip": 1.04737592,
      "balance_loss_mlp": 1.02103972,
      "epoch": 0.19515421150724463,
      "flos": 21033023892480.0,
      "grad_norm": 1.6229701700855272,
      "language_loss": 0.80526686,
      "learning_rate": 3.7207778190364437e-06,
      "loss": 0.82667577,
      "num_input_tokens_seen": 34345615,
      "step": 1623,
      "time_per_iteration": 2.632573366165161
    },
    {
      "auxiliary_loss_clip": 0.01130734,
      "auxiliary_loss_mlp": 0.01033525,
      "balance_loss_clip": 1.04932904,
      "balance_loss_mlp": 1.02382112,
      "epoch": 0.1952744543978837,
      "flos": 32961255143040.0,
      "grad_norm": 1.6046977402444162,
      "language_loss": 0.73752093,
      "learning_rate": 3.720380694948302e-06,
      "loss": 0.7591635,
      "num_input_tokens_seen": 34368500,
      "step": 1624,
      "time_per_iteration": 2.6935505867004395
    },
    {
      "auxiliary_loss_clip": 0.01083639,
      "auxiliary_loss_mlp": 0.01007573,
      "balance_loss_clip": 1.03001809,
      "balance_loss_mlp": 1.00430667,
      "epoch": 0.19539469728852282,
      "flos": 64044312030720.0,
      "grad_norm": 1.0340483389483592,
      "language_loss": 0.71240377,
      "learning_rate": 3.719983309879777e-06,
      "loss": 0.73331594,
      "num_input_tokens_seen": 34428280,
      "step": 1625,
      "time_per_iteration": 3.9505069255828857
    },
    {
      "auxiliary_loss_clip": 0.01164246,
      "auxiliary_loss_mlp": 0.0104103,
      "balance_loss_clip": 1.05362272,
      "balance_loss_mlp": 1.03160608,
      "epoch": 0.1955149401791619,
      "flos": 13370908078080.0,
      "grad_norm": 1.9448949063990069,
      "language_loss": 0.7744208,
      "learning_rate": 3.719585663891151e-06,
      "loss": 0.7964735,
      "num_input_tokens_seen": 34445815,
      "step": 1626,
      "time_per_iteration": 3.36405873298645
    },
    {
      "auxiliary_loss_clip": 0.0115292,
      "auxiliary_loss_mlp": 0.0104182,
      "balance_loss_clip": 1.05643702,
      "balance_loss_mlp": 1.03162766,
      "epoch": 0.195635183069801,
      "flos": 18728887184640.0,
      "grad_norm": 2.104964249852706,
      "language_loss": 0.78847951,
      "learning_rate": 3.719187757042747e-06,
      "loss": 0.81042689,
      "num_input_tokens_seen": 34463635,
      "step": 1627,
      "time_per_iteration": 3.3201677799224854
    },
    {
      "auxiliary_loss_clip": 0.01104896,
      "auxiliary_loss_mlp": 0.01003333,
      "balance_loss_clip": 1.0356195,
      "balance_loss_mlp": 1.00038815,
      "epoch": 0.1957554259604401,
      "flos": 69313952615040.0,
      "grad_norm": 0.7295797716334137,
      "language_loss": 0.54937702,
      "learning_rate": 3.7187895893949275e-06,
      "loss": 0.57045931,
      "num_input_tokens_seen": 34530105,
      "step": 1628,
      "time_per_iteration": 3.1895103454589844
    },
    {
      "auxiliary_loss_clip": 0.01145832,
      "auxiliary_loss_mlp": 0.01025827,
      "balance_loss_clip": 1.05101264,
      "balance_loss_mlp": 1.01543212,
      "epoch": 0.19587566885107918,
      "flos": 21069257736960.0,
      "grad_norm": 2.2061824573171993,
      "language_loss": 0.76364571,
      "learning_rate": 3.7183911610080937e-06,
      "loss": 0.78536224,
      "num_input_tokens_seen": 34546970,
      "step": 1629,
      "time_per_iteration": 2.5896711349487305
    },
    {
      "auxiliary_loss_clip": 0.01178498,
      "auxiliary_loss_mlp": 0.01042951,
      "balance_loss_clip": 1.05810952,
      "balance_loss_mlp": 1.03204989,
      "epoch": 0.19599591174171827,
      "flos": 22194661731840.0,
      "grad_norm": 2.9276473469108653,
      "language_loss": 0.74557209,
      "learning_rate": 3.7179924719426872e-06,
      "loss": 0.76778662,
      "num_input_tokens_seen": 34564865,
      "step": 1630,
      "time_per_iteration": 2.5846400260925293
    },
    {
      "auxiliary_loss_clip": 0.01208447,
      "auxiliary_loss_mlp": 0.01040735,
      "balance_loss_clip": 1.06222546,
      "balance_loss_mlp": 1.03079903,
      "epoch": 0.19611615463235738,
      "flos": 23768375374080.0,
      "grad_norm": 2.511109352186563,
      "language_loss": 0.76104778,
      "learning_rate": 3.7175935222591885e-06,
      "loss": 0.78353965,
      "num_input_tokens_seen": 34584165,
      "step": 1631,
      "time_per_iteration": 2.536773204803467
    },
    {
      "auxiliary_loss_clip": 0.0119398,
      "auxiliary_loss_mlp": 0.01034552,
      "balance_loss_clip": 1.06454563,
      "balance_loss_mlp": 1.02465725,
      "epoch": 0.19623639752299646,
      "flos": 28618218731520.0,
      "grad_norm": 1.8385144209183029,
      "language_loss": 0.74802679,
      "learning_rate": 3.717194312018118e-06,
      "loss": 0.77031207,
      "num_input_tokens_seen": 34603150,
      "step": 1632,
      "time_per_iteration": 2.6672165393829346
    },
    {
      "auxiliary_loss_clip": 0.01203381,
      "auxiliary_loss_mlp": 0.01037981,
      "balance_loss_clip": 1.0590229,
      "balance_loss_mlp": 1.02780676,
      "epoch": 0.19635664041363554,
      "flos": 21032700670080.0,
      "grad_norm": 2.1660042650376092,
      "language_loss": 0.76431143,
      "learning_rate": 3.716794841280036e-06,
      "loss": 0.78672504,
      "num_input_tokens_seen": 34621855,
      "step": 1633,
      "time_per_iteration": 2.50923752784729
    },
    {
      "auxiliary_loss_clip": 0.0121039,
      "auxiliary_loss_mlp": 0.01038689,
      "balance_loss_clip": 1.06045151,
      "balance_loss_mlp": 1.02867508,
      "epoch": 0.19647688330427462,
      "flos": 18879748306560.0,
      "grad_norm": 2.0475445430633057,
      "language_loss": 0.77557826,
      "learning_rate": 3.7163951101055407e-06,
      "loss": 0.798069,
      "num_input_tokens_seen": 34639915,
      "step": 1634,
      "time_per_iteration": 2.4987168312072754
    },
    {
      "auxiliary_loss_clip": 0.01188349,
      "auxiliary_loss_mlp": 0.01036922,
      "balance_loss_clip": 1.06042349,
      "balance_loss_mlp": 1.02673519,
      "epoch": 0.19659712619491373,
      "flos": 24242503921920.0,
      "grad_norm": 1.9019399445160583,
      "language_loss": 0.78896469,
      "learning_rate": 3.715995118555273e-06,
      "loss": 0.81121737,
      "num_input_tokens_seen": 34659890,
      "step": 1635,
      "time_per_iteration": 2.5771420001983643
    },
    {
      "auxiliary_loss_clip": 0.01154421,
      "auxiliary_loss_mlp": 0.01039155,
      "balance_loss_clip": 1.05317211,
      "balance_loss_mlp": 1.02812254,
      "epoch": 0.19671736908555282,
      "flos": 24717422568960.0,
      "grad_norm": 2.4529292534742893,
      "language_loss": 0.85550624,
      "learning_rate": 3.71559486668991e-06,
      "loss": 0.877442,
      "num_input_tokens_seen": 34678750,
      "step": 1636,
      "time_per_iteration": 2.63238263130188
    },
    {
      "auxiliary_loss_clip": 0.01211982,
      "auxiliary_loss_mlp": 0.00765143,
      "balance_loss_clip": 1.06431162,
      "balance_loss_mlp": 1.00102079,
      "epoch": 0.1968376119761919,
      "flos": 23842279607040.0,
      "grad_norm": 1.6884853463991911,
      "language_loss": 0.7729736,
      "learning_rate": 3.715194354570169e-06,
      "loss": 0.79274487,
      "num_input_tokens_seen": 34698755,
      "step": 1637,
      "time_per_iteration": 2.5613176822662354
    },
    {
      "auxiliary_loss_clip": 0.01206988,
      "auxiliary_loss_mlp": 0.01042607,
      "balance_loss_clip": 1.06504273,
      "balance_loss_mlp": 1.03295088,
      "epoch": 0.196957854866831,
      "flos": 18113917409280.0,
      "grad_norm": 1.9691690305393585,
      "language_loss": 0.83363259,
      "learning_rate": 3.714793582256809e-06,
      "loss": 0.85612857,
      "num_input_tokens_seen": 34715820,
      "step": 1638,
      "time_per_iteration": 2.583327054977417
    },
    {
      "auxiliary_loss_clip": 0.01218352,
      "auxiliary_loss_mlp": 0.01036213,
      "balance_loss_clip": 1.06208348,
      "balance_loss_mlp": 1.02621758,
      "epoch": 0.1970780977574701,
      "flos": 21653129312640.0,
      "grad_norm": 2.6975099761931673,
      "language_loss": 0.85075057,
      "learning_rate": 3.7143925498106253e-06,
      "loss": 0.87329626,
      "num_input_tokens_seen": 34734360,
      "step": 1639,
      "time_per_iteration": 2.4852707386016846
    },
    {
      "auxiliary_loss_clip": 0.01189501,
      "auxiliary_loss_mlp": 0.01035805,
      "balance_loss_clip": 1.05576229,
      "balance_loss_mlp": 1.02508259,
      "epoch": 0.19719834064810918,
      "flos": 20811813984000.0,
      "grad_norm": 1.9507284402425875,
      "language_loss": 0.79450566,
      "learning_rate": 3.7139912572924558e-06,
      "loss": 0.81675875,
      "num_input_tokens_seen": 34753390,
      "step": 1640,
      "time_per_iteration": 2.5805928707122803
    },
    {
      "auxiliary_loss_clip": 0.01201007,
      "auxiliary_loss_mlp": 0.01038118,
      "balance_loss_clip": 1.05786896,
      "balance_loss_mlp": 1.02830696,
      "epoch": 0.19731858353874826,
      "flos": 23434800744960.0,
      "grad_norm": 4.183664415689264,
      "language_loss": 0.79915917,
      "learning_rate": 3.7135897047631744e-06,
      "loss": 0.82155037,
      "num_input_tokens_seen": 34771275,
      "step": 1641,
      "time_per_iteration": 2.5684657096862793
    },
    {
      "auxiliary_loss_clip": 0.01192026,
      "auxiliary_loss_mlp": 0.01036715,
      "balance_loss_clip": 1.06010377,
      "balance_loss_mlp": 1.02636778,
      "epoch": 0.19743882642938737,
      "flos": 23988184652160.0,
      "grad_norm": 2.4634363500286907,
      "language_loss": 0.76119924,
      "learning_rate": 3.713187892283698e-06,
      "loss": 0.7834866,
      "num_input_tokens_seen": 34790885,
      "step": 1642,
      "time_per_iteration": 2.5714285373687744
    },
    {
      "auxiliary_loss_clip": 0.01158354,
      "auxiliary_loss_mlp": 0.01039186,
      "balance_loss_clip": 1.05249333,
      "balance_loss_mlp": 1.02901709,
      "epoch": 0.19755906932002645,
      "flos": 15004340081280.0,
      "grad_norm": 2.459635447959844,
      "language_loss": 0.87307298,
      "learning_rate": 3.71278581991498e-06,
      "loss": 0.89504838,
      "num_input_tokens_seen": 34806745,
      "step": 1643,
      "time_per_iteration": 2.5991883277893066
    },
    {
      "auxiliary_loss_clip": 0.0118069,
      "auxiliary_loss_mlp": 0.00766174,
      "balance_loss_clip": 1.06415129,
      "balance_loss_mlp": 1.00091124,
      "epoch": 0.19767931221066554,
      "flos": 19494466686720.0,
      "grad_norm": 1.7661277831995152,
      "language_loss": 0.79050279,
      "learning_rate": 3.712383487718015e-06,
      "loss": 0.80997145,
      "num_input_tokens_seen": 34824985,
      "step": 1644,
      "time_per_iteration": 2.565838098526001
    },
    {
      "auxiliary_loss_clip": 0.01141436,
      "auxiliary_loss_mlp": 0.01034648,
      "balance_loss_clip": 1.05368018,
      "balance_loss_mlp": 1.02492642,
      "epoch": 0.19779955510130465,
      "flos": 25737895958400.0,
      "grad_norm": 1.9384558475600515,
      "language_loss": 0.86899418,
      "learning_rate": 3.7119808957538365e-06,
      "loss": 0.89075506,
      "num_input_tokens_seen": 34843980,
      "step": 1645,
      "time_per_iteration": 2.6641485691070557
    },
    {
      "auxiliary_loss_clip": 0.01185913,
      "auxiliary_loss_mlp": 0.01035776,
      "balance_loss_clip": 1.05651164,
      "balance_loss_mlp": 1.02508879,
      "epoch": 0.19791979799194373,
      "flos": 20777699041920.0,
      "grad_norm": 8.451098249241808,
      "language_loss": 0.80159289,
      "learning_rate": 3.711578044083517e-06,
      "loss": 0.8238098,
      "num_input_tokens_seen": 34860780,
      "step": 1646,
      "time_per_iteration": 3.3448476791381836
    },
    {
      "auxiliary_loss_clip": 0.01192336,
      "auxiliary_loss_mlp": 0.01041315,
      "balance_loss_clip": 1.06008196,
      "balance_loss_mlp": 1.03087234,
      "epoch": 0.1980400408825828,
      "flos": 25589010084480.0,
      "grad_norm": 1.9040656801999716,
      "language_loss": 0.74541759,
      "learning_rate": 3.7111749327681698e-06,
      "loss": 0.76775408,
      "num_input_tokens_seen": 34880815,
      "step": 1647,
      "time_per_iteration": 2.5679314136505127
    },
    {
      "auxiliary_loss_clip": 0.0121259,
      "auxiliary_loss_mlp": 0.01030518,
      "balance_loss_clip": 1.06656599,
      "balance_loss_mlp": 1.02118969,
      "epoch": 0.1981602837732219,
      "flos": 23513840622720.0,
      "grad_norm": 3.0242349250946368,
      "language_loss": 0.86099494,
      "learning_rate": 3.7107715618689455e-06,
      "loss": 0.88342601,
      "num_input_tokens_seen": 34899790,
      "step": 1648,
      "time_per_iteration": 2.5200653076171875
    },
    {
      "auxiliary_loss_clip": 0.01202204,
      "auxiliary_loss_mlp": 0.01033351,
      "balance_loss_clip": 1.0612092,
      "balance_loss_mlp": 1.02311718,
      "epoch": 0.198280526663861,
      "flos": 23185365724800.0,
      "grad_norm": 1.5420652078259156,
      "language_loss": 0.83605528,
      "learning_rate": 3.710367931447035e-06,
      "loss": 0.85841078,
      "num_input_tokens_seen": 34921570,
      "step": 1649,
      "time_per_iteration": 2.5283823013305664
    },
    {
      "auxiliary_loss_clip": 0.01214658,
      "auxiliary_loss_mlp": 0.01041979,
      "balance_loss_clip": 1.0640049,
      "balance_loss_mlp": 1.03121471,
      "epoch": 0.1984007695545001,
      "flos": 21689470897920.0,
      "grad_norm": 2.1688839824402373,
      "language_loss": 0.86143672,
      "learning_rate": 3.70996404156367e-06,
      "loss": 0.88400304,
      "num_input_tokens_seen": 34941205,
      "step": 1650,
      "time_per_iteration": 2.5186758041381836
    },
    {
      "auxiliary_loss_clip": 0.01152221,
      "auxiliary_loss_mlp": 0.01038833,
      "balance_loss_clip": 1.05473626,
      "balance_loss_mlp": 1.02929068,
      "epoch": 0.19852101244513917,
      "flos": 36064008887040.0,
      "grad_norm": 1.6877518690527404,
      "language_loss": 0.7288053,
      "learning_rate": 3.7095598922801187e-06,
      "loss": 0.75071585,
      "num_input_tokens_seen": 34963280,
      "step": 1651,
      "time_per_iteration": 2.7213032245635986
    },
    {
      "auxiliary_loss_clip": 0.01221892,
      "auxiliary_loss_mlp": 0.01036918,
      "balance_loss_clip": 1.06495404,
      "balance_loss_mlp": 1.02660072,
      "epoch": 0.19864125533577828,
      "flos": 23105894883840.0,
      "grad_norm": 2.998001063498881,
      "language_loss": 0.7636205,
      "learning_rate": 3.7091554836576914e-06,
      "loss": 0.78620857,
      "num_input_tokens_seen": 34979955,
      "step": 1652,
      "time_per_iteration": 3.3028557300567627
    },
    {
      "auxiliary_loss_clip": 0.01205783,
      "auxiliary_loss_mlp": 0.00765017,
      "balance_loss_clip": 1.0644052,
      "balance_loss_mlp": 1.00104976,
      "epoch": 0.19876149822641737,
      "flos": 24608505553920.0,
      "grad_norm": 1.7078364518663895,
      "language_loss": 0.82894939,
      "learning_rate": 3.708750815757736e-06,
      "loss": 0.84865743,
      "num_input_tokens_seen": 35000725,
      "step": 1653,
      "time_per_iteration": 4.1630539894104
    },
    {
      "auxiliary_loss_clip": 0.01208338,
      "auxiliary_loss_mlp": 0.01040195,
      "balance_loss_clip": 1.06394792,
      "balance_loss_mlp": 1.03007448,
      "epoch": 0.19888174111705645,
      "flos": 32196645308160.0,
      "grad_norm": 2.4355750053133414,
      "language_loss": 0.73480725,
      "learning_rate": 3.7083458886416407e-06,
      "loss": 0.75729263,
      "num_input_tokens_seen": 35019920,
      "step": 1654,
      "time_per_iteration": 2.5967397689819336
    },
    {
      "auxiliary_loss_clip": 0.01152681,
      "auxiliary_loss_mlp": 0.01034596,
      "balance_loss_clip": 1.05905902,
      "balance_loss_mlp": 1.02455306,
      "epoch": 0.19900198400769553,
      "flos": 24608469640320.0,
      "grad_norm": 2.2523856304074594,
      "language_loss": 0.87917298,
      "learning_rate": 3.707940702370832e-06,
      "loss": 0.90104574,
      "num_input_tokens_seen": 35040765,
      "step": 1655,
      "time_per_iteration": 2.7132151126861572
    },
    {
      "auxiliary_loss_clip": 0.01111372,
      "auxiliary_loss_mlp": 0.01003066,
      "balance_loss_clip": 1.03698659,
      "balance_loss_mlp": 1.00069416,
      "epoch": 0.19912222689833464,
      "flos": 67915805673600.0,
      "grad_norm": 0.7581153224962618,
      "language_loss": 0.58303392,
      "learning_rate": 3.707535257006777e-06,
      "loss": 0.60417831,
      "num_input_tokens_seen": 35106390,
      "step": 1656,
      "time_per_iteration": 3.175356149673462
    },
    {
      "auxiliary_loss_clip": 0.01193514,
      "auxiliary_loss_mlp": 0.01039223,
      "balance_loss_clip": 1.06086111,
      "balance_loss_mlp": 1.02853,
      "epoch": 0.19924246978897373,
      "flos": 15742340916480.0,
      "grad_norm": 2.4797533285055735,
      "language_loss": 0.88435018,
      "learning_rate": 3.707129552610981e-06,
      "loss": 0.90667754,
      "num_input_tokens_seen": 35125040,
      "step": 1657,
      "time_per_iteration": 2.524670362472534
    },
    {
      "auxiliary_loss_clip": 0.01187576,
      "auxiliary_loss_mlp": 0.01031936,
      "balance_loss_clip": 1.06300759,
      "balance_loss_mlp": 1.02218461,
      "epoch": 0.1993627126796128,
      "flos": 17566566986880.0,
      "grad_norm": 2.0764022080352387,
      "language_loss": 0.73549104,
      "learning_rate": 3.70672358924499e-06,
      "loss": 0.75768614,
      "num_input_tokens_seen": 35144280,
      "step": 1658,
      "time_per_iteration": 2.53963303565979
    },
    {
      "auxiliary_loss_clip": 0.01176565,
      "auxiliary_loss_mlp": 0.01037206,
      "balance_loss_clip": 1.06365061,
      "balance_loss_mlp": 1.0269897,
      "epoch": 0.19948295557025192,
      "flos": 40843826680320.0,
      "grad_norm": 2.197653805299495,
      "language_loss": 0.78508693,
      "learning_rate": 3.706317366970386e-06,
      "loss": 0.80722463,
      "num_input_tokens_seen": 35165280,
      "step": 1659,
      "time_per_iteration": 2.7581794261932373
    },
    {
      "auxiliary_loss_clip": 0.01223309,
      "auxiliary_loss_mlp": 0.00765974,
      "balance_loss_clip": 1.06246698,
      "balance_loss_mlp": 1.00098264,
      "epoch": 0.199603198460891,
      "flos": 25082418620160.0,
      "grad_norm": 6.106228511336119,
      "language_loss": 0.83624744,
      "learning_rate": 3.705910885848795e-06,
      "loss": 0.85614032,
      "num_input_tokens_seen": 35183655,
      "step": 1660,
      "time_per_iteration": 2.5417072772979736
    },
    {
      "auxiliary_loss_clip": 0.01207101,
      "auxiliary_loss_mlp": 0.01031039,
      "balance_loss_clip": 1.06376636,
      "balance_loss_mlp": 1.02113295,
      "epoch": 0.19972344135153008,
      "flos": 20084120352000.0,
      "grad_norm": 2.050175655351116,
      "language_loss": 0.8430661,
      "learning_rate": 3.705504145941879e-06,
      "loss": 0.86544752,
      "num_input_tokens_seen": 35201825,
      "step": 1661,
      "time_per_iteration": 2.493687868118286
    },
    {
      "auxiliary_loss_clip": 0.01220762,
      "auxiliary_loss_mlp": 0.01031462,
      "balance_loss_clip": 1.06453264,
      "balance_loss_mlp": 1.02156162,
      "epoch": 0.1998436842421692,
      "flos": 23727472761600.0,
      "grad_norm": 1.8797041062696531,
      "language_loss": 0.78811651,
      "learning_rate": 3.7050971473113403e-06,
      "loss": 0.81063873,
      "num_input_tokens_seen": 35221600,
      "step": 1662,
      "time_per_iteration": 2.49241304397583
    },
    {
      "auxiliary_loss_clip": 0.01201331,
      "auxiliary_loss_mlp": 0.00764981,
      "balance_loss_clip": 1.06013894,
      "balance_loss_mlp": 1.00094926,
      "epoch": 0.19996392713280828,
      "flos": 36102361633920.0,
      "grad_norm": 1.6599697666414834,
      "language_loss": 0.80064654,
      "learning_rate": 3.7046898900189196e-06,
      "loss": 0.82030964,
      "num_input_tokens_seen": 35245935,
      "step": 1663,
      "time_per_iteration": 2.6219301223754883
    },
    {
      "auxiliary_loss_clip": 0.01181284,
      "auxiliary_loss_mlp": 0.01040233,
      "balance_loss_clip": 1.06198764,
      "balance_loss_mlp": 1.03007054,
      "epoch": 0.20008417002344736,
      "flos": 23657662679040.0,
      "grad_norm": 1.6288642803221676,
      "language_loss": 0.82824743,
      "learning_rate": 3.704282374126398e-06,
      "loss": 0.85046256,
      "num_input_tokens_seen": 35265615,
      "step": 1664,
      "time_per_iteration": 2.5889265537261963
    },
    {
      "auxiliary_loss_clip": 0.0117494,
      "auxiliary_loss_mlp": 0.01032182,
      "balance_loss_clip": 1.05887675,
      "balance_loss_mlp": 1.02223396,
      "epoch": 0.20020441291408644,
      "flos": 21872076664320.0,
      "grad_norm": 1.6582016618941289,
      "language_loss": 0.87228185,
      "learning_rate": 3.7038745996955954e-06,
      "loss": 0.89435303,
      "num_input_tokens_seen": 35284960,
      "step": 1665,
      "time_per_iteration": 2.676917791366577
    },
    {
      "auxiliary_loss_clip": 0.01181586,
      "auxiliary_loss_mlp": 0.01035357,
      "balance_loss_clip": 1.05892849,
      "balance_loss_mlp": 1.02605915,
      "epoch": 0.20032465580472555,
      "flos": 23179691376000.0,
      "grad_norm": 8.364605105437795,
      "language_loss": 0.71882677,
      "learning_rate": 3.703466566788371e-06,
      "loss": 0.74099624,
      "num_input_tokens_seen": 35304090,
      "step": 1666,
      "time_per_iteration": 2.6527786254882812
    },
    {
      "auxiliary_loss_clip": 0.01185144,
      "auxiliary_loss_mlp": 0.01034719,
      "balance_loss_clip": 1.06267095,
      "balance_loss_mlp": 1.02418709,
      "epoch": 0.20044489869536464,
      "flos": 23873521461120.0,
      "grad_norm": 1.7802398091804892,
      "language_loss": 0.74326873,
      "learning_rate": 3.703058275466622e-06,
      "loss": 0.76546741,
      "num_input_tokens_seen": 35323325,
      "step": 1667,
      "time_per_iteration": 2.6319797039031982
    },
    {
      "auxiliary_loss_clip": 0.01190139,
      "auxiliary_loss_mlp": 0.01037229,
      "balance_loss_clip": 1.06007433,
      "balance_loss_mlp": 1.02771592,
      "epoch": 0.20056514158600372,
      "flos": 21945226711680.0,
      "grad_norm": 1.6500551770118452,
      "language_loss": 0.77622211,
      "learning_rate": 3.7026497257922877e-06,
      "loss": 0.79849571,
      "num_input_tokens_seen": 35343635,
      "step": 1668,
      "time_per_iteration": 2.636244535446167
    },
    {
      "auxiliary_loss_clip": 0.01156455,
      "auxiliary_loss_mlp": 0.01047609,
      "balance_loss_clip": 1.0545131,
      "balance_loss_mlp": 1.03779793,
      "epoch": 0.20068538447664283,
      "flos": 23879159896320.0,
      "grad_norm": 1.7287193903547278,
      "language_loss": 0.8523429,
      "learning_rate": 3.7022409178273436e-06,
      "loss": 0.87438351,
      "num_input_tokens_seen": 35364615,
      "step": 1669,
      "time_per_iteration": 2.7075541019439697
    },
    {
      "auxiliary_loss_clip": 0.0120001,
      "auxiliary_loss_mlp": 0.01029436,
      "balance_loss_clip": 1.06105685,
      "balance_loss_mlp": 1.02031028,
      "epoch": 0.2008056273672819,
      "flos": 18442823270400.0,
      "grad_norm": 1.744699902509725,
      "language_loss": 0.78678107,
      "learning_rate": 3.7018318516338054e-06,
      "loss": 0.80907547,
      "num_input_tokens_seen": 35383775,
      "step": 1670,
      "time_per_iteration": 2.5497567653656006
    },
    {
      "auxiliary_loss_clip": 0.01208669,
      "auxiliary_loss_mlp": 0.01028947,
      "balance_loss_clip": 1.062778,
      "balance_loss_mlp": 1.01976764,
      "epoch": 0.200925870257921,
      "flos": 23659530186240.0,
      "grad_norm": 2.87900845657148,
      "language_loss": 0.81513917,
      "learning_rate": 3.7014225272737284e-06,
      "loss": 0.83751535,
      "num_input_tokens_seen": 35403000,
      "step": 1671,
      "time_per_iteration": 2.586712121963501
    },
    {
      "auxiliary_loss_clip": 0.01198212,
      "auxiliary_loss_mlp": 0.01034641,
      "balance_loss_clip": 1.06055844,
      "balance_loss_mlp": 1.02484179,
      "epoch": 0.20104611314856008,
      "flos": 16217115909120.0,
      "grad_norm": 2.2723517604150048,
      "language_loss": 0.7415908,
      "learning_rate": 3.701012944809207e-06,
      "loss": 0.76391935,
      "num_input_tokens_seen": 35420115,
      "step": 1672,
      "time_per_iteration": 3.333911418914795
    },
    {
      "auxiliary_loss_clip": 0.01188707,
      "auxiliary_loss_mlp": 0.00764452,
      "balance_loss_clip": 1.06163788,
      "balance_loss_mlp": 1.00105047,
      "epoch": 0.2011663560391992,
      "flos": 21397373498880.0,
      "grad_norm": 2.3451975569181904,
      "language_loss": 0.79037344,
      "learning_rate": 3.700603104302374e-06,
      "loss": 0.80990499,
      "num_input_tokens_seen": 35439925,
      "step": 1673,
      "time_per_iteration": 2.618485689163208
    },
    {
      "auxiliary_loss_clip": 0.01070271,
      "auxiliary_loss_mlp": 0.01003165,
      "balance_loss_clip": 1.02871227,
      "balance_loss_mlp": 1.00064969,
      "epoch": 0.20128659892983827,
      "flos": 62229459409920.0,
      "grad_norm": 0.8881138279454771,
      "language_loss": 0.55909812,
      "learning_rate": 3.7001930058154027e-06,
      "loss": 0.57983243,
      "num_input_tokens_seen": 35504885,
      "step": 1674,
      "time_per_iteration": 3.2131967544555664
    },
    {
      "auxiliary_loss_clip": 0.01173404,
      "auxiliary_loss_mlp": 0.0103804,
      "balance_loss_clip": 1.05735135,
      "balance_loss_mlp": 1.02799642,
      "epoch": 0.20140684182047736,
      "flos": 28438737448320.0,
      "grad_norm": 2.282931835369254,
      "language_loss": 0.80218577,
      "learning_rate": 3.6997826494105037e-06,
      "loss": 0.82430023,
      "num_input_tokens_seen": 35525330,
      "step": 1675,
      "time_per_iteration": 2.6722702980041504
    },
    {
      "auxiliary_loss_clip": 0.01189185,
      "auxiliary_loss_mlp": 0.01029578,
      "balance_loss_clip": 1.06025648,
      "balance_loss_mlp": 1.02014899,
      "epoch": 0.20152708471111647,
      "flos": 28074064619520.0,
      "grad_norm": 2.0277036115078615,
      "language_loss": 0.69524276,
      "learning_rate": 3.6993720351499286e-06,
      "loss": 0.71743041,
      "num_input_tokens_seen": 35546455,
      "step": 1676,
      "time_per_iteration": 2.6323435306549072
    },
    {
      "auxiliary_loss_clip": 0.01183767,
      "auxiliary_loss_mlp": 0.01033742,
      "balance_loss_clip": 1.06196809,
      "balance_loss_mlp": 1.02442551,
      "epoch": 0.20164732760175555,
      "flos": 23549751244800.0,
      "grad_norm": 1.7119437772356263,
      "language_loss": 0.77048182,
      "learning_rate": 3.6989611630959666e-06,
      "loss": 0.79265696,
      "num_input_tokens_seen": 35565010,
      "step": 1677,
      "time_per_iteration": 2.6108388900756836
    },
    {
      "auxiliary_loss_clip": 0.01103065,
      "auxiliary_loss_mlp": 0.01000476,
      "balance_loss_clip": 1.02454388,
      "balance_loss_mlp": 0.99802065,
      "epoch": 0.20176757049239463,
      "flos": 71100616037760.0,
      "grad_norm": 0.6932006752150882,
      "language_loss": 0.58356351,
      "learning_rate": 3.6985500333109474e-06,
      "loss": 0.60459894,
      "num_input_tokens_seen": 35633340,
      "step": 1678,
      "time_per_iteration": 3.211085081100464
    },
    {
      "auxiliary_loss_clip": 0.01165351,
      "auxiliary_loss_mlp": 0.01034402,
      "balance_loss_clip": 1.0536654,
      "balance_loss_mlp": 1.0252645,
      "epoch": 0.20188781338303372,
      "flos": 21430159637760.0,
      "grad_norm": 2.292766218170489,
      "language_loss": 0.7631526,
      "learning_rate": 3.6981386458572385e-06,
      "loss": 0.78515017,
      "num_input_tokens_seen": 35651315,
      "step": 1679,
      "time_per_iteration": 5.008856773376465
    },
    {
      "auxiliary_loss_clip": 0.01169122,
      "auxiliary_loss_mlp": 0.0103906,
      "balance_loss_clip": 1.05624056,
      "balance_loss_mlp": 1.02888584,
      "epoch": 0.20200805627367283,
      "flos": 11546215130880.0,
      "grad_norm": 2.6081241469431906,
      "language_loss": 0.7658906,
      "learning_rate": 3.6977270007972468e-06,
      "loss": 0.78797245,
      "num_input_tokens_seen": 35668850,
      "step": 1680,
      "time_per_iteration": 3.4652955532073975
    },
    {
      "auxiliary_loss_clip": 0.01191109,
      "auxiliary_loss_mlp": 0.01033774,
      "balance_loss_clip": 1.06098557,
      "balance_loss_mlp": 1.02419519,
      "epoch": 0.2021282991643119,
      "flos": 28545391906560.0,
      "grad_norm": 3.541670296225184,
      "language_loss": 0.7194438,
      "learning_rate": 3.6973150981934196e-06,
      "loss": 0.7416926,
      "num_input_tokens_seen": 35690080,
      "step": 1681,
      "time_per_iteration": 2.653881549835205
    },
    {
      "auxiliary_loss_clip": 0.01221046,
      "auxiliary_loss_mlp": 0.01035603,
      "balance_loss_clip": 1.06283867,
      "balance_loss_mlp": 1.025846,
      "epoch": 0.202248542054951,
      "flos": 17923446564480.0,
      "grad_norm": 2.4605282363226184,
      "language_loss": 0.84155953,
      "learning_rate": 3.6969029381082415e-06,
      "loss": 0.86412597,
      "num_input_tokens_seen": 35706075,
      "step": 1682,
      "time_per_iteration": 2.495000123977661
    },
    {
      "auxiliary_loss_clip": 0.01184506,
      "auxiliary_loss_mlp": 0.01031277,
      "balance_loss_clip": 1.05899298,
      "balance_loss_mlp": 1.02191353,
      "epoch": 0.2023687849455901,
      "flos": 19864634296320.0,
      "grad_norm": 3.9453847076596618,
      "language_loss": 0.79655206,
      "learning_rate": 3.696490520604237e-06,
      "loss": 0.81870985,
      "num_input_tokens_seen": 35724765,
      "step": 1683,
      "time_per_iteration": 2.578317403793335
    },
    {
      "auxiliary_loss_clip": 0.01198847,
      "auxiliary_loss_mlp": 0.01025917,
      "balance_loss_clip": 1.0611608,
      "balance_loss_mlp": 1.01688099,
      "epoch": 0.20248902783622919,
      "flos": 22564721600640.0,
      "grad_norm": 1.7052939037775134,
      "language_loss": 0.80656451,
      "learning_rate": 3.696077845743968e-06,
      "loss": 0.82881212,
      "num_input_tokens_seen": 35744355,
      "step": 1684,
      "time_per_iteration": 2.5487425327301025
    },
    {
      "auxiliary_loss_clip": 0.01221345,
      "auxiliary_loss_mlp": 0.01033491,
      "balance_loss_clip": 1.06353426,
      "balance_loss_mlp": 1.02342391,
      "epoch": 0.20260927072686827,
      "flos": 22709728805760.0,
      "grad_norm": 2.3815747828739964,
      "language_loss": 0.73337722,
      "learning_rate": 3.69566491359004e-06,
      "loss": 0.75592554,
      "num_input_tokens_seen": 35761000,
      "step": 1685,
      "time_per_iteration": 2.528510093688965
    },
    {
      "auxiliary_loss_clip": 0.0118516,
      "auxiliary_loss_mlp": 0.01034886,
      "balance_loss_clip": 1.05817342,
      "balance_loss_mlp": 1.02492011,
      "epoch": 0.20272951361750738,
      "flos": 51023998650240.0,
      "grad_norm": 3.933181495170971,
      "language_loss": 0.69612551,
      "learning_rate": 3.695251724205092e-06,
      "loss": 0.71832597,
      "num_input_tokens_seen": 35785360,
      "step": 1686,
      "time_per_iteration": 2.8178625106811523
    },
    {
      "auxiliary_loss_clip": 0.01216492,
      "auxiliary_loss_mlp": 0.01033008,
      "balance_loss_clip": 1.06177151,
      "balance_loss_mlp": 1.02323294,
      "epoch": 0.20284975650814646,
      "flos": 26578133879040.0,
      "grad_norm": 1.6286031159580288,
      "language_loss": 0.86638433,
      "learning_rate": 3.6948382776518054e-06,
      "loss": 0.8888793,
      "num_input_tokens_seen": 35806065,
      "step": 1687,
      "time_per_iteration": 2.5669915676116943
    },
    {
      "auxiliary_loss_clip": 0.01181965,
      "auxiliary_loss_mlp": 0.01039167,
      "balance_loss_clip": 1.05671179,
      "balance_loss_mlp": 1.02941513,
      "epoch": 0.20296999939878554,
      "flos": 16034222833920.0,
      "grad_norm": 4.514600189293757,
      "language_loss": 0.79371703,
      "learning_rate": 3.6944245739929e-06,
      "loss": 0.81592834,
      "num_input_tokens_seen": 35822225,
      "step": 1688,
      "time_per_iteration": 2.549323797225952
    },
    {
      "auxiliary_loss_clip": 0.01201791,
      "auxiliary_loss_mlp": 0.01040372,
      "balance_loss_clip": 1.06029034,
      "balance_loss_mlp": 1.03046536,
      "epoch": 0.20309024228942463,
      "flos": 19203374868480.0,
      "grad_norm": 6.2101087334099425,
      "language_loss": 0.71964628,
      "learning_rate": 3.6940106132911332e-06,
      "loss": 0.74206793,
      "num_input_tokens_seen": 35839410,
      "step": 1689,
      "time_per_iteration": 2.533395767211914
    },
    {
      "auxiliary_loss_clip": 0.01207532,
      "auxiliary_loss_mlp": 0.01034316,
      "balance_loss_clip": 1.06277561,
      "balance_loss_mlp": 1.02512527,
      "epoch": 0.20321048518006374,
      "flos": 22821087945600.0,
      "grad_norm": 1.7585589232562562,
      "language_loss": 0.88923246,
      "learning_rate": 3.6935963956093037e-06,
      "loss": 0.91165096,
      "num_input_tokens_seen": 35859495,
      "step": 1690,
      "time_per_iteration": 2.545464038848877
    },
    {
      "auxiliary_loss_clip": 0.01194583,
      "auxiliary_loss_mlp": 0.01032331,
      "balance_loss_clip": 1.05826819,
      "balance_loss_mlp": 1.02330697,
      "epoch": 0.20333072807070282,
      "flos": 19096397187840.0,
      "grad_norm": 1.734803479224798,
      "language_loss": 0.69162923,
      "learning_rate": 3.6931819210102474e-06,
      "loss": 0.71389842,
      "num_input_tokens_seen": 35878890,
      "step": 1691,
      "time_per_iteration": 2.5537819862365723
    },
    {
      "auxiliary_loss_clip": 0.0122165,
      "auxiliary_loss_mlp": 0.0103516,
      "balance_loss_clip": 1.06419921,
      "balance_loss_mlp": 1.02537251,
      "epoch": 0.2034509709613419,
      "flos": 18180962144640.0,
      "grad_norm": 2.022279199970276,
      "language_loss": 0.8474915,
      "learning_rate": 3.6927671895568402e-06,
      "loss": 0.87005955,
      "num_input_tokens_seen": 35897950,
      "step": 1692,
      "time_per_iteration": 2.5202362537384033
    },
    {
      "auxiliary_loss_clip": 0.01218279,
      "auxiliary_loss_mlp": 0.01035747,
      "balance_loss_clip": 1.0636524,
      "balance_loss_mlp": 1.02619874,
      "epoch": 0.20357121385198101,
      "flos": 22923899648640.0,
      "grad_norm": 1.8241207289876609,
      "language_loss": 0.86901534,
      "learning_rate": 3.692352201311996e-06,
      "loss": 0.89155555,
      "num_input_tokens_seen": 35916800,
      "step": 1693,
      "time_per_iteration": 2.512848138809204
    },
    {
      "auxiliary_loss_clip": 0.01169803,
      "auxiliary_loss_mlp": 0.01029335,
      "balance_loss_clip": 1.05633664,
      "balance_loss_mlp": 1.02000093,
      "epoch": 0.2036914567426201,
      "flos": 20922131629440.0,
      "grad_norm": 1.7842334458437972,
      "language_loss": 0.76321661,
      "learning_rate": 3.6919369563386687e-06,
      "loss": 0.78520793,
      "num_input_tokens_seen": 35936600,
      "step": 1694,
      "time_per_iteration": 2.620120048522949
    },
    {
      "auxiliary_loss_clip": 0.01187043,
      "auxiliary_loss_mlp": 0.01032737,
      "balance_loss_clip": 1.06135082,
      "balance_loss_mlp": 1.02387381,
      "epoch": 0.20381169963325918,
      "flos": 15519155760000.0,
      "grad_norm": 2.6611281919122014,
      "language_loss": 0.78831178,
      "learning_rate": 3.69152145469985e-06,
      "loss": 0.81050956,
      "num_input_tokens_seen": 35953645,
      "step": 1695,
      "time_per_iteration": 2.553910255432129
    },
    {
      "auxiliary_loss_clip": 0.01162611,
      "auxiliary_loss_mlp": 0.01044852,
      "balance_loss_clip": 1.05558705,
      "balance_loss_mlp": 1.03397417,
      "epoch": 0.20393194252389826,
      "flos": 28833143760000.0,
      "grad_norm": 2.0520999050451434,
      "language_loss": 0.81947726,
      "learning_rate": 3.691105696458572e-06,
      "loss": 0.8415519,
      "num_input_tokens_seen": 35970940,
      "step": 1696,
      "time_per_iteration": 2.6742448806762695
    },
    {
      "auxiliary_loss_clip": 0.01220367,
      "auxiliary_loss_mlp": 0.01030749,
      "balance_loss_clip": 1.06654644,
      "balance_loss_mlp": 1.02135515,
      "epoch": 0.20405218541453737,
      "flos": 22488554810880.0,
      "grad_norm": 4.328252715624741,
      "language_loss": 0.67752624,
      "learning_rate": 3.690689681677904e-06,
      "loss": 0.70003736,
      "num_input_tokens_seen": 35989410,
      "step": 1697,
      "time_per_iteration": 2.527660846710205
    },
    {
      "auxiliary_loss_clip": 0.01188608,
      "auxiliary_loss_mlp": 0.01031746,
      "balance_loss_clip": 1.05847466,
      "balance_loss_mlp": 1.02272201,
      "epoch": 0.20417242830517646,
      "flos": 25374408278400.0,
      "grad_norm": 1.7069664040136512,
      "language_loss": 0.88590586,
      "learning_rate": 3.690273410420956e-06,
      "loss": 0.90810943,
      "num_input_tokens_seen": 36009175,
      "step": 1698,
      "time_per_iteration": 2.61661434173584
    },
    {
      "auxiliary_loss_clip": 0.01202719,
      "auxiliary_loss_mlp": 0.0103629,
      "balance_loss_clip": 1.06076145,
      "balance_loss_mlp": 1.02700412,
      "epoch": 0.20429267119581554,
      "flos": 14793078240000.0,
      "grad_norm": 4.383743480557219,
      "language_loss": 0.76848876,
      "learning_rate": 3.689856882750875e-06,
      "loss": 0.79087889,
      "num_input_tokens_seen": 36024375,
      "step": 1699,
      "time_per_iteration": 3.3093175888061523
    },
    {
      "auxiliary_loss_clip": 0.01200932,
      "auxiliary_loss_mlp": 0.010327,
      "balance_loss_clip": 1.06264687,
      "balance_loss_mlp": 1.02391386,
      "epoch": 0.20441291408645465,
      "flos": 17781851151360.0,
      "grad_norm": 1.8709501683404441,
      "language_loss": 0.78793871,
      "learning_rate": 3.6894400987308486e-06,
      "loss": 0.81027502,
      "num_input_tokens_seen": 36041895,
      "step": 1700,
      "time_per_iteration": 2.4629805088043213
    },
    {
      "auxiliary_loss_clip": 0.01207364,
      "auxiliary_loss_mlp": 0.01038483,
      "balance_loss_clip": 1.06224144,
      "balance_loss_mlp": 1.02873802,
      "epoch": 0.20453315697709373,
      "flos": 16435668211200.0,
      "grad_norm": 1.8963789630471082,
      "language_loss": 0.84851086,
      "learning_rate": 3.6890230584241024e-06,
      "loss": 0.87096936,
      "num_input_tokens_seen": 36058825,
      "step": 1701,
      "time_per_iteration": 2.473630905151367
    },
    {
      "auxiliary_loss_clip": 0.01117082,
      "auxiliary_loss_mlp": 0.01017989,
      "balance_loss_clip": 1.02653635,
      "balance_loss_mlp": 1.01595056,
      "epoch": 0.20465339986773282,
      "flos": 66713085653760.0,
      "grad_norm": 1.079628427469279,
      "language_loss": 0.66417402,
      "learning_rate": 3.6886057618939016e-06,
      "loss": 0.6855247,
      "num_input_tokens_seen": 36121645,
      "step": 1702,
      "time_per_iteration": 3.0877153873443604
    },
    {
      "auxiliary_loss_clip": 0.01166953,
      "auxiliary_loss_mlp": 0.01043209,
      "balance_loss_clip": 1.0561502,
      "balance_loss_mlp": 1.03309953,
      "epoch": 0.2047736427583719,
      "flos": 41974114924800.0,
      "grad_norm": 1.9040943911125248,
      "language_loss": 0.6910609,
      "learning_rate": 3.6881882092035492e-06,
      "loss": 0.71316254,
      "num_input_tokens_seen": 36143030,
      "step": 1703,
      "time_per_iteration": 2.7542049884796143
    },
    {
      "auxiliary_loss_clip": 0.01088371,
      "auxiliary_loss_mlp": 0.00755769,
      "balance_loss_clip": 1.02766347,
      "balance_loss_mlp": 1.00053215,
      "epoch": 0.204893885649011,
      "flos": 69940878641280.0,
      "grad_norm": 0.9258463559652654,
      "language_loss": 0.61184669,
      "learning_rate": 3.6877704004163873e-06,
      "loss": 0.630288,
      "num_input_tokens_seen": 36203435,
      "step": 1704,
      "time_per_iteration": 3.2626121044158936
    },
    {
      "auxiliary_loss_clip": 0.01220419,
      "auxiliary_loss_mlp": 0.01033827,
      "balance_loss_clip": 1.06428337,
      "balance_loss_mlp": 1.02348018,
      "epoch": 0.2050141285396501,
      "flos": 22200012858240.0,
      "grad_norm": 2.175912443947148,
      "language_loss": 0.77588618,
      "learning_rate": 3.6873523355957984e-06,
      "loss": 0.79842865,
      "num_input_tokens_seen": 36222435,
      "step": 1705,
      "time_per_iteration": 3.327622413635254
    },
    {
      "auxiliary_loss_clip": 0.01115365,
      "auxiliary_loss_mlp": 0.01006685,
      "balance_loss_clip": 1.02520096,
      "balance_loss_mlp": 1.0046463,
      "epoch": 0.20513437143028918,
      "flos": 46283721730560.0,
      "grad_norm": 0.9799203017007736,
      "language_loss": 0.6413393,
      "learning_rate": 3.686934014805201e-06,
      "loss": 0.66255981,
      "num_input_tokens_seen": 36273065,
      "step": 1706,
      "time_per_iteration": 3.7156033515930176
    },
    {
      "auxiliary_loss_clip": 0.01202934,
      "auxiliary_loss_mlp": 0.01038778,
      "balance_loss_clip": 1.06356168,
      "balance_loss_mlp": 1.02921748,
      "epoch": 0.20525461432092829,
      "flos": 21904324099200.0,
      "grad_norm": 1.8423043649697464,
      "language_loss": 0.81273705,
      "learning_rate": 3.6865154381080552e-06,
      "loss": 0.83515418,
      "num_input_tokens_seen": 36293750,
      "step": 1707,
      "time_per_iteration": 3.3514809608459473
    },
    {
      "auxiliary_loss_clip": 0.01128011,
      "auxiliary_loss_mlp": 0.010299,
      "balance_loss_clip": 1.05236316,
      "balance_loss_mlp": 1.02069092,
      "epoch": 0.20537485721156737,
      "flos": 21214264942080.0,
      "grad_norm": 1.94272526529882,
      "language_loss": 0.82480317,
      "learning_rate": 3.6860966055678585e-06,
      "loss": 0.84638226,
      "num_input_tokens_seen": 36310105,
      "step": 1708,
      "time_per_iteration": 2.6890857219696045
    },
    {
      "auxiliary_loss_clip": 0.01205871,
      "auxiliary_loss_mlp": 0.01041815,
      "balance_loss_clip": 1.06422508,
      "balance_loss_mlp": 1.0317837,
      "epoch": 0.20549510010220645,
      "flos": 20191205773440.0,
      "grad_norm": 1.8327083175605325,
      "language_loss": 0.86357588,
      "learning_rate": 3.685677517248147e-06,
      "loss": 0.88605273,
      "num_input_tokens_seen": 36328995,
      "step": 1709,
      "time_per_iteration": 2.5011420249938965
    },
    {
      "auxiliary_loss_clip": 0.01189419,
      "auxiliary_loss_mlp": 0.00764406,
      "balance_loss_clip": 1.06558514,
      "balance_loss_mlp": 1.00087571,
      "epoch": 0.20561534299284553,
      "flos": 17016702612480.0,
      "grad_norm": 1.8940528465644986,
      "language_loss": 0.80435836,
      "learning_rate": 3.6852581732124967e-06,
      "loss": 0.82389659,
      "num_input_tokens_seen": 36346340,
      "step": 1710,
      "time_per_iteration": 2.5553958415985107
    },
    {
      "auxiliary_loss_clip": 0.01206969,
      "auxiliary_loss_mlp": 0.01033567,
      "balance_loss_clip": 1.06480241,
      "balance_loss_mlp": 1.02351213,
      "epoch": 0.20573558588348465,
      "flos": 22890467064960.0,
      "grad_norm": 2.9550287906302084,
      "language_loss": 0.76077873,
      "learning_rate": 3.6848385735245213e-06,
      "loss": 0.78318411,
      "num_input_tokens_seen": 36365430,
      "step": 1711,
      "time_per_iteration": 2.545814275741577
    },
    {
      "auxiliary_loss_clip": 0.01186558,
      "auxiliary_loss_mlp": 0.01032695,
      "balance_loss_clip": 1.05517054,
      "balance_loss_mlp": 1.02336097,
      "epoch": 0.20585582877412373,
      "flos": 24643123286400.0,
      "grad_norm": 1.816164872469563,
      "language_loss": 0.86060917,
      "learning_rate": 3.6844187182478734e-06,
      "loss": 0.88280165,
      "num_input_tokens_seen": 36386285,
      "step": 1712,
      "time_per_iteration": 2.5832128524780273
    },
    {
      "auxiliary_loss_clip": 0.01178493,
      "auxiliary_loss_mlp": 0.01030458,
      "balance_loss_clip": 1.05583549,
      "balance_loss_mlp": 1.02099848,
      "epoch": 0.2059760716647628,
      "flos": 24206952435840.0,
      "grad_norm": 1.656136867252933,
      "language_loss": 0.74730504,
      "learning_rate": 3.683998607446246e-06,
      "loss": 0.76939458,
      "num_input_tokens_seen": 36404935,
      "step": 1713,
      "time_per_iteration": 2.5760860443115234
    },
    {
      "auxiliary_loss_clip": 0.01204658,
      "auxiliary_loss_mlp": 0.01045292,
      "balance_loss_clip": 1.06433654,
      "balance_loss_mlp": 1.03669727,
      "epoch": 0.20609631455540192,
      "flos": 20229522606720.0,
      "grad_norm": 1.8677170672975212,
      "language_loss": 0.75019348,
      "learning_rate": 3.6835782411833686e-06,
      "loss": 0.77269304,
      "num_input_tokens_seen": 36424455,
      "step": 1714,
      "time_per_iteration": 2.557457447052002
    },
    {
      "auxiliary_loss_clip": 0.01163789,
      "auxiliary_loss_mlp": 0.01034639,
      "balance_loss_clip": 1.05579591,
      "balance_loss_mlp": 1.02487016,
      "epoch": 0.206216557446041,
      "flos": 19864957518720.0,
      "grad_norm": 1.9216832351580975,
      "language_loss": 0.74329787,
      "learning_rate": 3.68315761952301e-06,
      "loss": 0.76528215,
      "num_input_tokens_seen": 36441685,
      "step": 1715,
      "time_per_iteration": 2.557408094406128
    },
    {
      "auxiliary_loss_clip": 0.01220315,
      "auxiliary_loss_mlp": 0.01037572,
      "balance_loss_clip": 1.06523728,
      "balance_loss_mlp": 1.02794552,
      "epoch": 0.2063368003366801,
      "flos": 24096311568000.0,
      "grad_norm": 2.007141782840156,
      "language_loss": 0.83047593,
      "learning_rate": 3.6827367425289797e-06,
      "loss": 0.85305476,
      "num_input_tokens_seen": 36461460,
      "step": 1716,
      "time_per_iteration": 2.5541040897369385
    },
    {
      "auxiliary_loss_clip": 0.01190372,
      "auxiliary_loss_mlp": 0.01030896,
      "balance_loss_clip": 1.06114793,
      "balance_loss_mlp": 1.02061439,
      "epoch": 0.2064570432273192,
      "flos": 20340163474560.0,
      "grad_norm": 2.7823186274286016,
      "language_loss": 0.72970492,
      "learning_rate": 3.6823156102651225e-06,
      "loss": 0.7519176,
      "num_input_tokens_seen": 36479615,
      "step": 1717,
      "time_per_iteration": 2.5223388671875
    },
    {
      "auxiliary_loss_clip": 0.01134096,
      "auxiliary_loss_mlp": 0.010311,
      "balance_loss_clip": 1.05721724,
      "balance_loss_mlp": 1.02146244,
      "epoch": 0.20657728611795828,
      "flos": 20520363029760.0,
      "grad_norm": 1.8698432655575277,
      "language_loss": 0.70648134,
      "learning_rate": 3.6818942227953257e-06,
      "loss": 0.7281332,
      "num_input_tokens_seen": 36500160,
      "step": 1718,
      "time_per_iteration": 2.614090919494629
    },
    {
      "auxiliary_loss_clip": 0.01175611,
      "auxiliary_loss_mlp": 0.01031033,
      "balance_loss_clip": 1.0603174,
      "balance_loss_mlp": 1.02132297,
      "epoch": 0.20669752900859736,
      "flos": 21799285752960.0,
      "grad_norm": 1.9106012471330371,
      "language_loss": 0.69166601,
      "learning_rate": 3.681472580183512e-06,
      "loss": 0.71373242,
      "num_input_tokens_seen": 36518810,
      "step": 1719,
      "time_per_iteration": 2.5682260990142822
    },
    {
      "auxiliary_loss_clip": 0.0120147,
      "auxiliary_loss_mlp": 0.01032327,
      "balance_loss_clip": 1.06394529,
      "balance_loss_mlp": 1.02333891,
      "epoch": 0.20681777189923645,
      "flos": 15122020014720.0,
      "grad_norm": 1.7915004781149155,
      "language_loss": 0.86325067,
      "learning_rate": 3.6810506824936455e-06,
      "loss": 0.88558865,
      "num_input_tokens_seen": 36536890,
      "step": 1720,
      "time_per_iteration": 2.47111177444458
    },
    {
      "auxiliary_loss_clip": 0.01087482,
      "auxiliary_loss_mlp": 0.010051,
      "balance_loss_clip": 1.02257824,
      "balance_loss_mlp": 1.00282288,
      "epoch": 0.20693801478987556,
      "flos": 56481021509760.0,
      "grad_norm": 1.1317507893838084,
      "language_loss": 0.62530267,
      "learning_rate": 3.680628529789726e-06,
      "loss": 0.64622855,
      "num_input_tokens_seen": 36589300,
      "step": 1721,
      "time_per_iteration": 2.94596266746521
    },
    {
      "auxiliary_loss_clip": 0.012262,
      "auxiliary_loss_mlp": 0.01039448,
      "balance_loss_clip": 1.0663687,
      "balance_loss_mlp": 1.02882624,
      "epoch": 0.20705825768051464,
      "flos": 21614201948160.0,
      "grad_norm": 1.8765563320858571,
      "language_loss": 0.85612869,
      "learning_rate": 3.680206122135796e-06,
      "loss": 0.87878513,
      "num_input_tokens_seen": 36609905,
      "step": 1722,
      "time_per_iteration": 2.4761531352996826
    },
    {
      "auxiliary_loss_clip": 0.0116922,
      "auxiliary_loss_mlp": 0.010387,
      "balance_loss_clip": 1.06507683,
      "balance_loss_mlp": 1.02997386,
      "epoch": 0.20717850057115372,
      "flos": 25848895962240.0,
      "grad_norm": 1.803188406557687,
      "language_loss": 0.78218549,
      "learning_rate": 3.6797834595959323e-06,
      "loss": 0.80426466,
      "num_input_tokens_seen": 36629805,
      "step": 1723,
      "time_per_iteration": 2.61983323097229
    },
    {
      "auxiliary_loss_clip": 0.01147727,
      "auxiliary_loss_mlp": 0.01042117,
      "balance_loss_clip": 1.05313325,
      "balance_loss_mlp": 1.03179991,
      "epoch": 0.20729874346179283,
      "flos": 29130807767040.0,
      "grad_norm": 2.4710442945737277,
      "language_loss": 0.78367615,
      "learning_rate": 3.679360542234254e-06,
      "loss": 0.80557466,
      "num_input_tokens_seen": 36649150,
      "step": 1724,
      "time_per_iteration": 2.654662847518921
    },
    {
      "auxiliary_loss_clip": 0.01182316,
      "auxiliary_loss_mlp": 0.00764963,
      "balance_loss_clip": 1.05590117,
      "balance_loss_mlp": 1.00089419,
      "epoch": 0.20741898635243192,
      "flos": 29023363209600.0,
      "grad_norm": 1.6746600180368396,
      "language_loss": 0.72186583,
      "learning_rate": 3.678937370114916e-06,
      "loss": 0.74133861,
      "num_input_tokens_seen": 36668955,
      "step": 1725,
      "time_per_iteration": 3.48898983001709
    },
    {
      "auxiliary_loss_clip": 0.01184175,
      "auxiliary_loss_mlp": 0.0102942,
      "balance_loss_clip": 1.06155932,
      "balance_loss_mlp": 1.02088475,
      "epoch": 0.207539229243071,
      "flos": 15559447841280.0,
      "grad_norm": 1.9454122270823844,
      "language_loss": 0.78826487,
      "learning_rate": 3.678513943302114e-06,
      "loss": 0.81040084,
      "num_input_tokens_seen": 36685730,
      "step": 1726,
      "time_per_iteration": 2.545203924179077
    },
    {
      "auxiliary_loss_clip": 0.0121827,
      "auxiliary_loss_mlp": 0.01035791,
      "balance_loss_clip": 1.06433117,
      "balance_loss_mlp": 1.02687383,
      "epoch": 0.20765947213371008,
      "flos": 20521081301760.0,
      "grad_norm": 1.6933242353386795,
      "language_loss": 0.8538264,
      "learning_rate": 3.678090261860082e-06,
      "loss": 0.87636709,
      "num_input_tokens_seen": 36705460,
      "step": 1727,
      "time_per_iteration": 2.5430078506469727
    },
    {
      "auxiliary_loss_clip": 0.01172423,
      "auxiliary_loss_mlp": 0.01035899,
      "balance_loss_clip": 1.05392444,
      "balance_loss_mlp": 1.02663016,
      "epoch": 0.2077797150243492,
      "flos": 19354415558400.0,
      "grad_norm": 1.8073339637475292,
      "language_loss": 0.77326918,
      "learning_rate": 3.6776663258530906e-06,
      "loss": 0.7953524,
      "num_input_tokens_seen": 36724110,
      "step": 1728,
      "time_per_iteration": 2.6801679134368896
    },
    {
      "auxiliary_loss_clip": 0.01206503,
      "auxiliary_loss_mlp": 0.01032931,
      "balance_loss_clip": 1.06291127,
      "balance_loss_mlp": 1.0240382,
      "epoch": 0.20789995791498828,
      "flos": 21829952989440.0,
      "grad_norm": 1.8307166220351654,
      "language_loss": 0.71375144,
      "learning_rate": 3.6772421353454516e-06,
      "loss": 0.73614573,
      "num_input_tokens_seen": 36742705,
      "step": 1729,
      "time_per_iteration": 2.6156044006347656
    },
    {
      "auxiliary_loss_clip": 0.01202153,
      "auxiliary_loss_mlp": 0.01032222,
      "balance_loss_clip": 1.06386757,
      "balance_loss_mlp": 1.02293587,
      "epoch": 0.20802020080562736,
      "flos": 23148844571520.0,
      "grad_norm": 8.882434085461293,
      "language_loss": 0.8839817,
      "learning_rate": 3.6768176904015153e-06,
      "loss": 0.90632546,
      "num_input_tokens_seen": 36762510,
      "step": 1730,
      "time_per_iteration": 2.529400587081909
    },
    {
      "auxiliary_loss_clip": 0.01203104,
      "auxiliary_loss_mlp": 0.01038505,
      "balance_loss_clip": 1.06127071,
      "balance_loss_mlp": 1.02921844,
      "epoch": 0.20814044369626647,
      "flos": 23072677781760.0,
      "grad_norm": 2.6368028659940492,
      "language_loss": 0.60422206,
      "learning_rate": 3.6763929910856674e-06,
      "loss": 0.62663811,
      "num_input_tokens_seen": 36780960,
      "step": 1731,
      "time_per_iteration": 3.3543660640716553
    },
    {
      "auxiliary_loss_clip": 0.01203673,
      "auxiliary_loss_mlp": 0.01041385,
      "balance_loss_clip": 1.06389642,
      "balance_loss_mlp": 1.03196144,
      "epoch": 0.20826068658690555,
      "flos": 19608016556160.0,
      "grad_norm": 2.414765034132618,
      "language_loss": 0.77754235,
      "learning_rate": 3.6759680374623365e-06,
      "loss": 0.79999292,
      "num_input_tokens_seen": 36798875,
      "step": 1732,
      "time_per_iteration": 3.3706953525543213
    },
    {
      "auxiliary_loss_clip": 0.01216353,
      "auxiliary_loss_mlp": 0.01030399,
      "balance_loss_clip": 1.06389928,
      "balance_loss_mlp": 1.02096367,
      "epoch": 0.20838092947754464,
      "flos": 25374049142400.0,
      "grad_norm": 2.383951389535304,
      "language_loss": 0.75453377,
      "learning_rate": 3.675542829595986e-06,
      "loss": 0.7770012,
      "num_input_tokens_seen": 36818540,
      "step": 1733,
      "time_per_iteration": 3.331346035003662
    },
    {
      "auxiliary_loss_clip": 0.01187756,
      "auxiliary_loss_mlp": 0.01034533,
      "balance_loss_clip": 1.05992317,
      "balance_loss_mlp": 1.02521074,
      "epoch": 0.20850117236818372,
      "flos": 24061729749120.0,
      "grad_norm": 1.437901828774705,
      "language_loss": 0.79396731,
      "learning_rate": 3.6751173675511213e-06,
      "loss": 0.81619018,
      "num_input_tokens_seen": 36840585,
      "step": 1734,
      "time_per_iteration": 2.581486463546753
    },
    {
      "auxiliary_loss_clip": 0.01182469,
      "auxiliary_loss_mlp": 0.01036183,
      "balance_loss_clip": 1.05398667,
      "balance_loss_mlp": 1.02736163,
      "epoch": 0.20862141525882283,
      "flos": 20077799558400.0,
      "grad_norm": 2.1602936115016993,
      "language_loss": 0.87514347,
      "learning_rate": 3.674691651392283e-06,
      "loss": 0.89733005,
      "num_input_tokens_seen": 36858255,
      "step": 1735,
      "time_per_iteration": 2.5433480739593506
    },
    {
      "auxiliary_loss_clip": 0.01194135,
      "auxiliary_loss_mlp": 0.01045702,
      "balance_loss_clip": 1.0634383,
      "balance_loss_mlp": 1.03649926,
      "epoch": 0.2087416581494619,
      "flos": 39015183237120.0,
      "grad_norm": 2.4436272545325513,
      "language_loss": 0.76214147,
      "learning_rate": 3.674265681184053e-06,
      "loss": 0.78453982,
      "num_input_tokens_seen": 36881515,
      "step": 1736,
      "time_per_iteration": 2.6752898693084717
    },
    {
      "auxiliary_loss_clip": 0.01188291,
      "auxiliary_loss_mlp": 0.01031949,
      "balance_loss_clip": 1.05806065,
      "balance_loss_mlp": 1.02281713,
      "epoch": 0.208861901040101,
      "flos": 26101994169600.0,
      "grad_norm": 1.6540068303848314,
      "language_loss": 0.86296463,
      "learning_rate": 3.6738394569910504e-06,
      "loss": 0.885167,
      "num_input_tokens_seen": 36902055,
      "step": 1737,
      "time_per_iteration": 2.56394362449646
    },
    {
      "auxiliary_loss_clip": 0.01202918,
      "auxiliary_loss_mlp": 0.01032384,
      "balance_loss_clip": 1.06375539,
      "balance_loss_mlp": 1.02349091,
      "epoch": 0.2089821439307401,
      "flos": 28398732675840.0,
      "grad_norm": 2.2776346492230375,
      "language_loss": 0.82504511,
      "learning_rate": 3.6734129788779333e-06,
      "loss": 0.8473981,
      "num_input_tokens_seen": 36921230,
      "step": 1738,
      "time_per_iteration": 2.5582215785980225
    },
    {
      "auxiliary_loss_clip": 0.01170748,
      "auxiliary_loss_mlp": 0.01031851,
      "balance_loss_clip": 1.06071949,
      "balance_loss_mlp": 1.02264166,
      "epoch": 0.2091023868213792,
      "flos": 21069616872960.0,
      "grad_norm": 1.6406682291254882,
      "language_loss": 0.90560412,
      "learning_rate": 3.6729862469093976e-06,
      "loss": 0.92763019,
      "num_input_tokens_seen": 36940325,
      "step": 1739,
      "time_per_iteration": 2.5706984996795654
    },
    {
      "auxiliary_loss_clip": 0.01172463,
      "auxiliary_loss_mlp": 0.0103843,
      "balance_loss_clip": 1.05536759,
      "balance_loss_mlp": 1.02911961,
      "epoch": 0.20922262971201827,
      "flos": 22455481363200.0,
      "grad_norm": 2.4190153924251176,
      "language_loss": 0.82254064,
      "learning_rate": 3.6725592611501782e-06,
      "loss": 0.84464955,
      "num_input_tokens_seen": 36959000,
      "step": 1740,
      "time_per_iteration": 2.5368447303771973
    },
    {
      "auxiliary_loss_clip": 0.01200981,
      "auxiliary_loss_mlp": 0.01039083,
      "balance_loss_clip": 1.06038165,
      "balance_loss_mlp": 1.02958798,
      "epoch": 0.20934287260265738,
      "flos": 27852244179840.0,
      "grad_norm": 3.252599703194495,
      "language_loss": 0.7698943,
      "learning_rate": 3.6721320216650496e-06,
      "loss": 0.79229492,
      "num_input_tokens_seen": 36979615,
      "step": 1741,
      "time_per_iteration": 2.5715086460113525
    },
    {
      "auxiliary_loss_clip": 0.01185693,
      "auxiliary_loss_mlp": 0.0104033,
      "balance_loss_clip": 1.0593679,
      "balance_loss_mlp": 1.03079963,
      "epoch": 0.20946311549329646,
      "flos": 16435309075200.0,
      "grad_norm": 1.6848133030444423,
      "language_loss": 0.83857214,
      "learning_rate": 3.6717045285188215e-06,
      "loss": 0.86083245,
      "num_input_tokens_seen": 36997310,
      "step": 1742,
      "time_per_iteration": 2.538525342941284
    },
    {
      "auxiliary_loss_clip": 0.01141515,
      "auxiliary_loss_mlp": 0.01036919,
      "balance_loss_clip": 1.05084956,
      "balance_loss_mlp": 1.02732861,
      "epoch": 0.20958335838393555,
      "flos": 22492720788480.0,
      "grad_norm": 2.1099686823320654,
      "language_loss": 0.87020957,
      "learning_rate": 3.671276781776346e-06,
      "loss": 0.89199394,
      "num_input_tokens_seen": 37015965,
      "step": 1743,
      "time_per_iteration": 2.6134045124053955
    },
    {
      "auxiliary_loss_clip": 0.01176265,
      "auxiliary_loss_mlp": 0.01033339,
      "balance_loss_clip": 1.05311918,
      "balance_loss_mlp": 1.02415419,
      "epoch": 0.20970360127457463,
      "flos": 25224768218880.0,
      "grad_norm": 1.906691330481457,
      "language_loss": 0.67145264,
      "learning_rate": 3.6708487815025128e-06,
      "loss": 0.69354868,
      "num_input_tokens_seen": 37036545,
      "step": 1744,
      "time_per_iteration": 2.58872652053833
    },
    {
      "auxiliary_loss_clip": 0.01172732,
      "auxiliary_loss_mlp": 0.01030618,
      "balance_loss_clip": 1.05657196,
      "balance_loss_mlp": 1.02116442,
      "epoch": 0.20982384416521374,
      "flos": 18479164855680.0,
      "grad_norm": 2.292926979330607,
      "language_loss": 0.74404144,
      "learning_rate": 3.6704205277622463e-06,
      "loss": 0.7660749,
      "num_input_tokens_seen": 37054985,
      "step": 1745,
      "time_per_iteration": 2.6021785736083984
    },
    {
      "auxiliary_loss_clip": 0.01187591,
      "auxiliary_loss_mlp": 0.01033413,
      "balance_loss_clip": 1.05626488,
      "balance_loss_mlp": 1.0242219,
      "epoch": 0.20994408705585282,
      "flos": 25373546352000.0,
      "grad_norm": 1.6762954296293981,
      "language_loss": 0.80512577,
      "learning_rate": 3.6699920206205146e-06,
      "loss": 0.82733583,
      "num_input_tokens_seen": 37075725,
      "step": 1746,
      "time_per_iteration": 2.5860376358032227
    },
    {
      "auxiliary_loss_clip": 0.01201971,
      "auxiliary_loss_mlp": 0.01034074,
      "balance_loss_clip": 1.05970824,
      "balance_loss_mlp": 1.02510393,
      "epoch": 0.2100643299464919,
      "flos": 21320955313920.0,
      "grad_norm": 2.0023847954338843,
      "language_loss": 0.81598699,
      "learning_rate": 3.669563260142321e-06,
      "loss": 0.83834743,
      "num_input_tokens_seen": 37094615,
      "step": 1747,
      "time_per_iteration": 2.5021493434906006
    },
    {
      "auxiliary_loss_clip": 0.01182758,
      "auxiliary_loss_mlp": 0.01038151,
      "balance_loss_clip": 1.06117511,
      "balance_loss_mlp": 1.02884698,
      "epoch": 0.21018457283713102,
      "flos": 19354379644800.0,
      "grad_norm": 1.9329789875094434,
      "language_loss": 0.84209371,
      "learning_rate": 3.6691342463927083e-06,
      "loss": 0.86430287,
      "num_input_tokens_seen": 37113610,
      "step": 1748,
      "time_per_iteration": 2.5173652172088623
    },
    {
      "auxiliary_loss_clip": 0.01174863,
      "auxiliary_loss_mlp": 0.01038985,
      "balance_loss_clip": 1.05768466,
      "balance_loss_mlp": 1.02957892,
      "epoch": 0.2103048157277701,
      "flos": 28330035914880.0,
      "grad_norm": 1.619241627535905,
      "language_loss": 0.81551063,
      "learning_rate": 3.668704979436758e-06,
      "loss": 0.83764911,
      "num_input_tokens_seen": 37133705,
      "step": 1749,
      "time_per_iteration": 2.626819610595703
    },
    {
      "auxiliary_loss_clip": 0.01176972,
      "auxiliary_loss_mlp": 0.01035693,
      "balance_loss_clip": 1.05405307,
      "balance_loss_mlp": 1.02638865,
      "epoch": 0.21042505861840918,
      "flos": 17457290835840.0,
      "grad_norm": 1.9597630624351245,
      "language_loss": 0.78709507,
      "learning_rate": 3.668275459339588e-06,
      "loss": 0.80922174,
      "num_input_tokens_seen": 37152185,
      "step": 1750,
      "time_per_iteration": 2.5133371353149414
    },
    {
      "auxiliary_loss_clip": 0.01216895,
      "auxiliary_loss_mlp": 0.01037369,
      "balance_loss_clip": 1.0639286,
      "balance_loss_mlp": 1.02714109,
      "epoch": 0.21054530150904827,
      "flos": 14209817195520.0,
      "grad_norm": 2.112450244642857,
      "language_loss": 0.80269277,
      "learning_rate": 3.667845686166358e-06,
      "loss": 0.82523543,
      "num_input_tokens_seen": 37169110,
      "step": 1751,
      "time_per_iteration": 2.430022716522217
    },
    {
      "auxiliary_loss_clip": 0.01153772,
      "auxiliary_loss_mlp": 0.01030117,
      "balance_loss_clip": 1.05169058,
      "balance_loss_mlp": 1.02035403,
      "epoch": 0.21066554439968738,
      "flos": 18618210403200.0,
      "grad_norm": 1.7866486114255284,
      "language_loss": 0.85896385,
      "learning_rate": 3.6674156599822634e-06,
      "loss": 0.88080275,
      "num_input_tokens_seen": 37184905,
      "step": 1752,
      "time_per_iteration": 3.3723833560943604
    },
    {
      "auxiliary_loss_clip": 0.01157478,
      "auxiliary_loss_mlp": 0.0103964,
      "balance_loss_clip": 1.0509603,
      "balance_loss_mlp": 1.02996576,
      "epoch": 0.21078578729032646,
      "flos": 23658883741440.0,
      "grad_norm": 2.2056655646680534,
      "language_loss": 0.81837642,
      "learning_rate": 3.666985380852539e-06,
      "loss": 0.84034765,
      "num_input_tokens_seen": 37203910,
      "step": 1753,
      "time_per_iteration": 2.600553512573242
    },
    {
      "auxiliary_loss_clip": 0.01185862,
      "auxiliary_loss_mlp": 0.0103255,
      "balance_loss_clip": 1.0602572,
      "balance_loss_mlp": 1.02281022,
      "epoch": 0.21090603018096554,
      "flos": 29346379240320.0,
      "grad_norm": 3.260311458547385,
      "language_loss": 0.74748254,
      "learning_rate": 3.6665548488424576e-06,
      "loss": 0.76966667,
      "num_input_tokens_seen": 37222670,
      "step": 1754,
      "time_per_iteration": 2.575789451599121
    },
    {
      "auxiliary_loss_clip": 0.01217189,
      "auxiliary_loss_mlp": 0.01037331,
      "balance_loss_clip": 1.06208384,
      "balance_loss_mlp": 1.02701926,
      "epoch": 0.21102627307160465,
      "flos": 23261245205760.0,
      "grad_norm": 2.0270978935915496,
      "language_loss": 0.88008004,
      "learning_rate": 3.6661240640173307e-06,
      "loss": 0.90262526,
      "num_input_tokens_seen": 37244140,
      "step": 1755,
      "time_per_iteration": 2.5638747215270996
    },
    {
      "auxiliary_loss_clip": 0.01083785,
      "auxiliary_loss_mlp": 0.01010325,
      "balance_loss_clip": 1.02759123,
      "balance_loss_mlp": 1.00809598,
      "epoch": 0.21114651596224374,
      "flos": 54633454577280.0,
      "grad_norm": 0.8889622053571035,
      "language_loss": 0.57912737,
      "learning_rate": 3.6656930264425085e-06,
      "loss": 0.60006845,
      "num_input_tokens_seen": 37308185,
      "step": 1756,
      "time_per_iteration": 3.1736364364624023
    },
    {
      "auxiliary_loss_clip": 0.01217146,
      "auxiliary_loss_mlp": 0.01037821,
      "balance_loss_clip": 1.06263375,
      "balance_loss_mlp": 1.02786684,
      "epoch": 0.21126675885288282,
      "flos": 21543314457600.0,
      "grad_norm": 1.7376468322884182,
      "language_loss": 0.75861591,
      "learning_rate": 3.665261736183378e-06,
      "loss": 0.7811656,
      "num_input_tokens_seen": 37328220,
      "step": 1757,
      "time_per_iteration": 2.507418394088745
    },
    {
      "auxiliary_loss_clip": 0.01172478,
      "auxiliary_loss_mlp": 0.01031661,
      "balance_loss_clip": 1.05848813,
      "balance_loss_mlp": 1.02159965,
      "epoch": 0.2113870017435219,
      "flos": 10961876678400.0,
      "grad_norm": 2.39046261252039,
      "language_loss": 0.89167386,
      "learning_rate": 3.664830193305366e-06,
      "loss": 0.91371524,
      "num_input_tokens_seen": 37345995,
      "step": 1758,
      "time_per_iteration": 4.308273792266846
    },
    {
      "auxiliary_loss_clip": 0.01166234,
      "auxiliary_loss_mlp": 0.01036465,
      "balance_loss_clip": 1.05290079,
      "balance_loss_mlp": 1.02668428,
      "epoch": 0.211507244634161,
      "flos": 16653825463680.0,
      "grad_norm": 6.170319221508419,
      "language_loss": 0.76876199,
      "learning_rate": 3.6643983978739373e-06,
      "loss": 0.79078901,
      "num_input_tokens_seen": 37362610,
      "step": 1759,
      "time_per_iteration": 3.3425939083099365
    },
    {
      "auxiliary_loss_clip": 0.01180341,
      "auxiliary_loss_mlp": 0.01033525,
      "balance_loss_clip": 1.05904484,
      "balance_loss_mlp": 1.02352321,
      "epoch": 0.2116274875248001,
      "flos": 20954091755520.0,
      "grad_norm": 1.9004249772441613,
      "language_loss": 0.8203029,
      "learning_rate": 3.663966349954596e-06,
      "loss": 0.84244156,
      "num_input_tokens_seen": 37382790,
      "step": 1760,
      "time_per_iteration": 2.5548155307769775
    },
    {
      "auxiliary_loss_clip": 0.01105109,
      "auxiliary_loss_mlp": 0.01001855,
      "balance_loss_clip": 1.02659369,
      "balance_loss_mlp": 0.99964952,
      "epoch": 0.21174773041543918,
      "flos": 68196949424640.0,
      "grad_norm": 0.7922156201813727,
      "language_loss": 0.59808755,
      "learning_rate": 3.6635340496128816e-06,
      "loss": 0.6191572,
      "num_input_tokens_seen": 37439720,
      "step": 1761,
      "time_per_iteration": 2.9759645462036133
    },
    {
      "auxiliary_loss_clip": 0.01154979,
      "auxiliary_loss_mlp": 0.01034496,
      "balance_loss_clip": 1.05480206,
      "balance_loss_mlp": 1.02534103,
      "epoch": 0.2118679733060783,
      "flos": 20668315150080.0,
      "grad_norm": 1.6493272588694963,
      "language_loss": 0.92545718,
      "learning_rate": 3.6631014969143747e-06,
      "loss": 0.94735187,
      "num_input_tokens_seen": 37459410,
      "step": 1762,
      "time_per_iteration": 2.632343292236328
    },
    {
      "auxiliary_loss_clip": 0.01206052,
      "auxiliary_loss_mlp": 0.01043486,
      "balance_loss_clip": 1.06574392,
      "balance_loss_mlp": 1.03408027,
      "epoch": 0.21198821619671737,
      "flos": 23223431162880.0,
      "grad_norm": 1.7015162526646155,
      "language_loss": 0.89152634,
      "learning_rate": 3.662668691924693e-06,
      "loss": 0.91402173,
      "num_input_tokens_seen": 37480460,
      "step": 1763,
      "time_per_iteration": 2.518904685974121
    },
    {
      "auxiliary_loss_clip": 0.01170215,
      "auxiliary_loss_mlp": 0.01039916,
      "balance_loss_clip": 1.05585933,
      "balance_loss_mlp": 1.02950943,
      "epoch": 0.21210845908735645,
      "flos": 24498547044480.0,
      "grad_norm": 1.944626822653477,
      "language_loss": 0.72100049,
      "learning_rate": 3.6622356347094927e-06,
      "loss": 0.74310178,
      "num_input_tokens_seen": 37502025,
      "step": 1764,
      "time_per_iteration": 2.589101791381836
    },
    {
      "auxiliary_loss_clip": 0.01171676,
      "auxiliary_loss_mlp": 0.01037009,
      "balance_loss_clip": 1.05284286,
      "balance_loss_mlp": 1.02623296,
      "epoch": 0.21222870197799554,
      "flos": 27089789160960.0,
      "grad_norm": 1.945570693918676,
      "language_loss": 0.7870971,
      "learning_rate": 3.6618023253344684e-06,
      "loss": 0.80918401,
      "num_input_tokens_seen": 37520885,
      "step": 1765,
      "time_per_iteration": 2.5923502445220947
    },
    {
      "auxiliary_loss_clip": 0.01200865,
      "auxiliary_loss_mlp": 0.01041109,
      "balance_loss_clip": 1.0592438,
      "balance_loss_mlp": 1.030792,
      "epoch": 0.21234894486863465,
      "flos": 16873850223360.0,
      "grad_norm": 1.4454107411192525,
      "language_loss": 0.83385426,
      "learning_rate": 3.6613687638653527e-06,
      "loss": 0.85627401,
      "num_input_tokens_seen": 37539055,
      "step": 1766,
      "time_per_iteration": 2.4639527797698975
    },
    {
      "auxiliary_loss_clip": 0.01181866,
      "auxiliary_loss_mlp": 0.01035126,
      "balance_loss_clip": 1.05760121,
      "balance_loss_mlp": 1.02518988,
      "epoch": 0.21246918775927373,
      "flos": 23474949171840.0,
      "grad_norm": 1.8126336215905998,
      "language_loss": 0.7757827,
      "learning_rate": 3.660934950367916e-06,
      "loss": 0.79795271,
      "num_input_tokens_seen": 37558300,
      "step": 1767,
      "time_per_iteration": 2.5510809421539307
    },
    {
      "auxiliary_loss_clip": 0.01204323,
      "auxiliary_loss_mlp": 0.01036614,
      "balance_loss_clip": 1.06220198,
      "balance_loss_mlp": 1.02671957,
      "epoch": 0.21258943064991281,
      "flos": 22382295402240.0,
      "grad_norm": 1.699834246609623,
      "language_loss": 0.83455169,
      "learning_rate": 3.660500884907968e-06,
      "loss": 0.85696107,
      "num_input_tokens_seen": 37579040,
      "step": 1768,
      "time_per_iteration": 2.5239593982696533
    },
    {
      "auxiliary_loss_clip": 0.01069117,
      "auxiliary_loss_mlp": 0.01003329,
      "balance_loss_clip": 1.02535892,
      "balance_loss_mlp": 1.00082552,
      "epoch": 0.21270967354055192,
      "flos": 59440168679040.0,
      "grad_norm": 0.8216218626058457,
      "language_loss": 0.60000062,
      "learning_rate": 3.660066567551356e-06,
      "loss": 0.6207251,
      "num_input_tokens_seen": 37639185,
      "step": 1769,
      "time_per_iteration": 3.0549306869506836
    },
    {
      "auxiliary_loss_clip": 0.01200479,
      "auxiliary_loss_mlp": 0.00765161,
      "balance_loss_clip": 1.06048334,
      "balance_loss_mlp": 1.00098419,
      "epoch": 0.212829916431191,
      "flos": 21544032729600.0,
      "grad_norm": 2.228823779654595,
      "language_loss": 0.84556282,
      "learning_rate": 3.6596319983639657e-06,
      "loss": 0.86521918,
      "num_input_tokens_seen": 37657765,
      "step": 1770,
      "time_per_iteration": 2.515789747238159
    },
    {
      "auxiliary_loss_clip": 0.0117351,
      "auxiliary_loss_mlp": 0.00765912,
      "balance_loss_clip": 1.05885124,
      "balance_loss_mlp": 1.00096118,
      "epoch": 0.2129501593218301,
      "flos": 28987739896320.0,
      "grad_norm": 1.550211201985683,
      "language_loss": 0.86196232,
      "learning_rate": 3.6591971774117214e-06,
      "loss": 0.88135654,
      "num_input_tokens_seen": 37680740,
      "step": 1771,
      "time_per_iteration": 2.6453745365142822
    },
    {
      "auxiliary_loss_clip": 0.01207496,
      "auxiliary_loss_mlp": 0.01040615,
      "balance_loss_clip": 1.0632937,
      "balance_loss_mlp": 1.03048182,
      "epoch": 0.2130704022124692,
      "flos": 18806993308800.0,
      "grad_norm": 2.059605367534018,
      "language_loss": 0.80518997,
      "learning_rate": 3.6587621047605833e-06,
      "loss": 0.82767105,
      "num_input_tokens_seen": 37697910,
      "step": 1772,
      "time_per_iteration": 2.4984118938446045
    },
    {
      "auxiliary_loss_clip": 0.0120295,
      "auxiliary_loss_mlp": 0.01039117,
      "balance_loss_clip": 1.06244361,
      "balance_loss_mlp": 1.02975905,
      "epoch": 0.21319064510310828,
      "flos": 13918150759680.0,
      "grad_norm": 1.9476953152537653,
      "language_loss": 0.86659396,
      "learning_rate": 3.6583267804765542e-06,
      "loss": 0.8890146,
      "num_input_tokens_seen": 37712245,
      "step": 1773,
      "time_per_iteration": 2.5339155197143555
    },
    {
      "auxiliary_loss_clip": 0.01198622,
      "auxiliary_loss_mlp": 0.01037087,
      "balance_loss_clip": 1.05881619,
      "balance_loss_mlp": 1.02641821,
      "epoch": 0.21331088799374737,
      "flos": 20959694277120.0,
      "grad_norm": 1.685265008435033,
      "language_loss": 0.85512364,
      "learning_rate": 3.6578912046256702e-06,
      "loss": 0.87748063,
      "num_input_tokens_seen": 37730765,
      "step": 1774,
      "time_per_iteration": 2.5187294483184814
    },
    {
      "auxiliary_loss_clip": 0.0116821,
      "auxiliary_loss_mlp": 0.01033041,
      "balance_loss_clip": 1.05329263,
      "balance_loss_mlp": 1.02243781,
      "epoch": 0.21343113088438645,
      "flos": 18624638937600.0,
      "grad_norm": 1.95408503356112,
      "language_loss": 0.75892329,
      "learning_rate": 3.6574553772740083e-06,
      "loss": 0.78093576,
      "num_input_tokens_seen": 37748695,
      "step": 1775,
      "time_per_iteration": 2.541170597076416
    },
    {
      "auxiliary_loss_clip": 0.01094954,
      "auxiliary_loss_mlp": 0.01002763,
      "balance_loss_clip": 1.0269798,
      "balance_loss_mlp": 1.00051045,
      "epoch": 0.21355137377502556,
      "flos": 67413128791680.0,
      "grad_norm": 0.8493705330258402,
      "language_loss": 0.61886555,
      "learning_rate": 3.657019298487684e-06,
      "loss": 0.63984269,
      "num_input_tokens_seen": 37813705,
      "step": 1776,
      "time_per_iteration": 3.1203174591064453
    },
    {
      "auxiliary_loss_clip": 0.01210159,
      "auxiliary_loss_mlp": 0.0076584,
      "balance_loss_clip": 1.06136084,
      "balance_loss_mlp": 1.00101912,
      "epoch": 0.21367161666566464,
      "flos": 34532095697280.0,
      "grad_norm": 2.6303326203727138,
      "language_loss": 0.83458847,
      "learning_rate": 3.6565829683328495e-06,
      "loss": 0.85434848,
      "num_input_tokens_seen": 37836330,
      "step": 1777,
      "time_per_iteration": 2.6514554023742676
    },
    {
      "auxiliary_loss_clip": 0.01196245,
      "auxiliary_loss_mlp": 0.0103759,
      "balance_loss_clip": 1.05905044,
      "balance_loss_mlp": 1.02734363,
      "epoch": 0.21379185955630373,
      "flos": 18989347680000.0,
      "grad_norm": 1.6819733295980346,
      "language_loss": 0.85998476,
      "learning_rate": 3.6561463868756965e-06,
      "loss": 0.88232303,
      "num_input_tokens_seen": 37855030,
      "step": 1778,
      "time_per_iteration": 2.454549551010132
    },
    {
      "auxiliary_loss_clip": 0.01202991,
      "auxiliary_loss_mlp": 0.01036865,
      "balance_loss_clip": 1.0637095,
      "balance_loss_mlp": 1.0266782,
      "epoch": 0.21391210244694284,
      "flos": 28218497207040.0,
      "grad_norm": 1.7194331359769421,
      "language_loss": 0.78008235,
      "learning_rate": 3.655709554182452e-06,
      "loss": 0.80248088,
      "num_input_tokens_seen": 37875370,
      "step": 1779,
      "time_per_iteration": 3.3246679306030273
    },
    {
      "auxiliary_loss_clip": 0.01204383,
      "auxiliary_loss_mlp": 0.01033702,
      "balance_loss_clip": 1.05911446,
      "balance_loss_mlp": 1.02421927,
      "epoch": 0.21403234533758192,
      "flos": 17455064192640.0,
      "grad_norm": 2.5177189526327997,
      "language_loss": 0.8467648,
      "learning_rate": 3.6552724703193855e-06,
      "loss": 0.86914563,
      "num_input_tokens_seen": 37892560,
      "step": 1780,
      "time_per_iteration": 2.4566287994384766
    },
    {
      "auxiliary_loss_clip": 0.01063446,
      "auxiliary_loss_mlp": 0.01008573,
      "balance_loss_clip": 1.02203,
      "balance_loss_mlp": 1.00595033,
      "epoch": 0.214152588228221,
      "flos": 51637606686720.0,
      "grad_norm": 0.7924905630335375,
      "language_loss": 0.55932772,
      "learning_rate": 3.654835135352801e-06,
      "loss": 0.58004791,
      "num_input_tokens_seen": 37947370,
      "step": 1781,
      "time_per_iteration": 3.0200095176696777
    },
    {
      "auxiliary_loss_clip": 0.01154921,
      "auxiliary_loss_mlp": 0.01032253,
      "balance_loss_clip": 1.05036926,
      "balance_loss_mlp": 1.02206063,
      "epoch": 0.21427283111886009,
      "flos": 19496154625920.0,
      "grad_norm": 1.7512825860459846,
      "language_loss": 0.87445223,
      "learning_rate": 3.654397549349043e-06,
      "loss": 0.89632404,
      "num_input_tokens_seen": 37964745,
      "step": 1782,
      "time_per_iteration": 2.579237699508667
    },
    {
      "auxiliary_loss_clip": 0.01185015,
      "auxiliary_loss_mlp": 0.01034363,
      "balance_loss_clip": 1.06101263,
      "balance_loss_mlp": 1.0240159,
      "epoch": 0.2143930740094992,
      "flos": 20084802710400.0,
      "grad_norm": 2.2447810221344535,
      "language_loss": 0.75453168,
      "learning_rate": 3.653959712374491e-06,
      "loss": 0.77672553,
      "num_input_tokens_seen": 37982850,
      "step": 1783,
      "time_per_iteration": 2.521242141723633
    },
    {
      "auxiliary_loss_clip": 0.01166413,
      "auxiliary_loss_mlp": 0.01026669,
      "balance_loss_clip": 1.05910063,
      "balance_loss_mlp": 1.01706624,
      "epoch": 0.21451331690013828,
      "flos": 21798603394560.0,
      "grad_norm": 1.5610972534755763,
      "language_loss": 0.82655156,
      "learning_rate": 3.6535216244955663e-06,
      "loss": 0.84848237,
      "num_input_tokens_seen": 38002745,
      "step": 1784,
      "time_per_iteration": 3.4302234649658203
    },
    {
      "auxiliary_loss_clip": 0.01185738,
      "auxiliary_loss_mlp": 0.01037906,
      "balance_loss_clip": 1.05997849,
      "balance_loss_mlp": 1.02786827,
      "epoch": 0.21463355979077736,
      "flos": 32853882412800.0,
      "grad_norm": 1.6081607400772948,
      "language_loss": 0.70927966,
      "learning_rate": 3.653083285778726e-06,
      "loss": 0.73151606,
      "num_input_tokens_seen": 38024115,
      "step": 1785,
      "time_per_iteration": 3.5211477279663086
    },
    {
      "auxiliary_loss_clip": 0.01206595,
      "auxiliary_loss_mlp": 0.01032194,
      "balance_loss_clip": 1.06041694,
      "balance_loss_mlp": 1.02140558,
      "epoch": 0.21475380268141647,
      "flos": 21543817248000.0,
      "grad_norm": 1.9384667437333827,
      "language_loss": 0.811643,
      "learning_rate": 3.6526446962904653e-06,
      "loss": 0.83403093,
      "num_input_tokens_seen": 38042830,
      "step": 1786,
      "time_per_iteration": 3.3201074600219727
    },
    {
      "auxiliary_loss_clip": 0.01197391,
      "auxiliary_loss_mlp": 0.0104066,
      "balance_loss_clip": 1.06145024,
      "balance_loss_mlp": 1.03102171,
      "epoch": 0.21487404557205556,
      "flos": 32159082660480.0,
      "grad_norm": 1.513096369347035,
      "language_loss": 0.74209732,
      "learning_rate": 3.652205856097318e-06,
      "loss": 0.76447779,
      "num_input_tokens_seen": 38066015,
      "step": 1787,
      "time_per_iteration": 2.612100124359131
    },
    {
      "auxiliary_loss_clip": 0.01180835,
      "auxiliary_loss_mlp": 0.00765009,
      "balance_loss_clip": 1.05796349,
      "balance_loss_mlp": 1.00116944,
      "epoch": 0.21499428846269464,
      "flos": 12673091583360.0,
      "grad_norm": 1.8831345578204703,
      "language_loss": 0.79402864,
      "learning_rate": 3.651766765265856e-06,
      "loss": 0.81348705,
      "num_input_tokens_seen": 38083025,
      "step": 1788,
      "time_per_iteration": 2.5232691764831543
    },
    {
      "auxiliary_loss_clip": 0.01180869,
      "auxiliary_loss_mlp": 0.01026411,
      "balance_loss_clip": 1.05589414,
      "balance_loss_mlp": 1.01684487,
      "epoch": 0.21511453135333372,
      "flos": 23471573293440.0,
      "grad_norm": 2.041857349656563,
      "language_loss": 0.81488442,
      "learning_rate": 3.65132742386269e-06,
      "loss": 0.83695722,
      "num_input_tokens_seen": 38098245,
      "step": 1789,
      "time_per_iteration": 2.5078086853027344
    },
    {
      "auxiliary_loss_clip": 0.01214362,
      "auxiliary_loss_mlp": 0.0103262,
      "balance_loss_clip": 1.05942988,
      "balance_loss_mlp": 1.02215934,
      "epoch": 0.21523477424397283,
      "flos": 26943560893440.0,
      "grad_norm": 1.662498619104773,
      "language_loss": 0.84748101,
      "learning_rate": 3.6508878319544656e-06,
      "loss": 0.86995089,
      "num_input_tokens_seen": 38118460,
      "step": 1790,
      "time_per_iteration": 2.4993860721588135
    },
    {
      "auxiliary_loss_clip": 0.0117633,
      "auxiliary_loss_mlp": 0.01045007,
      "balance_loss_clip": 1.05865979,
      "balance_loss_mlp": 1.03489256,
      "epoch": 0.21535501713461191,
      "flos": 18916161719040.0,
      "grad_norm": 2.563312368127213,
      "language_loss": 0.81922615,
      "learning_rate": 3.65044798960787e-06,
      "loss": 0.84143949,
      "num_input_tokens_seen": 38136800,
      "step": 1791,
      "time_per_iteration": 2.5290913581848145
    },
    {
      "auxiliary_loss_clip": 0.01164793,
      "auxiliary_loss_mlp": 0.01031974,
      "balance_loss_clip": 1.05520153,
      "balance_loss_mlp": 1.02253819,
      "epoch": 0.215475260025251,
      "flos": 17895113712000.0,
      "grad_norm": 1.6737757317399367,
      "language_loss": 0.78327453,
      "learning_rate": 3.650007896889627e-06,
      "loss": 0.80524218,
      "num_input_tokens_seen": 38155380,
      "step": 1792,
      "time_per_iteration": 2.5355114936828613
    },
    {
      "auxiliary_loss_clip": 0.01214479,
      "auxiliary_loss_mlp": 0.01039045,
      "balance_loss_clip": 1.06281829,
      "balance_loss_mlp": 1.02914512,
      "epoch": 0.2155955029158901,
      "flos": 16654292340480.0,
      "grad_norm": 2.1883727892734024,
      "language_loss": 0.80680132,
      "learning_rate": 3.6495675538664974e-06,
      "loss": 0.82933658,
      "num_input_tokens_seen": 38174395,
      "step": 1793,
      "time_per_iteration": 2.457888126373291
    },
    {
      "auxiliary_loss_clip": 0.01185336,
      "auxiliary_loss_mlp": 0.01034756,
      "balance_loss_clip": 1.05595183,
      "balance_loss_mlp": 1.02495694,
      "epoch": 0.2157157458065292,
      "flos": 23621213352960.0,
      "grad_norm": 1.7575446136644943,
      "language_loss": 0.82871157,
      "learning_rate": 3.649126960605282e-06,
      "loss": 0.85091251,
      "num_input_tokens_seen": 38195380,
      "step": 1794,
      "time_per_iteration": 2.544405221939087
    },
    {
      "auxiliary_loss_clip": 0.01182188,
      "auxiliary_loss_mlp": 0.01033776,
      "balance_loss_clip": 1.05798399,
      "balance_loss_mlp": 1.02397752,
      "epoch": 0.21583598869716827,
      "flos": 22127078292480.0,
      "grad_norm": 2.3241727635455027,
      "language_loss": 0.83531022,
      "learning_rate": 3.6486861171728174e-06,
      "loss": 0.85746992,
      "num_input_tokens_seen": 38213775,
      "step": 1795,
      "time_per_iteration": 2.516519546508789
    },
    {
      "auxiliary_loss_clip": 0.01167489,
      "auxiliary_loss_mlp": 0.01029691,
      "balance_loss_clip": 1.05154955,
      "balance_loss_mlp": 1.01995194,
      "epoch": 0.21595623158780738,
      "flos": 23441229279360.0,
      "grad_norm": 1.6675853419057018,
      "language_loss": 0.7864151,
      "learning_rate": 3.6482450236359803e-06,
      "loss": 0.80838692,
      "num_input_tokens_seen": 38235630,
      "step": 1796,
      "time_per_iteration": 2.6114914417266846
    },
    {
      "auxiliary_loss_clip": 0.01198251,
      "auxiliary_loss_mlp": 0.01039043,
      "balance_loss_clip": 1.06090307,
      "balance_loss_mlp": 1.02982223,
      "epoch": 0.21607647447844647,
      "flos": 26906501036160.0,
      "grad_norm": 2.642848341706185,
      "language_loss": 0.77897894,
      "learning_rate": 3.647803680061683e-06,
      "loss": 0.80135179,
      "num_input_tokens_seen": 38256045,
      "step": 1797,
      "time_per_iteration": 2.5336523056030273
    },
    {
      "auxiliary_loss_clip": 0.01186294,
      "auxiliary_loss_mlp": 0.01036353,
      "balance_loss_clip": 1.05849743,
      "balance_loss_mlp": 1.02542746,
      "epoch": 0.21619671736908555,
      "flos": 14495378319360.0,
      "grad_norm": 2.4854600263610647,
      "language_loss": 0.74615562,
      "learning_rate": 3.6473620865168776e-06,
      "loss": 0.76838207,
      "num_input_tokens_seen": 38272915,
      "step": 1798,
      "time_per_iteration": 2.494514226913452
    },
    {
      "auxiliary_loss_clip": 0.01184594,
      "auxiliary_loss_mlp": 0.0103556,
      "balance_loss_clip": 1.0608511,
      "balance_loss_mlp": 1.02644014,
      "epoch": 0.21631696025972463,
      "flos": 17931096161280.0,
      "grad_norm": 1.9309158295459796,
      "language_loss": 0.81498837,
      "learning_rate": 3.646920243068554e-06,
      "loss": 0.83718991,
      "num_input_tokens_seen": 38290810,
      "step": 1799,
      "time_per_iteration": 2.5032858848571777
    },
    {
      "auxiliary_loss_clip": 0.01167867,
      "auxiliary_loss_mlp": 0.01034563,
      "balance_loss_clip": 1.05375743,
      "balance_loss_mlp": 1.02534819,
      "epoch": 0.21643720315036374,
      "flos": 24462385027200.0,
      "grad_norm": 1.5824022339318022,
      "language_loss": 0.74510413,
      "learning_rate": 3.6464781497837384e-06,
      "loss": 0.76712847,
      "num_input_tokens_seen": 38312785,
      "step": 1800,
      "time_per_iteration": 2.5370969772338867
    },
    {
      "auxiliary_loss_clip": 0.0118437,
      "auxiliary_loss_mlp": 0.01045752,
      "balance_loss_clip": 1.05401719,
      "balance_loss_mlp": 1.03638232,
      "epoch": 0.21655744604100283,
      "flos": 28474432588800.0,
      "grad_norm": 1.8600297324915298,
      "language_loss": 0.72813314,
      "learning_rate": 3.6460358067294965e-06,
      "loss": 0.75043434,
      "num_input_tokens_seen": 38334015,
      "step": 1801,
      "time_per_iteration": 2.576519727706909
    },
    {
      "auxiliary_loss_clip": 0.01216647,
      "auxiliary_loss_mlp": 0.01030826,
      "balance_loss_clip": 1.06006932,
      "balance_loss_mlp": 1.02083635,
      "epoch": 0.2166776889316419,
      "flos": 20152960767360.0,
      "grad_norm": 1.955194711043629,
      "language_loss": 0.78240514,
      "learning_rate": 3.645593213972932e-06,
      "loss": 0.80487984,
      "num_input_tokens_seen": 38352920,
      "step": 1802,
      "time_per_iteration": 2.4722437858581543
    },
    {
      "auxiliary_loss_clip": 0.01194934,
      "auxiliary_loss_mlp": 0.01032185,
      "balance_loss_clip": 1.05827296,
      "balance_loss_mlp": 1.02185571,
      "epoch": 0.21679793182228102,
      "flos": 15193482122880.0,
      "grad_norm": 2.605012971886413,
      "language_loss": 0.79298192,
      "learning_rate": 3.6451503715811852e-06,
      "loss": 0.81525314,
      "num_input_tokens_seen": 38371230,
      "step": 1803,
      "time_per_iteration": 2.4734320640563965
    },
    {
      "auxiliary_loss_clip": 0.01182008,
      "auxiliary_loss_mlp": 0.01033363,
      "balance_loss_clip": 1.05973446,
      "balance_loss_mlp": 1.0253222,
      "epoch": 0.2169181747129201,
      "flos": 17384464010880.0,
      "grad_norm": 1.9668271057411717,
      "language_loss": 0.80384636,
      "learning_rate": 3.6447072796214345e-06,
      "loss": 0.82600009,
      "num_input_tokens_seen": 38389795,
      "step": 1804,
      "time_per_iteration": 2.4981882572174072
    },
    {
      "auxiliary_loss_clip": 0.01065626,
      "auxiliary_loss_mlp": 0.01005924,
      "balance_loss_clip": 1.02624571,
      "balance_loss_mlp": 1.00354028,
      "epoch": 0.21703841760355919,
      "flos": 58760955429120.0,
      "grad_norm": 0.9554364486366573,
      "language_loss": 0.63303971,
      "learning_rate": 3.644263938160898e-06,
      "loss": 0.65375525,
      "num_input_tokens_seen": 38445760,
      "step": 1805,
      "time_per_iteration": 3.0463926792144775
    },
    {
      "auxiliary_loss_clip": 0.01166202,
      "auxiliary_loss_mlp": 0.01031654,
      "balance_loss_clip": 1.05590272,
      "balance_loss_mlp": 1.02158642,
      "epoch": 0.21715866049419827,
      "flos": 22418457419520.0,
      "grad_norm": 1.8374340320280655,
      "language_loss": 0.72122657,
      "learning_rate": 3.6438203472668293e-06,
      "loss": 0.74320513,
      "num_input_tokens_seen": 38465405,
      "step": 1806,
      "time_per_iteration": 3.380117177963257
    },
    {
      "auxiliary_loss_clip": 0.01186358,
      "auxiliary_loss_mlp": 0.01034154,
      "balance_loss_clip": 1.05735385,
      "balance_loss_mlp": 1.0243547,
      "epoch": 0.21727890338483738,
      "flos": 17237732952960.0,
      "grad_norm": 1.9070799122138213,
      "language_loss": 0.82056355,
      "learning_rate": 3.6433765070065206e-06,
      "loss": 0.84276867,
      "num_input_tokens_seen": 38483195,
      "step": 1807,
      "time_per_iteration": 2.499148368835449
    },
    {
      "auxiliary_loss_clip": 0.01213145,
      "auxiliary_loss_mlp": 0.01031619,
      "balance_loss_clip": 1.05952454,
      "balance_loss_mlp": 1.02148652,
      "epoch": 0.21739914627547646,
      "flos": 13434792416640.0,
      "grad_norm": 2.5394212463227146,
      "language_loss": 0.87596506,
      "learning_rate": 3.6429324174473025e-06,
      "loss": 0.8984127,
      "num_input_tokens_seen": 38496735,
      "step": 1808,
      "time_per_iteration": 2.4230000972747803
    },
    {
      "auxiliary_loss_clip": 0.01199368,
      "auxiliary_loss_mlp": 0.01035993,
      "balance_loss_clip": 1.05679536,
      "balance_loss_mlp": 1.02656329,
      "epoch": 0.21751938916611555,
      "flos": 20959514709120.0,
      "grad_norm": 1.9673286826152745,
      "language_loss": 0.84808159,
      "learning_rate": 3.6424880786565425e-06,
      "loss": 0.87043512,
      "num_input_tokens_seen": 38512880,
      "step": 1809,
      "time_per_iteration": 2.478933095932007
    },
    {
      "auxiliary_loss_clip": 0.01152118,
      "auxiliary_loss_mlp": 0.01037819,
      "balance_loss_clip": 1.05597627,
      "balance_loss_mlp": 1.0272212,
      "epoch": 0.21763963205675466,
      "flos": 27599936071680.0,
      "grad_norm": 2.2724983028377452,
      "language_loss": 0.79833269,
      "learning_rate": 3.6420434907016482e-06,
      "loss": 0.82023209,
      "num_input_tokens_seen": 38532570,
      "step": 1810,
      "time_per_iteration": 3.4623312950134277
    },
    {
      "auxiliary_loss_clip": 0.01200193,
      "auxiliary_loss_mlp": 0.01033967,
      "balance_loss_clip": 1.06220508,
      "balance_loss_mlp": 1.02466893,
      "epoch": 0.21775987494739374,
      "flos": 21430411032960.0,
      "grad_norm": 1.5040179017112,
      "language_loss": 0.8115114,
      "learning_rate": 3.6415986536500606e-06,
      "loss": 0.83385295,
      "num_input_tokens_seen": 38550900,
      "step": 1811,
      "time_per_iteration": 2.485686779022217
    },
    {
      "auxiliary_loss_clip": 0.01147597,
      "auxiliary_loss_mlp": 0.01038427,
      "balance_loss_clip": 1.05770361,
      "balance_loss_mlp": 1.02853298,
      "epoch": 0.21788011783803282,
      "flos": 18332972501760.0,
      "grad_norm": 6.0255268028168105,
      "language_loss": 0.80824983,
      "learning_rate": 3.641153567569263e-06,
      "loss": 0.83011007,
      "num_input_tokens_seen": 38569215,
      "step": 1812,
      "time_per_iteration": 3.4030940532684326
    },
    {
      "auxiliary_loss_clip": 0.01194622,
      "auxiliary_loss_mlp": 0.0102929,
      "balance_loss_clip": 1.05840492,
      "balance_loss_mlp": 1.01991379,
      "epoch": 0.2180003607286719,
      "flos": 30262748037120.0,
      "grad_norm": 2.333828781563822,
      "language_loss": 0.95938945,
      "learning_rate": 3.640708232526774e-06,
      "loss": 0.9816286,
      "num_input_tokens_seen": 38587870,
      "step": 1813,
      "time_per_iteration": 3.4120779037475586
    },
    {
      "auxiliary_loss_clip": 0.01132549,
      "auxiliary_loss_mlp": 0.0103344,
      "balance_loss_clip": 1.04485655,
      "balance_loss_mlp": 1.02314019,
      "epoch": 0.21812060361931102,
      "flos": 25480272637440.0,
      "grad_norm": 4.50554017108526,
      "language_loss": 0.78354979,
      "learning_rate": 3.6402626485901504e-06,
      "loss": 0.80520976,
      "num_input_tokens_seen": 38606965,
      "step": 1814,
      "time_per_iteration": 2.6621203422546387
    },
    {
      "auxiliary_loss_clip": 0.01194347,
      "auxiliary_loss_mlp": 0.0103815,
      "balance_loss_clip": 1.06085503,
      "balance_loss_mlp": 1.02885103,
      "epoch": 0.2182408465099501,
      "flos": 21908166854400.0,
      "grad_norm": 1.935069136220931,
      "language_loss": 0.77921939,
      "learning_rate": 3.639816815826988e-06,
      "loss": 0.80154437,
      "num_input_tokens_seen": 38626290,
      "step": 1815,
      "time_per_iteration": 2.5036466121673584
    },
    {
      "auxiliary_loss_clip": 0.01181054,
      "auxiliary_loss_mlp": 0.01028961,
      "balance_loss_clip": 1.0582943,
      "balance_loss_mlp": 1.01971006,
      "epoch": 0.21836108940058918,
      "flos": 23657339456640.0,
      "grad_norm": 1.7389176893223532,
      "language_loss": 0.77971947,
      "learning_rate": 3.6393707343049176e-06,
      "loss": 0.80181956,
      "num_input_tokens_seen": 38646620,
      "step": 1816,
      "time_per_iteration": 2.55485463142395
    },
    {
      "auxiliary_loss_clip": 0.01199444,
      "auxiliary_loss_mlp": 0.01029475,
      "balance_loss_clip": 1.0571363,
      "balance_loss_mlp": 1.02026582,
      "epoch": 0.2184813322912283,
      "flos": 24681009156480.0,
      "grad_norm": 2.750800225248675,
      "language_loss": 0.73751456,
      "learning_rate": 3.6389244040916104e-06,
      "loss": 0.75980377,
      "num_input_tokens_seen": 38665695,
      "step": 1817,
      "time_per_iteration": 2.543424606323242
    },
    {
      "auxiliary_loss_clip": 0.01175243,
      "auxiliary_loss_mlp": 0.00765681,
      "balance_loss_clip": 1.0555681,
      "balance_loss_mlp": 1.00089371,
      "epoch": 0.21860157518186737,
      "flos": 26574650259840.0,
      "grad_norm": 1.9858860385453883,
      "language_loss": 0.79217559,
      "learning_rate": 3.6384778252547747e-06,
      "loss": 0.81158483,
      "num_input_tokens_seen": 38681575,
      "step": 1818,
      "time_per_iteration": 2.546870231628418
    },
    {
      "auxiliary_loss_clip": 0.01179903,
      "auxiliary_loss_mlp": 0.00764651,
      "balance_loss_clip": 1.06010067,
      "balance_loss_mlp": 1.0009644,
      "epoch": 0.21872181807250646,
      "flos": 20886292834560.0,
      "grad_norm": 2.633498003300965,
      "language_loss": 0.78427494,
      "learning_rate": 3.638030997862155e-06,
      "loss": 0.80372047,
      "num_input_tokens_seen": 38700510,
      "step": 1819,
      "time_per_iteration": 2.5029172897338867
    },
    {
      "auxiliary_loss_clip": 0.01089882,
      "auxiliary_loss_mlp": 0.01008806,
      "balance_loss_clip": 1.02916503,
      "balance_loss_mlp": 1.00705338,
      "epoch": 0.21884206096314554,
      "flos": 61209452897280.0,
      "grad_norm": 0.7692780539040929,
      "language_loss": 0.59501928,
      "learning_rate": 3.6375839219815356e-06,
      "loss": 0.61600614,
      "num_input_tokens_seen": 38758310,
      "step": 1820,
      "time_per_iteration": 3.018749475479126
    },
    {
      "auxiliary_loss_clip": 0.0121308,
      "auxiliary_loss_mlp": 0.01031777,
      "balance_loss_clip": 1.06106544,
      "balance_loss_mlp": 1.02185249,
      "epoch": 0.21896230385378465,
      "flos": 23473835850240.0,
      "grad_norm": 1.9937184411250723,
      "language_loss": 0.82753086,
      "learning_rate": 3.6371365976807375e-06,
      "loss": 0.8499794,
      "num_input_tokens_seen": 38778705,
      "step": 1821,
      "time_per_iteration": 2.4649085998535156
    },
    {
      "auxiliary_loss_clip": 0.0114489,
      "auxiliary_loss_mlp": 0.01029548,
      "balance_loss_clip": 1.05391288,
      "balance_loss_mlp": 1.02007675,
      "epoch": 0.21908254674442373,
      "flos": 25081915829760.0,
      "grad_norm": 1.700915811783782,
      "language_loss": 0.84071362,
      "learning_rate": 3.6366890250276185e-06,
      "loss": 0.86245799,
      "num_input_tokens_seen": 38799660,
      "step": 1822,
      "time_per_iteration": 2.6104955673217773
    },
    {
      "auxiliary_loss_clip": 0.01212311,
      "auxiliary_loss_mlp": 0.01030653,
      "balance_loss_clip": 1.06049335,
      "balance_loss_mlp": 1.02105069,
      "epoch": 0.21920278963506282,
      "flos": 23513768795520.0,
      "grad_norm": 1.8820216773477807,
      "language_loss": 0.89986455,
      "learning_rate": 3.6362412040900764e-06,
      "loss": 0.9222942,
      "num_input_tokens_seen": 38819450,
      "step": 1823,
      "time_per_iteration": 2.475553274154663
    },
    {
      "auxiliary_loss_clip": 0.01199665,
      "auxiliary_loss_mlp": 0.01032477,
      "balance_loss_clip": 1.05791688,
      "balance_loss_mlp": 1.02274978,
      "epoch": 0.21932303252570193,
      "flos": 29242238734080.0,
      "grad_norm": 2.0366168150927515,
      "language_loss": 0.80926651,
      "learning_rate": 3.635793134936044e-06,
      "loss": 0.83158791,
      "num_input_tokens_seen": 38840460,
      "step": 1824,
      "time_per_iteration": 2.543588161468506
    },
    {
      "auxiliary_loss_clip": 0.01196323,
      "auxiliary_loss_mlp": 0.01036387,
      "balance_loss_clip": 1.05972886,
      "balance_loss_mlp": 1.02686214,
      "epoch": 0.219443275416341,
      "flos": 20806857907200.0,
      "grad_norm": 2.055728139162839,
      "language_loss": 0.73196483,
      "learning_rate": 3.635344817633494e-06,
      "loss": 0.75429189,
      "num_input_tokens_seen": 38859775,
      "step": 1825,
      "time_per_iteration": 2.595139265060425
    },
    {
      "auxiliary_loss_clip": 0.01192514,
      "auxiliary_loss_mlp": 0.01032715,
      "balance_loss_clip": 1.05707002,
      "balance_loss_mlp": 1.02314222,
      "epoch": 0.2195635183069801,
      "flos": 14501555458560.0,
      "grad_norm": 2.067874691579874,
      "language_loss": 0.75541836,
      "learning_rate": 3.634896252250436e-06,
      "loss": 0.77767074,
      "num_input_tokens_seen": 38876540,
      "step": 1826,
      "time_per_iteration": 2.4944369792938232
    },
    {
      "auxiliary_loss_clip": 0.01217211,
      "auxiliary_loss_mlp": 0.01042603,
      "balance_loss_clip": 1.06392622,
      "balance_loss_mlp": 1.03305995,
      "epoch": 0.2196837611976192,
      "flos": 24243473589120.0,
      "grad_norm": 1.9677866078938475,
      "language_loss": 0.82004106,
      "learning_rate": 3.6344474388549157e-06,
      "loss": 0.84263921,
      "num_input_tokens_seen": 38896195,
      "step": 1827,
      "time_per_iteration": 2.4724934101104736
    },
    {
      "auxiliary_loss_clip": 0.01203914,
      "auxiliary_loss_mlp": 0.01040499,
      "balance_loss_clip": 1.06372714,
      "balance_loss_mlp": 1.0298661,
      "epoch": 0.2198040040882583,
      "flos": 18074523168000.0,
      "grad_norm": 2.095655207348312,
      "language_loss": 0.80106449,
      "learning_rate": 3.6339983775150183e-06,
      "loss": 0.82350862,
      "num_input_tokens_seen": 38912755,
      "step": 1828,
      "time_per_iteration": 2.479191780090332
    },
    {
      "auxiliary_loss_clip": 0.01195874,
      "auxiliary_loss_mlp": 0.01030235,
      "balance_loss_clip": 1.06006789,
      "balance_loss_mlp": 1.02009618,
      "epoch": 0.21992424697889737,
      "flos": 17784185535360.0,
      "grad_norm": 2.5605731210981166,
      "language_loss": 0.83987868,
      "learning_rate": 3.6335490682988664e-06,
      "loss": 0.86213982,
      "num_input_tokens_seen": 38928365,
      "step": 1829,
      "time_per_iteration": 2.4454495906829834
    },
    {
      "auxiliary_loss_clip": 0.01130761,
      "auxiliary_loss_mlp": 0.0103396,
      "balance_loss_clip": 1.04988146,
      "balance_loss_mlp": 1.02451229,
      "epoch": 0.22004448986953645,
      "flos": 17638495971840.0,
      "grad_norm": 2.016747118862944,
      "language_loss": 0.82983798,
      "learning_rate": 3.63309951127462e-06,
      "loss": 0.85148519,
      "num_input_tokens_seen": 38945275,
      "step": 1830,
      "time_per_iteration": 2.604417324066162
    },
    {
      "auxiliary_loss_clip": 0.0116918,
      "auxiliary_loss_mlp": 0.01041439,
      "balance_loss_clip": 1.05855632,
      "balance_loss_mlp": 1.03133583,
      "epoch": 0.22016473276017556,
      "flos": 22275533203200.0,
      "grad_norm": 2.07099195142222,
      "language_loss": 0.75454307,
      "learning_rate": 3.6326497065104757e-06,
      "loss": 0.7766493,
      "num_input_tokens_seen": 38965740,
      "step": 1831,
      "time_per_iteration": 2.576331377029419
    },
    {
      "auxiliary_loss_clip": 0.01203776,
      "auxiliary_loss_mlp": 0.01032998,
      "balance_loss_clip": 1.06062341,
      "balance_loss_mlp": 1.0234791,
      "epoch": 0.22028497565081465,
      "flos": 25556259859200.0,
      "grad_norm": 1.962192022577612,
      "language_loss": 0.77963573,
      "learning_rate": 3.6321996540746697e-06,
      "loss": 0.8020035,
      "num_input_tokens_seen": 38984815,
      "step": 1832,
      "time_per_iteration": 3.343738555908203
    },
    {
      "auxiliary_loss_clip": 0.01168782,
      "auxiliary_loss_mlp": 0.01033061,
      "balance_loss_clip": 1.05751157,
      "balance_loss_mlp": 1.02288043,
      "epoch": 0.22040521854145373,
      "flos": 36247332925440.0,
      "grad_norm": 1.7843214336478292,
      "language_loss": 0.80798113,
      "learning_rate": 3.6317493540354733e-06,
      "loss": 0.82999957,
      "num_input_tokens_seen": 39008230,
      "step": 1833,
      "time_per_iteration": 2.691223621368408
    },
    {
      "auxiliary_loss_clip": 0.01196108,
      "auxiliary_loss_mlp": 0.01038222,
      "balance_loss_clip": 1.05848503,
      "balance_loss_mlp": 1.02853072,
      "epoch": 0.22052546143209284,
      "flos": 11838420270720.0,
      "grad_norm": 2.1075454989375526,
      "language_loss": 0.76603472,
      "learning_rate": 3.6312988064611976e-06,
      "loss": 0.788378,
      "num_input_tokens_seen": 39026540,
      "step": 1834,
      "time_per_iteration": 2.503697395324707
    },
    {
      "auxiliary_loss_clip": 0.01170335,
      "auxiliary_loss_mlp": 0.01037744,
      "balance_loss_clip": 1.05327177,
      "balance_loss_mlp": 1.02775407,
      "epoch": 0.22064570432273192,
      "flos": 24209250906240.0,
      "grad_norm": 1.7980106022524054,
      "language_loss": 0.81579375,
      "learning_rate": 3.6308480114201896e-06,
      "loss": 0.83787453,
      "num_input_tokens_seen": 39048460,
      "step": 1835,
      "time_per_iteration": 2.586886167526245
    },
    {
      "auxiliary_loss_clip": 0.01218161,
      "auxiliary_loss_mlp": 0.01039778,
      "balance_loss_clip": 1.06628919,
      "balance_loss_mlp": 1.02968121,
      "epoch": 0.220765947213371,
      "flos": 17931347556480.0,
      "grad_norm": 1.9866853062646816,
      "language_loss": 0.76503837,
      "learning_rate": 3.630396968980835e-06,
      "loss": 0.78761774,
      "num_input_tokens_seen": 39066335,
      "step": 1836,
      "time_per_iteration": 2.4501953125
    },
    {
      "auxiliary_loss_clip": 0.0118653,
      "auxiliary_loss_mlp": 0.01036977,
      "balance_loss_clip": 1.05829692,
      "balance_loss_mlp": 1.02668285,
      "epoch": 0.2208861901040101,
      "flos": 26757040544640.0,
      "grad_norm": 2.4257323987347736,
      "language_loss": 0.83769405,
      "learning_rate": 3.6299456792115575e-06,
      "loss": 0.85992903,
      "num_input_tokens_seen": 39087590,
      "step": 1837,
      "time_per_iteration": 3.375005006790161
    },
    {
      "auxiliary_loss_clip": 0.01110082,
      "auxiliary_loss_mlp": 0.01031275,
      "balance_loss_clip": 1.0461539,
      "balance_loss_mlp": 1.02144611,
      "epoch": 0.2210064329946492,
      "flos": 17817977255040.0,
      "grad_norm": 1.867138953688265,
      "language_loss": 0.81080282,
      "learning_rate": 3.629494142180815e-06,
      "loss": 0.83221638,
      "num_input_tokens_seen": 39106335,
      "step": 1838,
      "time_per_iteration": 3.5625452995300293
    },
    {
      "auxiliary_loss_clip": 0.01214356,
      "auxiliary_loss_mlp": 0.01033185,
      "balance_loss_clip": 1.06248856,
      "balance_loss_mlp": 1.02335024,
      "epoch": 0.22112667588528828,
      "flos": 17967401832960.0,
      "grad_norm": 2.4137200429991235,
      "language_loss": 0.8487736,
      "learning_rate": 3.6290423579571075e-06,
      "loss": 0.87124902,
      "num_input_tokens_seen": 39122875,
      "step": 1839,
      "time_per_iteration": 2.429676055908203
    },
    {
      "auxiliary_loss_clip": 0.0119767,
      "auxiliary_loss_mlp": 0.01033447,
      "balance_loss_clip": 1.06198418,
      "balance_loss_mlp": 1.02320123,
      "epoch": 0.22124691877592736,
      "flos": 18369206346240.0,
      "grad_norm": 1.5888362504762383,
      "language_loss": 0.80414015,
      "learning_rate": 3.6285903266089694e-06,
      "loss": 0.8264513,
      "num_input_tokens_seen": 39142150,
      "step": 1840,
      "time_per_iteration": 3.2666099071502686
    },
    {
      "auxiliary_loss_clip": 0.01188929,
      "auxiliary_loss_mlp": 0.01031047,
      "balance_loss_clip": 1.06119275,
      "balance_loss_mlp": 1.02119398,
      "epoch": 0.22136716166656648,
      "flos": 20813286441600.0,
      "grad_norm": 2.277933352628288,
      "language_loss": 0.77550155,
      "learning_rate": 3.628138048204974e-06,
      "loss": 0.79770124,
      "num_input_tokens_seen": 39162835,
      "step": 1841,
      "time_per_iteration": 2.5117993354797363
    },
    {
      "auxiliary_loss_clip": 0.01147157,
      "auxiliary_loss_mlp": 0.01035456,
      "balance_loss_clip": 1.05471373,
      "balance_loss_mlp": 1.02466178,
      "epoch": 0.22148740455720556,
      "flos": 17675699483520.0,
      "grad_norm": 1.7526688298215332,
      "language_loss": 0.75941789,
      "learning_rate": 3.6276855228137304e-06,
      "loss": 0.78124404,
      "num_input_tokens_seen": 39181040,
      "step": 1842,
      "time_per_iteration": 2.5626485347747803
    },
    {
      "auxiliary_loss_clip": 0.01215864,
      "auxiliary_loss_mlp": 0.00765363,
      "balance_loss_clip": 1.06377769,
      "balance_loss_mlp": 1.0009048,
      "epoch": 0.22160764744784464,
      "flos": 21726710323200.0,
      "grad_norm": 2.1723645940308165,
      "language_loss": 0.81869686,
      "learning_rate": 3.6272327505038874e-06,
      "loss": 0.83850914,
      "num_input_tokens_seen": 39197505,
      "step": 1843,
      "time_per_iteration": 2.474837303161621
    },
    {
      "auxiliary_loss_clip": 0.01158895,
      "auxiliary_loss_mlp": 0.01036233,
      "balance_loss_clip": 1.0536046,
      "balance_loss_mlp": 1.02705932,
      "epoch": 0.22172789033848372,
      "flos": 23764712186880.0,
      "grad_norm": 2.0657982927518783,
      "language_loss": 0.78608894,
      "learning_rate": 3.626779731344131e-06,
      "loss": 0.80804026,
      "num_input_tokens_seen": 39217295,
      "step": 1844,
      "time_per_iteration": 2.614419460296631
    },
    {
      "auxiliary_loss_clip": 0.01209412,
      "auxiliary_loss_mlp": 0.01039411,
      "balance_loss_clip": 1.06075335,
      "balance_loss_mlp": 1.03011298,
      "epoch": 0.22184813322912283,
      "flos": 16982300361600.0,
      "grad_norm": 1.8979379168343127,
      "language_loss": 0.85162294,
      "learning_rate": 3.6263264654031814e-06,
      "loss": 0.87411118,
      "num_input_tokens_seen": 39234195,
      "step": 1845,
      "time_per_iteration": 2.429943323135376
    },
    {
      "auxiliary_loss_clip": 0.0107815,
      "auxiliary_loss_mlp": 0.01002226,
      "balance_loss_clip": 1.02922654,
      "balance_loss_mlp": 1.00043821,
      "epoch": 0.22196837611976192,
      "flos": 61823740314240.0,
      "grad_norm": 0.7080897769965405,
      "language_loss": 0.59222591,
      "learning_rate": 3.6258729527498008e-06,
      "loss": 0.61302972,
      "num_input_tokens_seen": 39295040,
      "step": 1846,
      "time_per_iteration": 3.091163396835327
    },
    {
      "auxiliary_loss_clip": 0.01191191,
      "auxiliary_loss_mlp": 0.01031038,
      "balance_loss_clip": 1.06260633,
      "balance_loss_mlp": 1.02148342,
      "epoch": 0.222088619010401,
      "flos": 25558019625600.0,
      "grad_norm": 2.234913187081375,
      "language_loss": 0.64814746,
      "learning_rate": 3.6254191934527854e-06,
      "loss": 0.6703698,
      "num_input_tokens_seen": 39314395,
      "step": 1847,
      "time_per_iteration": 2.6546897888183594
    },
    {
      "auxiliary_loss_clip": 0.0116668,
      "auxiliary_loss_mlp": 0.0103203,
      "balance_loss_clip": 1.05999601,
      "balance_loss_mlp": 1.02168822,
      "epoch": 0.2222088619010401,
      "flos": 19318612677120.0,
      "grad_norm": 1.9880423692090115,
      "language_loss": 0.65021938,
      "learning_rate": 3.6249651875809715e-06,
      "loss": 0.67220646,
      "num_input_tokens_seen": 39334275,
      "step": 1848,
      "time_per_iteration": 2.595595598220825
    },
    {
      "auxiliary_loss_clip": 0.01179336,
      "auxiliary_loss_mlp": 0.01031689,
      "balance_loss_clip": 1.06053615,
      "balance_loss_mlp": 1.02221775,
      "epoch": 0.2223291047916792,
      "flos": 19099342103040.0,
      "grad_norm": 1.8900897185309564,
      "language_loss": 0.89523858,
      "learning_rate": 3.62451093520323e-06,
      "loss": 0.91734886,
      "num_input_tokens_seen": 39352180,
      "step": 1849,
      "time_per_iteration": 2.5012753009796143
    },
    {
      "auxiliary_loss_clip": 0.01151103,
      "auxiliary_loss_mlp": 0.01043861,
      "balance_loss_clip": 1.0532589,
      "balance_loss_mlp": 1.0345273,
      "epoch": 0.22244934768231828,
      "flos": 20850418126080.0,
      "grad_norm": 2.192864299218851,
      "language_loss": 0.90520245,
      "learning_rate": 3.6240564363884714e-06,
      "loss": 0.9271521,
      "num_input_tokens_seen": 39372125,
      "step": 1850,
      "time_per_iteration": 2.60660982131958
    },
    {
      "auxiliary_loss_clip": 0.01201212,
      "auxiliary_loss_mlp": 0.01038678,
      "balance_loss_clip": 1.05873311,
      "balance_loss_mlp": 1.02897382,
      "epoch": 0.2225695905729574,
      "flos": 15632921111040.0,
      "grad_norm": 2.1501431983204053,
      "language_loss": 0.70448303,
      "learning_rate": 3.623601691205643e-06,
      "loss": 0.72688192,
      "num_input_tokens_seen": 39391200,
      "step": 1851,
      "time_per_iteration": 2.502091646194458
    },
    {
      "auxiliary_loss_clip": 0.01196985,
      "auxiliary_loss_mlp": 0.01029312,
      "balance_loss_clip": 1.05827117,
      "balance_loss_mlp": 1.02013314,
      "epoch": 0.22268983346359647,
      "flos": 25373582265600.0,
      "grad_norm": 2.0433813144031587,
      "language_loss": 0.81309092,
      "learning_rate": 3.623146699723729e-06,
      "loss": 0.83535391,
      "num_input_tokens_seen": 39410660,
      "step": 1852,
      "time_per_iteration": 2.553537368774414
    },
    {
      "auxiliary_loss_clip": 0.01186827,
      "auxiliary_loss_mlp": 0.01037897,
      "balance_loss_clip": 1.06475699,
      "balance_loss_mlp": 1.02832437,
      "epoch": 0.22281007635423555,
      "flos": 13261452359040.0,
      "grad_norm": 1.7434932059917718,
      "language_loss": 0.77765834,
      "learning_rate": 3.6226914620117507e-06,
      "loss": 0.7999056,
      "num_input_tokens_seen": 39429280,
      "step": 1853,
      "time_per_iteration": 2.5166845321655273
    },
    {
      "auxiliary_loss_clip": 0.01171565,
      "auxiliary_loss_mlp": 0.01031864,
      "balance_loss_clip": 1.05404162,
      "balance_loss_mlp": 1.02259493,
      "epoch": 0.22293031924487464,
      "flos": 15340536403200.0,
      "grad_norm": 2.448947588991763,
      "language_loss": 0.81192482,
      "learning_rate": 3.622235978138768e-06,
      "loss": 0.8339591,
      "num_input_tokens_seen": 39446905,
      "step": 1854,
      "time_per_iteration": 2.531911849975586
    },
    {
      "auxiliary_loss_clip": 0.01200147,
      "auxiliary_loss_mlp": 0.01030829,
      "balance_loss_clip": 1.06383705,
      "balance_loss_mlp": 1.02135217,
      "epoch": 0.22305056213551375,
      "flos": 22564649773440.0,
      "grad_norm": 1.8666837020135476,
      "language_loss": 0.81472456,
      "learning_rate": 3.621780248173877e-06,
      "loss": 0.83703429,
      "num_input_tokens_seen": 39465105,
      "step": 1855,
      "time_per_iteration": 2.485137701034546
    },
    {
      "auxiliary_loss_clip": 0.01105582,
      "auxiliary_loss_mlp": 0.01004052,
      "balance_loss_clip": 1.028512,
      "balance_loss_mlp": 1.00253248,
      "epoch": 0.22317080502615283,
      "flos": 64880419887360.0,
      "grad_norm": 0.8282101929430018,
      "language_loss": 0.61029661,
      "learning_rate": 3.6213242721862125e-06,
      "loss": 0.63139296,
      "num_input_tokens_seen": 39523560,
      "step": 1856,
      "time_per_iteration": 3.068829298019409
    },
    {
      "auxiliary_loss_clip": 0.01173778,
      "auxiliary_loss_mlp": 0.01035931,
      "balance_loss_clip": 1.0582335,
      "balance_loss_mlp": 1.02625108,
      "epoch": 0.2232910479167919,
      "flos": 25775997310080.0,
      "grad_norm": 1.588392971160899,
      "language_loss": 0.75256741,
      "learning_rate": 3.620868050244945e-06,
      "loss": 0.77466452,
      "num_input_tokens_seen": 39544040,
      "step": 1857,
      "time_per_iteration": 2.5508782863616943
    },
    {
      "auxiliary_loss_clip": 0.01178659,
      "auxiliary_loss_mlp": 0.01030533,
      "balance_loss_clip": 1.05777502,
      "balance_loss_mlp": 1.02074552,
      "epoch": 0.22341129080743102,
      "flos": 23251799928960.0,
      "grad_norm": 1.7773498125523781,
      "language_loss": 0.7767967,
      "learning_rate": 3.6204115824192817e-06,
      "loss": 0.79888856,
      "num_input_tokens_seen": 39561515,
      "step": 1858,
      "time_per_iteration": 2.5358359813690186
    },
    {
      "auxiliary_loss_clip": 0.01173929,
      "auxiliary_loss_mlp": 0.01032516,
      "balance_loss_clip": 1.05437028,
      "balance_loss_mlp": 1.02241302,
      "epoch": 0.2235315336980701,
      "flos": 21214552250880.0,
      "grad_norm": 2.57898043621557,
      "language_loss": 0.76530206,
      "learning_rate": 3.619954868778471e-06,
      "loss": 0.78736651,
      "num_input_tokens_seen": 39578210,
      "step": 1859,
      "time_per_iteration": 3.29205584526062
    },
    {
      "auxiliary_loss_clip": 0.01183414,
      "auxiliary_loss_mlp": 0.01034268,
      "balance_loss_clip": 1.05726385,
      "balance_loss_mlp": 1.02543473,
      "epoch": 0.2236517765887092,
      "flos": 19901945548800.0,
      "grad_norm": 1.922907339410487,
      "language_loss": 0.82973999,
      "learning_rate": 3.6194979093917944e-06,
      "loss": 0.85191673,
      "num_input_tokens_seen": 39597625,
      "step": 1860,
      "time_per_iteration": 2.516542673110962
    },
    {
      "auxiliary_loss_clip": 0.01179317,
      "auxiliary_loss_mlp": 0.01037151,
      "balance_loss_clip": 1.05759752,
      "balance_loss_mlp": 1.02783465,
      "epoch": 0.22377201947934827,
      "flos": 23214847812480.0,
      "grad_norm": 1.8880782064336612,
      "language_loss": 0.87007153,
      "learning_rate": 3.6190407043285724e-06,
      "loss": 0.89223617,
      "num_input_tokens_seen": 39615360,
      "step": 1861,
      "time_per_iteration": 2.525892734527588
    },
    {
      "auxiliary_loss_clip": 0.01216472,
      "auxiliary_loss_mlp": 0.01034519,
      "balance_loss_clip": 1.06313014,
      "balance_loss_mlp": 1.02469587,
      "epoch": 0.22389226236998738,
      "flos": 26794244056320.0,
      "grad_norm": 1.8270058990913984,
      "language_loss": 0.75750405,
      "learning_rate": 3.618583253658163e-06,
      "loss": 0.78001392,
      "num_input_tokens_seen": 39635460,
      "step": 1862,
      "time_per_iteration": 2.4903433322906494
    },
    {
      "auxiliary_loss_clip": 0.0115348,
      "auxiliary_loss_mlp": 0.0076543,
      "balance_loss_clip": 1.05478442,
      "balance_loss_mlp": 1.00098884,
      "epoch": 0.22401250526062647,
      "flos": 24170359455360.0,
      "grad_norm": 2.0531138651682936,
      "language_loss": 0.86383408,
      "learning_rate": 3.618125557449961e-06,
      "loss": 0.88302314,
      "num_input_tokens_seen": 39653515,
      "step": 1863,
      "time_per_iteration": 2.628727436065674
    },
    {
      "auxiliary_loss_clip": 0.01195902,
      "auxiliary_loss_mlp": 0.01032405,
      "balance_loss_clip": 1.06027055,
      "balance_loss_mlp": 1.02274323,
      "epoch": 0.22413274815126555,
      "flos": 16759761649920.0,
      "grad_norm": 1.9826821231827214,
      "language_loss": 0.83092076,
      "learning_rate": 3.6176676157733983e-06,
      "loss": 0.85320383,
      "num_input_tokens_seen": 39668525,
      "step": 1864,
      "time_per_iteration": 3.395414113998413
    },
    {
      "auxiliary_loss_clip": 0.01163012,
      "auxiliary_loss_mlp": 0.0103444,
      "balance_loss_clip": 1.0552299,
      "balance_loss_mlp": 1.02475476,
      "epoch": 0.22425299104190466,
      "flos": 21360205900800.0,
      "grad_norm": 1.9861608042584011,
      "language_loss": 0.75771934,
      "learning_rate": 3.6172094286979443e-06,
      "loss": 0.77969384,
      "num_input_tokens_seen": 39685895,
      "step": 1865,
      "time_per_iteration": 2.538931131362915
    },
    {
      "auxiliary_loss_clip": 0.01180527,
      "auxiliary_loss_mlp": 0.01033329,
      "balance_loss_clip": 1.05447745,
      "balance_loss_mlp": 1.02385747,
      "epoch": 0.22437323393254374,
      "flos": 32165547108480.0,
      "grad_norm": 1.4542573198902833,
      "language_loss": 0.81347823,
      "learning_rate": 3.6167509962931064e-06,
      "loss": 0.83561677,
      "num_input_tokens_seen": 39711595,
      "step": 1866,
      "time_per_iteration": 3.361229181289673
    },
    {
      "auxiliary_loss_clip": 0.011636,
      "auxiliary_loss_mlp": 0.01033082,
      "balance_loss_clip": 1.05907941,
      "balance_loss_mlp": 1.02331305,
      "epoch": 0.22449347682318282,
      "flos": 18002809664640.0,
      "grad_norm": 2.425195390624701,
      "language_loss": 0.77168375,
      "learning_rate": 3.6162923186284276e-06,
      "loss": 0.79365057,
      "num_input_tokens_seen": 39727555,
      "step": 1867,
      "time_per_iteration": 2.542787790298462
    },
    {
      "auxiliary_loss_clip": 0.01184134,
      "auxiliary_loss_mlp": 0.01038992,
      "balance_loss_clip": 1.05755353,
      "balance_loss_mlp": 1.02940118,
      "epoch": 0.2246137197138219,
      "flos": 18697286194560.0,
      "grad_norm": 1.871388549069376,
      "language_loss": 0.85816681,
      "learning_rate": 3.6158333957734888e-06,
      "loss": 0.88039809,
      "num_input_tokens_seen": 39746145,
      "step": 1868,
      "time_per_iteration": 2.494091272354126
    },
    {
      "auxiliary_loss_clip": 0.01169593,
      "auxiliary_loss_mlp": 0.01033391,
      "balance_loss_clip": 1.0536294,
      "balance_loss_mlp": 1.02390838,
      "epoch": 0.22473396260446102,
      "flos": 15590653781760.0,
      "grad_norm": 2.5066638802733663,
      "language_loss": 0.82652211,
      "learning_rate": 3.6153742277979088e-06,
      "loss": 0.84855193,
      "num_input_tokens_seen": 39763575,
      "step": 1869,
      "time_per_iteration": 2.527780055999756
    },
    {
      "auxiliary_loss_clip": 0.01186785,
      "auxiliary_loss_mlp": 0.01038545,
      "balance_loss_clip": 1.05839658,
      "balance_loss_mlp": 1.02910352,
      "epoch": 0.2248542054951001,
      "flos": 14465501182080.0,
      "grad_norm": 1.8966546830711624,
      "language_loss": 0.78301626,
      "learning_rate": 3.6149148147713434e-06,
      "loss": 0.8052696,
      "num_input_tokens_seen": 39781810,
      "step": 1870,
      "time_per_iteration": 2.492715358734131
    },
    {
      "auxiliary_loss_clip": 0.01205259,
      "auxiliary_loss_mlp": 0.01036571,
      "balance_loss_clip": 1.06516886,
      "balance_loss_mlp": 1.02749372,
      "epoch": 0.22497444838573918,
      "flos": 19243882431360.0,
      "grad_norm": 1.948123388163964,
      "language_loss": 0.86649078,
      "learning_rate": 3.614455156763484e-06,
      "loss": 0.8889091,
      "num_input_tokens_seen": 39800115,
      "step": 1871,
      "time_per_iteration": 2.4802510738372803
    },
    {
      "auxiliary_loss_clip": 0.01147589,
      "auxiliary_loss_mlp": 0.01034369,
      "balance_loss_clip": 1.0484823,
      "balance_loss_mlp": 1.02514207,
      "epoch": 0.2250946912763783,
      "flos": 16910299549440.0,
      "grad_norm": 1.9783770824194185,
      "language_loss": 0.71753198,
      "learning_rate": 3.613995253844061e-06,
      "loss": 0.73935157,
      "num_input_tokens_seen": 39817795,
      "step": 1872,
      "time_per_iteration": 2.5490310192108154
    },
    {
      "auxiliary_loss_clip": 0.011987,
      "auxiliary_loss_mlp": 0.01037336,
      "balance_loss_clip": 1.06279373,
      "balance_loss_mlp": 1.02771521,
      "epoch": 0.22521493416701738,
      "flos": 24681368292480.0,
      "grad_norm": 1.8430744053092842,
      "language_loss": 0.81225878,
      "learning_rate": 3.6135351060828414e-06,
      "loss": 0.83461916,
      "num_input_tokens_seen": 39838270,
      "step": 1873,
      "time_per_iteration": 2.5368502140045166
    },
    {
      "auxiliary_loss_clip": 0.01220939,
      "auxiliary_loss_mlp": 0.01041575,
      "balance_loss_clip": 1.06565356,
      "balance_loss_mlp": 1.0313648,
      "epoch": 0.22533517705765646,
      "flos": 17821963664640.0,
      "grad_norm": 2.4485399220804536,
      "language_loss": 0.69037896,
      "learning_rate": 3.6130747135496285e-06,
      "loss": 0.71300411,
      "num_input_tokens_seen": 39857270,
      "step": 1874,
      "time_per_iteration": 2.5364906787872314
    },
    {
      "auxiliary_loss_clip": 0.0121332,
      "auxiliary_loss_mlp": 0.01033476,
      "balance_loss_clip": 1.06186152,
      "balance_loss_mlp": 1.02355158,
      "epoch": 0.22545541994829554,
      "flos": 33691390899840.0,
      "grad_norm": 1.912430356470123,
      "language_loss": 0.65819371,
      "learning_rate": 3.6126140763142646e-06,
      "loss": 0.68066162,
      "num_input_tokens_seen": 39882300,
      "step": 1875,
      "time_per_iteration": 2.5601909160614014
    },
    {
      "auxiliary_loss_clip": 0.01215015,
      "auxiliary_loss_mlp": 0.01039081,
      "balance_loss_clip": 1.06363487,
      "balance_loss_mlp": 1.02875769,
      "epoch": 0.22557566283893465,
      "flos": 19171594310400.0,
      "grad_norm": 2.4741014409101187,
      "language_loss": 0.85806561,
      "learning_rate": 3.6121531944466275e-06,
      "loss": 0.88060653,
      "num_input_tokens_seen": 39899625,
      "step": 1876,
      "time_per_iteration": 2.439235210418701
    },
    {
      "auxiliary_loss_clip": 0.01195966,
      "auxiliary_loss_mlp": 0.01037557,
      "balance_loss_clip": 1.06039047,
      "balance_loss_mlp": 1.02843761,
      "epoch": 0.22569590572957374,
      "flos": 20773281669120.0,
      "grad_norm": 2.0766529801959623,
      "language_loss": 0.78550196,
      "learning_rate": 3.611692068016633e-06,
      "loss": 0.80783719,
      "num_input_tokens_seen": 39915955,
      "step": 1877,
      "time_per_iteration": 2.5173697471618652
    },
    {
      "auxiliary_loss_clip": 0.01160749,
      "auxiliary_loss_mlp": 0.01040672,
      "balance_loss_clip": 1.05057383,
      "balance_loss_mlp": 1.03011644,
      "epoch": 0.22581614862021282,
      "flos": 18442715529600.0,
      "grad_norm": 2.0774131567164624,
      "language_loss": 0.74898779,
      "learning_rate": 3.611230697094233e-06,
      "loss": 0.77100205,
      "num_input_tokens_seen": 39932655,
      "step": 1878,
      "time_per_iteration": 2.56384015083313
    },
    {
      "auxiliary_loss_clip": 0.01185986,
      "auxiliary_loss_mlp": 0.01037462,
      "balance_loss_clip": 1.05694747,
      "balance_loss_mlp": 1.02830684,
      "epoch": 0.22593639151085193,
      "flos": 20048389297920.0,
      "grad_norm": 1.7982894101729847,
      "language_loss": 0.86887002,
      "learning_rate": 3.6107690817494173e-06,
      "loss": 0.89110446,
      "num_input_tokens_seen": 39952875,
      "step": 1879,
      "time_per_iteration": 2.598787307739258
    },
    {
      "auxiliary_loss_clip": 0.01148995,
      "auxiliary_loss_mlp": 0.01032976,
      "balance_loss_clip": 1.05222559,
      "balance_loss_mlp": 1.02327812,
      "epoch": 0.226056634401491,
      "flos": 13115116350720.0,
      "grad_norm": 2.589123386647954,
      "language_loss": 0.70996881,
      "learning_rate": 3.6103072220522117e-06,
      "loss": 0.73178852,
      "num_input_tokens_seen": 39968405,
      "step": 1880,
      "time_per_iteration": 2.5793960094451904
    },
    {
      "auxiliary_loss_clip": 0.01172021,
      "auxiliary_loss_mlp": 0.010378,
      "balance_loss_clip": 1.05532455,
      "balance_loss_mlp": 1.02825689,
      "epoch": 0.2261768772921301,
      "flos": 18988378012800.0,
      "grad_norm": 1.8618163450802068,
      "language_loss": 0.91884744,
      "learning_rate": 3.609845118072682e-06,
      "loss": 0.94094574,
      "num_input_tokens_seen": 39987075,
      "step": 1881,
      "time_per_iteration": 2.562549352645874
    },
    {
      "auxiliary_loss_clip": 0.01204044,
      "auxiliary_loss_mlp": 0.00765513,
      "balance_loss_clip": 1.06053567,
      "balance_loss_mlp": 1.00106442,
      "epoch": 0.2262971201827692,
      "flos": 19974054101760.0,
      "grad_norm": 1.8250034757698619,
      "language_loss": 0.79947442,
      "learning_rate": 3.6093827698809276e-06,
      "loss": 0.81917,
      "num_input_tokens_seen": 40006175,
      "step": 1882,
      "time_per_iteration": 2.489743947982788
    },
    {
      "auxiliary_loss_clip": 0.0119492,
      "auxiliary_loss_mlp": 0.01037179,
      "balance_loss_clip": 1.0564754,
      "balance_loss_mlp": 1.02767777,
      "epoch": 0.2264173630734083,
      "flos": 16654543735680.0,
      "grad_norm": 2.136312271798255,
      "language_loss": 0.84983349,
      "learning_rate": 3.6089201775470864e-06,
      "loss": 0.87215436,
      "num_input_tokens_seen": 40021630,
      "step": 1883,
      "time_per_iteration": 2.4470057487487793
    },
    {
      "auxiliary_loss_clip": 0.01158182,
      "auxiliary_loss_mlp": 0.01036651,
      "balance_loss_clip": 1.05583787,
      "balance_loss_mlp": 1.02687633,
      "epoch": 0.22653760596404737,
      "flos": 24389809597440.0,
      "grad_norm": 1.3768009123846234,
      "language_loss": 0.77395582,
      "learning_rate": 3.6084573411413334e-06,
      "loss": 0.79590416,
      "num_input_tokens_seen": 40041025,
      "step": 1884,
      "time_per_iteration": 2.5785439014434814
    },
    {
      "auxiliary_loss_clip": 0.01169279,
      "auxiliary_loss_mlp": 0.01040842,
      "balance_loss_clip": 1.05641735,
      "balance_loss_mlp": 1.03001213,
      "epoch": 0.22665784885468646,
      "flos": 18332541538560.0,
      "grad_norm": 1.9760189842238627,
      "language_loss": 0.80994856,
      "learning_rate": 3.607994260733881e-06,
      "loss": 0.83204985,
      "num_input_tokens_seen": 40060265,
      "step": 1885,
      "time_per_iteration": 3.3323817253112793
    },
    {
      "auxiliary_loss_clip": 0.01187462,
      "auxiliary_loss_mlp": 0.01031074,
      "balance_loss_clip": 1.05670381,
      "balance_loss_mlp": 1.02208531,
      "epoch": 0.22677809174532557,
      "flos": 24058102475520.0,
      "grad_norm": 1.634145050080432,
      "language_loss": 0.74713755,
      "learning_rate": 3.6075309363949776e-06,
      "loss": 0.76932299,
      "num_input_tokens_seen": 40079435,
      "step": 1886,
      "time_per_iteration": 2.514387845993042
    },
    {
      "auxiliary_loss_clip": 0.01213582,
      "auxiliary_loss_mlp": 0.0103372,
      "balance_loss_clip": 1.06082606,
      "balance_loss_mlp": 1.02393341,
      "epoch": 0.22689833463596465,
      "flos": 20374242503040.0,
      "grad_norm": 2.282287208111359,
      "language_loss": 0.81096125,
      "learning_rate": 3.6070673681949094e-06,
      "loss": 0.83343434,
      "num_input_tokens_seen": 40097800,
      "step": 1887,
      "time_per_iteration": 2.4556827545166016
    },
    {
      "auxiliary_loss_clip": 0.01185518,
      "auxiliary_loss_mlp": 0.00765134,
      "balance_loss_clip": 1.0596745,
      "balance_loss_mlp": 1.00110686,
      "epoch": 0.22701857752660373,
      "flos": 30120398438400.0,
      "grad_norm": 1.8463247542872159,
      "language_loss": 0.8146311,
      "learning_rate": 3.606603556203999e-06,
      "loss": 0.83413768,
      "num_input_tokens_seen": 40122745,
      "step": 1888,
      "time_per_iteration": 2.601454973220825
    },
    {
      "auxiliary_loss_clip": 0.01197506,
      "auxiliary_loss_mlp": 0.0103553,
      "balance_loss_clip": 1.05647182,
      "balance_loss_mlp": 1.02602863,
      "epoch": 0.22713882041724284,
      "flos": 22492182084480.0,
      "grad_norm": 1.811119368682901,
      "language_loss": 0.83463424,
      "learning_rate": 3.6061395004926066e-06,
      "loss": 0.85696459,
      "num_input_tokens_seen": 40141680,
      "step": 1889,
      "time_per_iteration": 2.480898141860962
    },
    {
      "auxiliary_loss_clip": 0.01213199,
      "auxiliary_loss_mlp": 0.01034164,
      "balance_loss_clip": 1.0600574,
      "balance_loss_mlp": 1.024544,
      "epoch": 0.22725906330788193,
      "flos": 20521548178560.0,
      "grad_norm": 1.9479741659128826,
      "language_loss": 0.84772599,
      "learning_rate": 3.605675201131129e-06,
      "loss": 0.87019956,
      "num_input_tokens_seen": 40160140,
      "step": 1890,
      "time_per_iteration": 2.4434306621551514
    },
    {
      "auxiliary_loss_clip": 0.01206992,
      "auxiliary_loss_mlp": 0.01033835,
      "balance_loss_clip": 1.06376541,
      "balance_loss_mlp": 1.02487016,
      "epoch": 0.227379306198521,
      "flos": 18989922297600.0,
      "grad_norm": 2.244446726712968,
      "language_loss": 0.79414457,
      "learning_rate": 3.60521065819e-06,
      "loss": 0.81655288,
      "num_input_tokens_seen": 40177450,
      "step": 1891,
      "time_per_iteration": 3.3893744945526123
    },
    {
      "auxiliary_loss_clip": 0.01186334,
      "auxiliary_loss_mlp": 0.01032853,
      "balance_loss_clip": 1.05716181,
      "balance_loss_mlp": 1.02424002,
      "epoch": 0.2274995490891601,
      "flos": 21798351999360.0,
      "grad_norm": 1.7231169556552874,
      "language_loss": 0.87688828,
      "learning_rate": 3.60474587173969e-06,
      "loss": 0.89908016,
      "num_input_tokens_seen": 40195935,
      "step": 1892,
      "time_per_iteration": 3.260382890701294
    },
    {
      "auxiliary_loss_clip": 0.01197128,
      "auxiliary_loss_mlp": 0.01035994,
      "balance_loss_clip": 1.0621841,
      "balance_loss_mlp": 1.02658868,
      "epoch": 0.2276197919797992,
      "flos": 19058654972160.0,
      "grad_norm": 1.9786237873562795,
      "language_loss": 0.84173048,
      "learning_rate": 3.6042808418507084e-06,
      "loss": 0.86406171,
      "num_input_tokens_seen": 40213620,
      "step": 1893,
      "time_per_iteration": 2.4510369300842285
    },
    {
      "auxiliary_loss_clip": 0.01200645,
      "auxiliary_loss_mlp": 0.01037345,
      "balance_loss_clip": 1.062029,
      "balance_loss_mlp": 1.02753949,
      "epoch": 0.22774003487043828,
      "flos": 18806777827200.0,
      "grad_norm": 3.832560721133269,
      "language_loss": 0.76831335,
      "learning_rate": 3.6038155685935976e-06,
      "loss": 0.79069322,
      "num_input_tokens_seen": 40230190,
      "step": 1894,
      "time_per_iteration": 2.454312562942505
    },
    {
      "auxiliary_loss_clip": 0.01195333,
      "auxiliary_loss_mlp": 0.01033483,
      "balance_loss_clip": 1.05904293,
      "balance_loss_mlp": 1.02455401,
      "epoch": 0.22786027776107737,
      "flos": 23002544476800.0,
      "grad_norm": 1.8478021553382742,
      "language_loss": 0.70841503,
      "learning_rate": 3.6033500520389404e-06,
      "loss": 0.73070323,
      "num_input_tokens_seen": 40246860,
      "step": 1895,
      "time_per_iteration": 2.4760398864746094
    },
    {
      "auxiliary_loss_clip": 0.01070673,
      "auxiliary_loss_mlp": 0.01006922,
      "balance_loss_clip": 1.02587962,
      "balance_loss_mlp": 1.00509787,
      "epoch": 0.22798052065171648,
      "flos": 66706872600960.0,
      "grad_norm": 0.8040780334635551,
      "language_loss": 0.64823902,
      "learning_rate": 3.6028842922573553e-06,
      "loss": 0.66901493,
      "num_input_tokens_seen": 40311005,
      "step": 1896,
      "time_per_iteration": 3.206712245941162
    },
    {
      "auxiliary_loss_clip": 0.01090266,
      "auxiliary_loss_mlp": 0.00754752,
      "balance_loss_clip": 1.03051019,
      "balance_loss_mlp": 1.00066173,
      "epoch": 0.22810076354235556,
      "flos": 62080896758400.0,
      "grad_norm": 0.8509710157333448,
      "language_loss": 0.62913644,
      "learning_rate": 3.602418289319497e-06,
      "loss": 0.64758658,
      "num_input_tokens_seen": 40369560,
      "step": 1897,
      "time_per_iteration": 3.1072418689727783
    },
    {
      "auxiliary_loss_clip": 0.01149405,
      "auxiliary_loss_mlp": 0.01034233,
      "balance_loss_clip": 1.05266404,
      "balance_loss_mlp": 1.02444577,
      "epoch": 0.22822100643299464,
      "flos": 23876358635520.0,
      "grad_norm": 1.9594700156480302,
      "language_loss": 0.73504794,
      "learning_rate": 3.601952043296059e-06,
      "loss": 0.75688434,
      "num_input_tokens_seen": 40389555,
      "step": 1898,
      "time_per_iteration": 2.6265969276428223
    },
    {
      "auxiliary_loss_clip": 0.01187299,
      "auxiliary_loss_mlp": 0.01032276,
      "balance_loss_clip": 1.05607986,
      "balance_loss_mlp": 1.02270961,
      "epoch": 0.22834124932363373,
      "flos": 20991331180800.0,
      "grad_norm": 2.209433595094319,
      "language_loss": 0.80496663,
      "learning_rate": 3.6014855542577696e-06,
      "loss": 0.82716238,
      "num_input_tokens_seen": 40406765,
      "step": 1899,
      "time_per_iteration": 2.5008046627044678
    },
    {
      "auxiliary_loss_clip": 0.01182195,
      "auxiliary_loss_mlp": 0.01028833,
      "balance_loss_clip": 1.05792546,
      "balance_loss_mlp": 1.01908135,
      "epoch": 0.22846149221427284,
      "flos": 24901572620160.0,
      "grad_norm": 2.22064301455328,
      "language_loss": 0.84333611,
      "learning_rate": 3.6010188222753943e-06,
      "loss": 0.86544639,
      "num_input_tokens_seen": 40427535,
      "step": 1900,
      "time_per_iteration": 2.5809316635131836
    },
    {
      "auxiliary_loss_clip": 0.01091189,
      "auxiliary_loss_mlp": 0.01011563,
      "balance_loss_clip": 1.02984834,
      "balance_loss_mlp": 1.00984609,
      "epoch": 0.22858173510491192,
      "flos": 56132294319360.0,
      "grad_norm": 0.9047666995725748,
      "language_loss": 0.64161056,
      "learning_rate": 3.6005518474197372e-06,
      "loss": 0.66263807,
      "num_input_tokens_seen": 40479580,
      "step": 1901,
      "time_per_iteration": 2.9828567504882812
    },
    {
      "auxiliary_loss_clip": 0.0119795,
      "auxiliary_loss_mlp": 0.01031531,
      "balance_loss_clip": 1.06150997,
      "balance_loss_mlp": 1.02176785,
      "epoch": 0.228701977995551,
      "flos": 24170826332160.0,
      "grad_norm": 2.096582661508518,
      "language_loss": 0.78212988,
      "learning_rate": 3.6000846297616373e-06,
      "loss": 0.80442476,
      "num_input_tokens_seen": 40497880,
      "step": 1902,
      "time_per_iteration": 2.5138919353485107
    },
    {
      "auxiliary_loss_clip": 0.01217016,
      "auxiliary_loss_mlp": 0.01035119,
      "balance_loss_clip": 1.06371641,
      "balance_loss_mlp": 1.02481353,
      "epoch": 0.22882222088619011,
      "flos": 21387892308480.0,
      "grad_norm": 2.208548764161951,
      "language_loss": 0.72412646,
      "learning_rate": 3.5996171693719717e-06,
      "loss": 0.74664783,
      "num_input_tokens_seen": 40513975,
      "step": 1903,
      "time_per_iteration": 2.4436137676239014
    },
    {
      "auxiliary_loss_clip": 0.01109584,
      "auxiliary_loss_mlp": 0.01005274,
      "balance_loss_clip": 1.03206694,
      "balance_loss_mlp": 1.00360513,
      "epoch": 0.2289424637768292,
      "flos": 64589615377920.0,
      "grad_norm": 0.8396527258145293,
      "language_loss": 0.64781499,
      "learning_rate": 3.5991494663216528e-06,
      "loss": 0.66896355,
      "num_input_tokens_seen": 40576960,
      "step": 1904,
      "time_per_iteration": 3.1305947303771973
    },
    {
      "auxiliary_loss_clip": 0.01213654,
      "auxiliary_loss_mlp": 0.01029727,
      "balance_loss_clip": 1.06340075,
      "balance_loss_mlp": 1.02005911,
      "epoch": 0.22906270666746828,
      "flos": 22163419877760.0,
      "grad_norm": 3.14637454918225,
      "language_loss": 0.87503016,
      "learning_rate": 3.5986815206816314e-06,
      "loss": 0.89746398,
      "num_input_tokens_seen": 40595780,
      "step": 1905,
      "time_per_iteration": 2.4825522899627686
    },
    {
      "auxiliary_loss_clip": 0.01212803,
      "auxiliary_loss_mlp": 0.01036475,
      "balance_loss_clip": 1.06230724,
      "balance_loss_mlp": 1.02745128,
      "epoch": 0.2291829495581074,
      "flos": 25772334122880.0,
      "grad_norm": 1.974033283095527,
      "language_loss": 0.74598277,
      "learning_rate": 3.598213332522895e-06,
      "loss": 0.76847559,
      "num_input_tokens_seen": 40615810,
      "step": 1906,
      "time_per_iteration": 2.4877636432647705
    },
    {
      "auxiliary_loss_clip": 0.01195163,
      "auxiliary_loss_mlp": 0.01033071,
      "balance_loss_clip": 1.05830097,
      "balance_loss_mlp": 1.0232898,
      "epoch": 0.22930319244874647,
      "flos": 31172760126720.0,
      "grad_norm": 1.821359731588867,
      "language_loss": 0.77216184,
      "learning_rate": 3.597744901916466e-06,
      "loss": 0.7944442,
      "num_input_tokens_seen": 40637095,
      "step": 1907,
      "time_per_iteration": 2.561657190322876
    },
    {
      "auxiliary_loss_clip": 0.01215777,
      "auxiliary_loss_mlp": 0.01035309,
      "balance_loss_clip": 1.06053555,
      "balance_loss_mlp": 1.02530122,
      "epoch": 0.22942343533938556,
      "flos": 23254098399360.0,
      "grad_norm": 3.3067372795388024,
      "language_loss": 0.77003586,
      "learning_rate": 3.5972762289334058e-06,
      "loss": 0.79254675,
      "num_input_tokens_seen": 40656725,
      "step": 1908,
      "time_per_iteration": 2.472668409347534
    },
    {
      "auxiliary_loss_clip": 0.01135152,
      "auxiliary_loss_mlp": 0.01029602,
      "balance_loss_clip": 1.05400538,
      "balance_loss_mlp": 1.01947498,
      "epoch": 0.22954367823002464,
      "flos": 14610903436800.0,
      "grad_norm": 2.24948213431362,
      "language_loss": 0.84914607,
      "learning_rate": 3.5968073136448116e-06,
      "loss": 0.87079358,
      "num_input_tokens_seen": 40674745,
      "step": 1909,
      "time_per_iteration": 2.600472927093506
    },
    {
      "auxiliary_loss_clip": 0.01202628,
      "auxiliary_loss_mlp": 0.01037943,
      "balance_loss_clip": 1.06018329,
      "balance_loss_mlp": 1.02745259,
      "epoch": 0.22966392112066375,
      "flos": 16763604405120.0,
      "grad_norm": 1.7186533335638219,
      "language_loss": 0.91463274,
      "learning_rate": 3.596338156121818e-06,
      "loss": 0.93703842,
      "num_input_tokens_seen": 40693630,
      "step": 1910,
      "time_per_iteration": 2.455132007598877
    },
    {
      "auxiliary_loss_clip": 0.0109297,
      "auxiliary_loss_mlp": 0.01001664,
      "balance_loss_clip": 1.02800202,
      "balance_loss_mlp": 0.99997121,
      "epoch": 0.22978416401130283,
      "flos": 67474247783040.0,
      "grad_norm": 0.7406317435003689,
      "language_loss": 0.59394455,
      "learning_rate": 3.595868756435595e-06,
      "loss": 0.61489081,
      "num_input_tokens_seen": 40761310,
      "step": 1911,
      "time_per_iteration": 4.0456647872924805
    },
    {
      "auxiliary_loss_clip": 0.01173517,
      "auxiliary_loss_mlp": 0.01034111,
      "balance_loss_clip": 1.06183112,
      "balance_loss_mlp": 1.02435338,
      "epoch": 0.22990440690194192,
      "flos": 19865137086720.0,
      "grad_norm": 2.649085475940073,
      "language_loss": 0.80906856,
      "learning_rate": 3.5953991146573504e-06,
      "loss": 0.83114481,
      "num_input_tokens_seen": 40779955,
      "step": 1912,
      "time_per_iteration": 2.576812505722046
    },
    {
      "auxiliary_loss_clip": 0.01199708,
      "auxiliary_loss_mlp": 0.01032644,
      "balance_loss_clip": 1.05798328,
      "balance_loss_mlp": 1.02240396,
      "epoch": 0.23002464979258103,
      "flos": 13289246507520.0,
      "grad_norm": 2.4439647535748903,
      "language_loss": 0.83580017,
      "learning_rate": 3.5949292308583294e-06,
      "loss": 0.85812366,
      "num_input_tokens_seen": 40793200,
      "step": 1913,
      "time_per_iteration": 2.4744374752044678
    },
    {
      "auxiliary_loss_clip": 0.01214594,
      "auxiliary_loss_mlp": 0.01034855,
      "balance_loss_clip": 1.06406713,
      "balance_loss_mlp": 1.0239476,
      "epoch": 0.2301448926832201,
      "flos": 22163779013760.0,
      "grad_norm": 2.0029949690347464,
      "language_loss": 0.80833125,
      "learning_rate": 3.594459105109811e-06,
      "loss": 0.83082575,
      "num_input_tokens_seen": 40812380,
      "step": 1914,
      "time_per_iteration": 2.452862501144409
    },
    {
      "auxiliary_loss_clip": 0.01202086,
      "auxiliary_loss_mlp": 0.01033229,
      "balance_loss_clip": 1.06376553,
      "balance_loss_mlp": 1.02444339,
      "epoch": 0.2302651355738592,
      "flos": 20704477167360.0,
      "grad_norm": 1.7716538625401077,
      "language_loss": 0.81096971,
      "learning_rate": 3.593988737483115e-06,
      "loss": 0.83332288,
      "num_input_tokens_seen": 40832320,
      "step": 1915,
      "time_per_iteration": 2.4938321113586426
    },
    {
      "auxiliary_loss_clip": 0.01182161,
      "auxiliary_loss_mlp": 0.01031473,
      "balance_loss_clip": 1.0580945,
      "balance_loss_mlp": 1.02159071,
      "epoch": 0.23038537846449827,
      "flos": 18588943797120.0,
      "grad_norm": 1.9160218149132646,
      "language_loss": 0.78065801,
      "learning_rate": 3.5935181280495947e-06,
      "loss": 0.8027944,
      "num_input_tokens_seen": 40850900,
      "step": 1916,
      "time_per_iteration": 2.5129659175872803
    },
    {
      "auxiliary_loss_clip": 0.01087867,
      "auxiliary_loss_mlp": 0.01001513,
      "balance_loss_clip": 1.02836192,
      "balance_loss_mlp": 0.99984437,
      "epoch": 0.23050562135513739,
      "flos": 64224260190720.0,
      "grad_norm": 0.9080719464070544,
      "language_loss": 0.54271203,
      "learning_rate": 3.5930472768806412e-06,
      "loss": 0.56360584,
      "num_input_tokens_seen": 40909570,
      "step": 1917,
      "time_per_iteration": 3.8663241863250732
    },
    {
      "auxiliary_loss_clip": 0.01211672,
      "auxiliary_loss_mlp": 0.01036062,
      "balance_loss_clip": 1.06318474,
      "balance_loss_mlp": 1.02603054,
      "epoch": 0.23062586424577647,
      "flos": 17313396952320.0,
      "grad_norm": 2.302850048551058,
      "language_loss": 0.77138782,
      "learning_rate": 3.5925761840476826e-06,
      "loss": 0.7938652,
      "num_input_tokens_seen": 40928180,
      "step": 1918,
      "time_per_iteration": 3.3247687816619873
    },
    {
      "auxiliary_loss_clip": 0.01178787,
      "auxiliary_loss_mlp": 0.01031142,
      "balance_loss_clip": 1.05813265,
      "balance_loss_mlp": 1.02196872,
      "epoch": 0.23074610713641555,
      "flos": 27855979194240.0,
      "grad_norm": 7.456738816837622,
      "language_loss": 0.81363225,
      "learning_rate": 3.592104849622183e-06,
      "loss": 0.83573157,
      "num_input_tokens_seen": 40950435,
      "step": 1919,
      "time_per_iteration": 3.3578546047210693
    },
    {
      "auxiliary_loss_clip": 0.01143675,
      "auxiliary_loss_mlp": 0.01032357,
      "balance_loss_clip": 1.05322242,
      "balance_loss_mlp": 1.02280807,
      "epoch": 0.23086635002705466,
      "flos": 28841798937600.0,
      "grad_norm": 1.5063473308876887,
      "language_loss": 0.73119915,
      "learning_rate": 3.591633273675644e-06,
      "loss": 0.75295943,
      "num_input_tokens_seen": 40972670,
      "step": 1920,
      "time_per_iteration": 2.6541922092437744
    },
    {
      "auxiliary_loss_clip": 0.01064178,
      "auxiliary_loss_mlp": 0.01008834,
      "balance_loss_clip": 1.02913833,
      "balance_loss_mlp": 1.00650966,
      "epoch": 0.23098659291769374,
      "flos": 62923681566720.0,
      "grad_norm": 0.9073545390728627,
      "language_loss": 0.58245361,
      "learning_rate": 3.591161456279602e-06,
      "loss": 0.60318375,
      "num_input_tokens_seen": 41018215,
      "step": 1921,
      "time_per_iteration": 2.92101788520813
    },
    {
      "auxiliary_loss_clip": 0.01186544,
      "auxiliary_loss_mlp": 0.01033526,
      "balance_loss_clip": 1.0567441,
      "balance_loss_mlp": 1.02385259,
      "epoch": 0.23110683580833283,
      "flos": 23476816679040.0,
      "grad_norm": 1.5056857106502466,
      "language_loss": 0.80467165,
      "learning_rate": 3.590689397505633e-06,
      "loss": 0.82687235,
      "num_input_tokens_seen": 41039125,
      "step": 1922,
      "time_per_iteration": 2.712777853012085
    },
    {
      "auxiliary_loss_clip": 0.01210273,
      "auxiliary_loss_mlp": 0.010294,
      "balance_loss_clip": 1.06129718,
      "balance_loss_mlp": 1.01991057,
      "epoch": 0.2312270786989719,
      "flos": 27271066124160.0,
      "grad_norm": 1.6353858722836059,
      "language_loss": 0.86771405,
      "learning_rate": 3.590217097425347e-06,
      "loss": 0.89011079,
      "num_input_tokens_seen": 41059025,
      "step": 1923,
      "time_per_iteration": 2.529050827026367
    },
    {
      "auxiliary_loss_clip": 0.01215289,
      "auxiliary_loss_mlp": 0.01032136,
      "balance_loss_clip": 1.06352818,
      "balance_loss_mlp": 1.02207446,
      "epoch": 0.23134732158961102,
      "flos": 13261344618240.0,
      "grad_norm": 1.7843459673148958,
      "language_loss": 0.71058226,
      "learning_rate": 3.589744556110391e-06,
      "loss": 0.73305655,
      "num_input_tokens_seen": 41077015,
      "step": 1924,
      "time_per_iteration": 2.439056158065796
    },
    {
      "auxiliary_loss_clip": 0.01179935,
      "auxiliary_loss_mlp": 0.01034317,
      "balance_loss_clip": 1.05626798,
      "balance_loss_mlp": 1.02564418,
      "epoch": 0.2314675644802501,
      "flos": 36977648250240.0,
      "grad_norm": 1.6877289085880027,
      "language_loss": 0.84338379,
      "learning_rate": 3.58927177363245e-06,
      "loss": 0.86552632,
      "num_input_tokens_seen": 41099840,
      "step": 1925,
      "time_per_iteration": 2.697558641433716
    },
    {
      "auxiliary_loss_clip": 0.01163788,
      "auxiliary_loss_mlp": 0.01035418,
      "balance_loss_clip": 1.05415154,
      "balance_loss_mlp": 1.02501678,
      "epoch": 0.2315878073708892,
      "flos": 23842207779840.0,
      "grad_norm": 2.3634779859918296,
      "language_loss": 0.72541994,
      "learning_rate": 3.5887987500632447e-06,
      "loss": 0.74741197,
      "num_input_tokens_seen": 41117845,
      "step": 1926,
      "time_per_iteration": 2.5524322986602783
    },
    {
      "auxiliary_loss_clip": 0.01169859,
      "auxiliary_loss_mlp": 0.010361,
      "balance_loss_clip": 1.05530918,
      "balance_loss_mlp": 1.02742767,
      "epoch": 0.2317080502615283,
      "flos": 23039424766080.0,
      "grad_norm": 1.8556639998048141,
      "language_loss": 0.84464228,
      "learning_rate": 3.5883254854745325e-06,
      "loss": 0.86670184,
      "num_input_tokens_seen": 41136235,
      "step": 1927,
      "time_per_iteration": 2.6146647930145264
    },
    {
      "auxiliary_loss_clip": 0.0119896,
      "auxiliary_loss_mlp": 0.01031803,
      "balance_loss_clip": 1.05650103,
      "balance_loss_mlp": 1.0221529,
      "epoch": 0.23182829315216738,
      "flos": 11254656435840.0,
      "grad_norm": 1.9891985581114355,
      "language_loss": 0.75201017,
      "learning_rate": 3.587851979938107e-06,
      "loss": 0.7743178,
      "num_input_tokens_seen": 41153125,
      "step": 1928,
      "time_per_iteration": 2.4627838134765625
    },
    {
      "auxiliary_loss_clip": 0.0119501,
      "auxiliary_loss_mlp": 0.01032014,
      "balance_loss_clip": 1.05958784,
      "balance_loss_mlp": 1.02259636,
      "epoch": 0.23194853604280646,
      "flos": 19828939155840.0,
      "grad_norm": 2.727395033535805,
      "language_loss": 0.77979016,
      "learning_rate": 3.5873782335257985e-06,
      "loss": 0.80206043,
      "num_input_tokens_seen": 41171290,
      "step": 1929,
      "time_per_iteration": 2.4804580211639404
    },
    {
      "auxiliary_loss_clip": 0.01163595,
      "auxiliary_loss_mlp": 0.01036443,
      "balance_loss_clip": 1.05667007,
      "balance_loss_mlp": 1.02674484,
      "epoch": 0.23206877893344555,
      "flos": 15305020830720.0,
      "grad_norm": 2.197442347653758,
      "language_loss": 0.78496575,
      "learning_rate": 3.5869042463094744e-06,
      "loss": 0.80696607,
      "num_input_tokens_seen": 41189005,
      "step": 1930,
      "time_per_iteration": 2.513066053390503
    },
    {
      "auxiliary_loss_clip": 0.01131983,
      "auxiliary_loss_mlp": 0.01038852,
      "balance_loss_clip": 1.04848385,
      "balance_loss_mlp": 1.02873111,
      "epoch": 0.23218902182408466,
      "flos": 22711488572160.0,
      "grad_norm": 1.9071195582850453,
      "language_loss": 0.76829469,
      "learning_rate": 3.586430018361038e-06,
      "loss": 0.79000306,
      "num_input_tokens_seen": 41208775,
      "step": 1931,
      "time_per_iteration": 2.6066949367523193
    },
    {
      "auxiliary_loss_clip": 0.01166604,
      "auxiliary_loss_mlp": 0.01033146,
      "balance_loss_clip": 1.05240917,
      "balance_loss_mlp": 1.02277541,
      "epoch": 0.23230926471472374,
      "flos": 22710734386560.0,
      "grad_norm": 2.443180941755307,
      "language_loss": 0.76346928,
      "learning_rate": 3.5859555497524283e-06,
      "loss": 0.78546685,
      "num_input_tokens_seen": 41226010,
      "step": 1932,
      "time_per_iteration": 2.5034677982330322
    },
    {
      "auxiliary_loss_clip": 0.01195427,
      "auxiliary_loss_mlp": 0.01036002,
      "balance_loss_clip": 1.05957794,
      "balance_loss_mlp": 1.02704334,
      "epoch": 0.23242950760536282,
      "flos": 20375499479040.0,
      "grad_norm": 1.9287028496982914,
      "language_loss": 0.91858327,
      "learning_rate": 3.5854808405556237e-06,
      "loss": 0.94089758,
      "num_input_tokens_seen": 41245245,
      "step": 1933,
      "time_per_iteration": 2.479234457015991
    },
    {
      "auxiliary_loss_clip": 0.01164682,
      "auxiliary_loss_mlp": 0.01035285,
      "balance_loss_clip": 1.05380058,
      "balance_loss_mlp": 1.02699971,
      "epoch": 0.23254975049600193,
      "flos": 16908324301440.0,
      "grad_norm": 2.2676536556648554,
      "language_loss": 0.74906296,
      "learning_rate": 3.5850058908426355e-06,
      "loss": 0.77106261,
      "num_input_tokens_seen": 41263795,
      "step": 1934,
      "time_per_iteration": 2.5206565856933594
    },
    {
      "auxiliary_loss_clip": 0.01182763,
      "auxiliary_loss_mlp": 0.01035244,
      "balance_loss_clip": 1.05431151,
      "balance_loss_mlp": 1.02611232,
      "epoch": 0.23266999338664102,
      "flos": 23294821443840.0,
      "grad_norm": 1.7005426418367304,
      "language_loss": 0.85302925,
      "learning_rate": 3.584530700685514e-06,
      "loss": 0.87520927,
      "num_input_tokens_seen": 41284055,
      "step": 1935,
      "time_per_iteration": 2.5419108867645264
    },
    {
      "auxiliary_loss_clip": 0.01181153,
      "auxiliary_loss_mlp": 0.01029347,
      "balance_loss_clip": 1.06176329,
      "balance_loss_mlp": 1.02011478,
      "epoch": 0.2327902362772801,
      "flos": 19569987031680.0,
      "grad_norm": 1.9727706416318,
      "language_loss": 0.88636446,
      "learning_rate": 3.5840552701563448e-06,
      "loss": 0.90846944,
      "num_input_tokens_seen": 41300255,
      "step": 1936,
      "time_per_iteration": 2.4959185123443604
    },
    {
      "auxiliary_loss_clip": 0.01184918,
      "auxiliary_loss_mlp": 0.01032039,
      "balance_loss_clip": 1.05404174,
      "balance_loss_mlp": 1.02241278,
      "epoch": 0.2329104791679192,
      "flos": 16727514215040.0,
      "grad_norm": 2.3995711887507736,
      "language_loss": 0.82066882,
      "learning_rate": 3.5835795993272513e-06,
      "loss": 0.84283841,
      "num_input_tokens_seen": 41318540,
      "step": 1937,
      "time_per_iteration": 2.5202860832214355
    },
    {
      "auxiliary_loss_clip": 0.01124891,
      "auxiliary_loss_mlp": 0.01043073,
      "balance_loss_clip": 1.05209231,
      "balance_loss_mlp": 1.03323257,
      "epoch": 0.2330307220585583,
      "flos": 22163743100160.0,
      "grad_norm": 4.447855821462046,
      "language_loss": 0.71046537,
      "learning_rate": 3.583103688270391e-06,
      "loss": 0.73214501,
      "num_input_tokens_seen": 41338320,
      "step": 1938,
      "time_per_iteration": 3.511669874191284
    },
    {
      "auxiliary_loss_clip": 0.01167642,
      "auxiliary_loss_mlp": 0.01036455,
      "balance_loss_clip": 1.05252004,
      "balance_loss_mlp": 1.02556562,
      "epoch": 0.23315096494919738,
      "flos": 19317319787520.0,
      "grad_norm": 2.0946630408222084,
      "language_loss": 0.89410293,
      "learning_rate": 3.58262753705796e-06,
      "loss": 0.91614395,
      "num_input_tokens_seen": 41353210,
      "step": 1939,
      "time_per_iteration": 2.519331693649292
    },
    {
      "auxiliary_loss_clip": 0.01094488,
      "auxiliary_loss_mlp": 0.01009485,
      "balance_loss_clip": 1.03496528,
      "balance_loss_mlp": 1.00788724,
      "epoch": 0.23327120783983646,
      "flos": 53031048946560.0,
      "grad_norm": 0.762076360197271,
      "language_loss": 0.55555022,
      "learning_rate": 3.5821511457621902e-06,
      "loss": 0.57658994,
      "num_input_tokens_seen": 41410510,
      "step": 1940,
      "time_per_iteration": 3.0885448455810547
    },
    {
      "auxiliary_loss_clip": 0.01178312,
      "auxiliary_loss_mlp": 0.0103693,
      "balance_loss_clip": 1.05692327,
      "balance_loss_mlp": 1.02673125,
      "epoch": 0.23339145073047557,
      "flos": 17126984344320.0,
      "grad_norm": 3.0561073140637784,
      "language_loss": 0.81302071,
      "learning_rate": 3.5816745144553497e-06,
      "loss": 0.83517313,
      "num_input_tokens_seen": 41425830,
      "step": 1941,
      "time_per_iteration": 2.4839398860931396
    },
    {
      "auxiliary_loss_clip": 0.01146039,
      "auxiliary_loss_mlp": 0.01027794,
      "balance_loss_clip": 1.05399966,
      "balance_loss_mlp": 1.01841247,
      "epoch": 0.23351169362111465,
      "flos": 13078918419840.0,
      "grad_norm": 2.814779265914372,
      "language_loss": 0.75493437,
      "learning_rate": 3.5811976432097424e-06,
      "loss": 0.77667266,
      "num_input_tokens_seen": 41443500,
      "step": 1942,
      "time_per_iteration": 2.5535335540771484
    },
    {
      "auxiliary_loss_clip": 0.01195001,
      "auxiliary_loss_mlp": 0.00764066,
      "balance_loss_clip": 1.06108856,
      "balance_loss_mlp": 1.00104523,
      "epoch": 0.23363193651175373,
      "flos": 15851257931520.0,
      "grad_norm": 2.2281377174652794,
      "language_loss": 0.84510922,
      "learning_rate": 3.58072053209771e-06,
      "loss": 0.86469984,
      "num_input_tokens_seen": 41460055,
      "step": 1943,
      "time_per_iteration": 2.499406099319458
    },
    {
      "auxiliary_loss_clip": 0.01173637,
      "auxiliary_loss_mlp": 0.01033634,
      "balance_loss_clip": 1.05330825,
      "balance_loss_mlp": 1.0238533,
      "epoch": 0.23375217940239285,
      "flos": 21025769345280.0,
      "grad_norm": 2.0944955210956797,
      "language_loss": 0.7908181,
      "learning_rate": 3.5802431811916296e-06,
      "loss": 0.81289077,
      "num_input_tokens_seen": 41476665,
      "step": 1944,
      "time_per_iteration": 4.196576356887817
    },
    {
      "auxiliary_loss_clip": 0.0117512,
      "auxiliary_loss_mlp": 0.01028221,
      "balance_loss_clip": 1.05613697,
      "balance_loss_mlp": 1.01920271,
      "epoch": 0.23387242229303193,
      "flos": 20594698225920.0,
      "grad_norm": 1.9070219817968121,
      "language_loss": 0.80690712,
      "learning_rate": 3.579765590563916e-06,
      "loss": 0.82894051,
      "num_input_tokens_seen": 41496065,
      "step": 1945,
      "time_per_iteration": 3.342613458633423
    },
    {
      "auxiliary_loss_clip": 0.01183395,
      "auxiliary_loss_mlp": 0.01032511,
      "balance_loss_clip": 1.05560887,
      "balance_loss_mlp": 1.02302241,
      "epoch": 0.233992665183671,
      "flos": 24279491952000.0,
      "grad_norm": 1.8303253083666764,
      "language_loss": 0.81812763,
      "learning_rate": 3.579287760287017e-06,
      "loss": 0.84028673,
      "num_input_tokens_seen": 41516815,
      "step": 1946,
      "time_per_iteration": 2.544931173324585
    },
    {
      "auxiliary_loss_clip": 0.0119233,
      "auxiliary_loss_mlp": 0.01034112,
      "balance_loss_clip": 1.05827081,
      "balance_loss_mlp": 1.02507567,
      "epoch": 0.2341129080743101,
      "flos": 30154621121280.0,
      "grad_norm": 1.8392033690714538,
      "language_loss": 0.72793335,
      "learning_rate": 3.578809690433421e-06,
      "loss": 0.75019777,
      "num_input_tokens_seen": 41538525,
      "step": 1947,
      "time_per_iteration": 2.5669567584991455
    },
    {
      "auxiliary_loss_clip": 0.01215407,
      "auxiliary_loss_mlp": 0.01036612,
      "balance_loss_clip": 1.06262004,
      "balance_loss_mlp": 1.02676547,
      "epoch": 0.2342331509649492,
      "flos": 22784135829120.0,
      "grad_norm": 4.005893996273609,
      "language_loss": 0.81113994,
      "learning_rate": 3.578331381075651e-06,
      "loss": 0.83366013,
      "num_input_tokens_seen": 41559025,
      "step": 1948,
      "time_per_iteration": 2.4530582427978516
    },
    {
      "auxiliary_loss_clip": 0.01196052,
      "auxiliary_loss_mlp": 0.01030069,
      "balance_loss_clip": 1.05800653,
      "balance_loss_mlp": 1.02032363,
      "epoch": 0.2343533938555883,
      "flos": 23623152687360.0,
      "grad_norm": 2.29998884192404,
      "language_loss": 0.69959247,
      "learning_rate": 3.5778528322862646e-06,
      "loss": 0.72185367,
      "num_input_tokens_seen": 41577845,
      "step": 1949,
      "time_per_iteration": 2.4965648651123047
    },
    {
      "auxiliary_loss_clip": 0.01196498,
      "auxiliary_loss_mlp": 0.01031312,
      "balance_loss_clip": 1.05714893,
      "balance_loss_mlp": 1.02231812,
      "epoch": 0.23447363674622737,
      "flos": 24570332375040.0,
      "grad_norm": 1.4333363379706736,
      "language_loss": 0.8650347,
      "learning_rate": 3.5773740441378585e-06,
      "loss": 0.88731283,
      "num_input_tokens_seen": 41598600,
      "step": 1950,
      "time_per_iteration": 2.4947404861450195
    },
    {
      "auxiliary_loss_clip": 0.01191019,
      "auxiliary_loss_mlp": 0.01036777,
      "balance_loss_clip": 1.05705655,
      "balance_loss_mlp": 1.0277946,
      "epoch": 0.23459387963686648,
      "flos": 53140322119680.0,
      "grad_norm": 1.7019286599252261,
      "language_loss": 0.73543864,
      "learning_rate": 3.5768950167030633e-06,
      "loss": 0.75771654,
      "num_input_tokens_seen": 41623300,
      "step": 1951,
      "time_per_iteration": 2.832526445388794
    },
    {
      "auxiliary_loss_clip": 0.01169855,
      "auxiliary_loss_mlp": 0.01038853,
      "balance_loss_clip": 1.05426359,
      "balance_loss_mlp": 1.02863121,
      "epoch": 0.23471412252750556,
      "flos": 23951412103680.0,
      "grad_norm": 1.8080582810706136,
      "language_loss": 0.78592414,
      "learning_rate": 3.576415750054548e-06,
      "loss": 0.80801123,
      "num_input_tokens_seen": 41643420,
      "step": 1952,
      "time_per_iteration": 2.5353944301605225
    },
    {
      "auxiliary_loss_clip": 0.01171031,
      "auxiliary_loss_mlp": 0.01027405,
      "balance_loss_clip": 1.05397022,
      "balance_loss_mlp": 1.01798141,
      "epoch": 0.23483436541814465,
      "flos": 15706573948800.0,
      "grad_norm": 2.0703214103623306,
      "language_loss": 0.85655081,
      "learning_rate": 3.5759362442650172e-06,
      "loss": 0.87853515,
      "num_input_tokens_seen": 41660170,
      "step": 1953,
      "time_per_iteration": 2.4940755367279053
    },
    {
      "auxiliary_loss_clip": 0.01197657,
      "auxiliary_loss_mlp": 0.01037699,
      "balance_loss_clip": 1.06208384,
      "balance_loss_mlp": 1.02786446,
      "epoch": 0.23495460830878373,
      "flos": 24936262179840.0,
      "grad_norm": 2.2820312120406805,
      "language_loss": 0.85677314,
      "learning_rate": 3.5754564994072113e-06,
      "loss": 0.87912667,
      "num_input_tokens_seen": 41679010,
      "step": 1954,
      "time_per_iteration": 2.507598876953125
    },
    {
      "auxiliary_loss_clip": 0.01177887,
      "auxiliary_loss_mlp": 0.01028535,
      "balance_loss_clip": 1.05579591,
      "balance_loss_mlp": 1.01818132,
      "epoch": 0.23507485119942284,
      "flos": 30482665056000.0,
      "grad_norm": 3.449529449489836,
      "language_loss": 0.59643817,
      "learning_rate": 3.5749765155539067e-06,
      "loss": 0.61850238,
      "num_input_tokens_seen": 41699495,
      "step": 1955,
      "time_per_iteration": 2.6668214797973633
    },
    {
      "auxiliary_loss_clip": 0.01162357,
      "auxiliary_loss_mlp": 0.01032017,
      "balance_loss_clip": 1.05296242,
      "balance_loss_mlp": 1.02195013,
      "epoch": 0.23519509409006192,
      "flos": 18329129746560.0,
      "grad_norm": 2.0889383424788512,
      "language_loss": 0.92178923,
      "learning_rate": 3.574496292777917e-06,
      "loss": 0.94373298,
      "num_input_tokens_seen": 41717705,
      "step": 1956,
      "time_per_iteration": 2.519028663635254
    },
    {
      "auxiliary_loss_clip": 0.01188259,
      "auxiliary_loss_mlp": 0.01039365,
      "balance_loss_clip": 1.05840158,
      "balance_loss_mlp": 1.02860069,
      "epoch": 0.235315336980701,
      "flos": 29643217234560.0,
      "grad_norm": 1.9944505516913964,
      "language_loss": 0.71760261,
      "learning_rate": 3.574015831152092e-06,
      "loss": 0.73987889,
      "num_input_tokens_seen": 41738120,
      "step": 1957,
      "time_per_iteration": 2.5755717754364014
    },
    {
      "auxiliary_loss_clip": 0.01171126,
      "auxiliary_loss_mlp": 0.0102877,
      "balance_loss_clip": 1.05480516,
      "balance_loss_mlp": 1.01953769,
      "epoch": 0.23543557987134012,
      "flos": 18551704371840.0,
      "grad_norm": 2.1748782295751448,
      "language_loss": 0.83523822,
      "learning_rate": 3.573535130749316e-06,
      "loss": 0.8572371,
      "num_input_tokens_seen": 41756070,
      "step": 1958,
      "time_per_iteration": 2.4874367713928223
    },
    {
      "auxiliary_loss_clip": 0.01171534,
      "auxiliary_loss_mlp": 0.0103606,
      "balance_loss_clip": 1.05612242,
      "balance_loss_mlp": 1.0266242,
      "epoch": 0.2355558227619792,
      "flos": 24679033908480.0,
      "grad_norm": 1.6505177091237302,
      "language_loss": 0.7380923,
      "learning_rate": 3.5730541916425127e-06,
      "loss": 0.76016819,
      "num_input_tokens_seen": 41777550,
      "step": 1959,
      "time_per_iteration": 2.5593602657318115
    },
    {
      "auxiliary_loss_clip": 0.0116974,
      "auxiliary_loss_mlp": 0.01030847,
      "balance_loss_clip": 1.05687869,
      "balance_loss_mlp": 1.02144718,
      "epoch": 0.23567606565261828,
      "flos": 21944795748480.0,
      "grad_norm": 2.191747657827774,
      "language_loss": 0.86217797,
      "learning_rate": 3.572573013904639e-06,
      "loss": 0.88418388,
      "num_input_tokens_seen": 41797460,
      "step": 1960,
      "time_per_iteration": 2.5693535804748535
    },
    {
      "auxiliary_loss_clip": 0.01209579,
      "auxiliary_loss_mlp": 0.0102695,
      "balance_loss_clip": 1.05912602,
      "balance_loss_mlp": 1.01752019,
      "epoch": 0.2357963085432574,
      "flos": 13589352639360.0,
      "grad_norm": 2.1483254493480497,
      "language_loss": 0.92030525,
      "learning_rate": 3.572091597608689e-06,
      "loss": 0.94267046,
      "num_input_tokens_seen": 41815585,
      "step": 1961,
      "time_per_iteration": 2.4453775882720947
    },
    {
      "auxiliary_loss_clip": 0.01187965,
      "auxiliary_loss_mlp": 0.01031834,
      "balance_loss_clip": 1.0585885,
      "balance_loss_mlp": 1.0214448,
      "epoch": 0.23591655143389648,
      "flos": 22088689632000.0,
      "grad_norm": 2.082642869489746,
      "language_loss": 0.73624814,
      "learning_rate": 3.571609942827694e-06,
      "loss": 0.75844616,
      "num_input_tokens_seen": 41834700,
      "step": 1962,
      "time_per_iteration": 2.519336700439453
    },
    {
      "auxiliary_loss_clip": 0.01180052,
      "auxiliary_loss_mlp": 0.0103066,
      "balance_loss_clip": 1.05754554,
      "balance_loss_mlp": 1.02139783,
      "epoch": 0.23603679432453556,
      "flos": 17017349057280.0,
      "grad_norm": 2.0141837876752233,
      "language_loss": 0.88374054,
      "learning_rate": 3.57112804963472e-06,
      "loss": 0.90584767,
      "num_input_tokens_seen": 41852915,
      "step": 1963,
      "time_per_iteration": 2.499077796936035
    },
    {
      "auxiliary_loss_clip": 0.01163569,
      "auxiliary_loss_mlp": 0.01032965,
      "balance_loss_clip": 1.05818379,
      "balance_loss_mlp": 1.02410722,
      "epoch": 0.23615703721517464,
      "flos": 19171307001600.0,
      "grad_norm": 1.8335817382633506,
      "language_loss": 0.76499212,
      "learning_rate": 3.57064591810287e-06,
      "loss": 0.7869575,
      "num_input_tokens_seen": 41870415,
      "step": 1964,
      "time_per_iteration": 2.5408272743225098
    },
    {
      "auxiliary_loss_clip": 0.01209766,
      "auxiliary_loss_mlp": 0.00764064,
      "balance_loss_clip": 1.06146097,
      "balance_loss_mlp": 1.00107098,
      "epoch": 0.23627728010581375,
      "flos": 19098803399040.0,
      "grad_norm": 2.091247943689354,
      "language_loss": 0.81079847,
      "learning_rate": 3.570163548305284e-06,
      "loss": 0.83053684,
      "num_input_tokens_seen": 41889345,
      "step": 1965,
      "time_per_iteration": 3.2580833435058594
    },
    {
      "auxiliary_loss_clip": 0.01179277,
      "auxiliary_loss_mlp": 0.01036764,
      "balance_loss_clip": 1.05619395,
      "balance_loss_mlp": 1.02648187,
      "epoch": 0.23639752299645284,
      "flos": 14282213057280.0,
      "grad_norm": 2.225712014217286,
      "language_loss": 0.69493705,
      "learning_rate": 3.569680940315135e-06,
      "loss": 0.7170974,
      "num_input_tokens_seen": 41905745,
      "step": 1966,
      "time_per_iteration": 2.5123424530029297
    },
    {
      "auxiliary_loss_clip": 0.01171246,
      "auxiliary_loss_mlp": 0.01036642,
      "balance_loss_clip": 1.05665433,
      "balance_loss_mlp": 1.02585959,
      "epoch": 0.23651776588709192,
      "flos": 22893411980160.0,
      "grad_norm": 1.8032376027295576,
      "language_loss": 0.82164556,
      "learning_rate": 3.5691980942056356e-06,
      "loss": 0.84372449,
      "num_input_tokens_seen": 41925115,
      "step": 1967,
      "time_per_iteration": 2.562812328338623
    },
    {
      "auxiliary_loss_clip": 0.01200229,
      "auxiliary_loss_mlp": 0.01029254,
      "balance_loss_clip": 1.05865979,
      "balance_loss_mlp": 1.01964009,
      "epoch": 0.23663800877773103,
      "flos": 18624531196800.0,
      "grad_norm": 1.8431660051754692,
      "language_loss": 0.79439557,
      "learning_rate": 3.5687150100500332e-06,
      "loss": 0.81669039,
      "num_input_tokens_seen": 41944815,
      "step": 1968,
      "time_per_iteration": 2.5219881534576416
    },
    {
      "auxiliary_loss_clip": 0.01194978,
      "auxiliary_loss_mlp": 0.01028519,
      "balance_loss_clip": 1.05712223,
      "balance_loss_mlp": 1.01887512,
      "epoch": 0.2367582516683701,
      "flos": 25555828896000.0,
      "grad_norm": 1.5966483987149835,
      "language_loss": 0.74536216,
      "learning_rate": 3.568231687921611e-06,
      "loss": 0.76759714,
      "num_input_tokens_seen": 41964990,
      "step": 1969,
      "time_per_iteration": 2.592735528945923
    },
    {
      "auxiliary_loss_clip": 0.01208422,
      "auxiliary_loss_mlp": 0.01032555,
      "balance_loss_clip": 1.06056547,
      "balance_loss_mlp": 1.02361417,
      "epoch": 0.2368784945590092,
      "flos": 23295072839040.0,
      "grad_norm": 1.4908503387488736,
      "language_loss": 0.80382669,
      "learning_rate": 3.5677481278936883e-06,
      "loss": 0.82623649,
      "num_input_tokens_seen": 41984570,
      "step": 1970,
      "time_per_iteration": 3.3571486473083496
    },
    {
      "auxiliary_loss_clip": 0.01106091,
      "auxiliary_loss_mlp": 0.01003246,
      "balance_loss_clip": 1.04770589,
      "balance_loss_mlp": 1.00119519,
      "epoch": 0.23699873744964828,
      "flos": 69859291875840.0,
      "grad_norm": 0.8305623939719402,
      "language_loss": 0.57858795,
      "learning_rate": 3.5672643300396214e-06,
      "loss": 0.59968132,
      "num_input_tokens_seen": 42053715,
      "step": 1971,
      "time_per_iteration": 4.0693888664245605
    },
    {
      "auxiliary_loss_clip": 0.01166607,
      "auxiliary_loss_mlp": 0.01032146,
      "balance_loss_clip": 1.0561471,
      "balance_loss_mlp": 1.02358079,
      "epoch": 0.2371189803402874,
      "flos": 21835052720640.0,
      "grad_norm": 2.1581035283230445,
      "language_loss": 0.67821872,
      "learning_rate": 3.566780294432802e-06,
      "loss": 0.70020628,
      "num_input_tokens_seen": 42070890,
      "step": 1972,
      "time_per_iteration": 3.3236582279205322
    },
    {
      "auxiliary_loss_clip": 0.012112,
      "auxiliary_loss_mlp": 0.01033971,
      "balance_loss_clip": 1.06122279,
      "balance_loss_mlp": 1.0253408,
      "epoch": 0.23723922323092647,
      "flos": 21908490076800.0,
      "grad_norm": 3.020341069181952,
      "language_loss": 0.74790835,
      "learning_rate": 3.566296021146657e-06,
      "loss": 0.77036005,
      "num_input_tokens_seen": 42090270,
      "step": 1973,
      "time_per_iteration": 2.476217746734619
    },
    {
      "auxiliary_loss_clip": 0.01215522,
      "auxiliary_loss_mlp": 0.01030141,
      "balance_loss_clip": 1.06330132,
      "balance_loss_mlp": 1.01981199,
      "epoch": 0.23735946612156555,
      "flos": 32708803380480.0,
      "grad_norm": 1.714424796244191,
      "language_loss": 0.73202336,
      "learning_rate": 3.565811510254652e-06,
      "loss": 0.75448,
      "num_input_tokens_seen": 42111150,
      "step": 1974,
      "time_per_iteration": 2.55385684967041
    },
    {
      "auxiliary_loss_clip": 0.01104221,
      "auxiliary_loss_mlp": 0.01022096,
      "balance_loss_clip": 1.0378381,
      "balance_loss_mlp": 1.01942599,
      "epoch": 0.23747970901220466,
      "flos": 70546944821760.0,
      "grad_norm": 0.8250801290332915,
      "language_loss": 0.58260167,
      "learning_rate": 3.5653267618302845e-06,
      "loss": 0.60386491,
      "num_input_tokens_seen": 42178730,
      "step": 1975,
      "time_per_iteration": 3.116600275039673
    },
    {
      "auxiliary_loss_clip": 0.01208879,
      "auxiliary_loss_mlp": 0.01032268,
      "balance_loss_clip": 1.05930519,
      "balance_loss_mlp": 1.02248716,
      "epoch": 0.23759995190284375,
      "flos": 20849807594880.0,
      "grad_norm": 2.0319083326706395,
      "language_loss": 0.85615057,
      "learning_rate": 3.564841775947093e-06,
      "loss": 0.87856209,
      "num_input_tokens_seen": 42199620,
      "step": 1976,
      "time_per_iteration": 2.463066577911377
    },
    {
      "auxiliary_loss_clip": 0.01164922,
      "auxiliary_loss_mlp": 0.01033955,
      "balance_loss_clip": 1.05365968,
      "balance_loss_mlp": 1.0239172,
      "epoch": 0.23772019479348283,
      "flos": 32921645420160.0,
      "grad_norm": 2.3155159880116485,
      "language_loss": 0.76085317,
      "learning_rate": 3.5643565526786475e-06,
      "loss": 0.78284192,
      "num_input_tokens_seen": 42219560,
      "step": 1977,
      "time_per_iteration": 2.707880735397339
    },
    {
      "auxiliary_loss_clip": 0.01212583,
      "auxiliary_loss_mlp": 0.01031647,
      "balance_loss_clip": 1.06224513,
      "balance_loss_mlp": 1.02216387,
      "epoch": 0.2378404376841219,
      "flos": 32342765834880.0,
      "grad_norm": 1.578334388010235,
      "language_loss": 0.77168584,
      "learning_rate": 3.5638710920985574e-06,
      "loss": 0.79412812,
      "num_input_tokens_seen": 42241020,
      "step": 1978,
      "time_per_iteration": 2.5489962100982666
    },
    {
      "auxiliary_loss_clip": 0.01202672,
      "auxiliary_loss_mlp": 0.00765138,
      "balance_loss_clip": 1.0582931,
      "balance_loss_mlp": 1.0010426,
      "epoch": 0.23796068057476102,
      "flos": 22997624313600.0,
      "grad_norm": 2.1064573105054776,
      "language_loss": 0.81533271,
      "learning_rate": 3.5633853942804655e-06,
      "loss": 0.83501077,
      "num_input_tokens_seen": 42259345,
      "step": 1979,
      "time_per_iteration": 2.4966232776641846
    },
    {
      "auxiliary_loss_clip": 0.01167636,
      "auxiliary_loss_mlp": 0.01033803,
      "balance_loss_clip": 1.05255985,
      "balance_loss_mlp": 1.02381372,
      "epoch": 0.2380809234654001,
      "flos": 13480938414720.0,
      "grad_norm": 3.389874887004538,
      "language_loss": 0.76879197,
      "learning_rate": 3.5628994592980527e-06,
      "loss": 0.79080635,
      "num_input_tokens_seen": 42277250,
      "step": 1980,
      "time_per_iteration": 2.5508618354797363
    },
    {
      "auxiliary_loss_clip": 0.0121223,
      "auxiliary_loss_mlp": 0.01031529,
      "balance_loss_clip": 1.06083322,
      "balance_loss_mlp": 1.02267718,
      "epoch": 0.2382011663560392,
      "flos": 16871803148160.0,
      "grad_norm": 1.9230295745879118,
      "language_loss": 0.70369422,
      "learning_rate": 3.562413287225034e-06,
      "loss": 0.7261318,
      "num_input_tokens_seen": 42295360,
      "step": 1981,
      "time_per_iteration": 2.4266204833984375
    },
    {
      "auxiliary_loss_clip": 0.01192312,
      "auxiliary_loss_mlp": 0.01029125,
      "balance_loss_clip": 1.05920398,
      "balance_loss_mlp": 1.01951706,
      "epoch": 0.2383214092466783,
      "flos": 18441135331200.0,
      "grad_norm": 2.189259462930049,
      "language_loss": 0.89515364,
      "learning_rate": 3.5619268781351623e-06,
      "loss": 0.91736799,
      "num_input_tokens_seen": 42313430,
      "step": 1982,
      "time_per_iteration": 2.4984307289123535
    },
    {
      "auxiliary_loss_clip": 0.01176238,
      "auxiliary_loss_mlp": 0.01031011,
      "balance_loss_clip": 1.05834305,
      "balance_loss_mlp": 1.02224338,
      "epoch": 0.23844165213731738,
      "flos": 19755717281280.0,
      "grad_norm": 2.076334799203889,
      "language_loss": 0.76870787,
      "learning_rate": 3.5614402321022256e-06,
      "loss": 0.79078031,
      "num_input_tokens_seen": 42331260,
      "step": 1983,
      "time_per_iteration": 2.522869110107422
    },
    {
      "auxiliary_loss_clip": 0.01143687,
      "auxiliary_loss_mlp": 0.01031858,
      "balance_loss_clip": 1.0510968,
      "balance_loss_mlp": 1.02254772,
      "epoch": 0.23856189502795647,
      "flos": 23367360960000.0,
      "grad_norm": 1.7014832108313558,
      "language_loss": 0.87191939,
      "learning_rate": 3.5609533492000463e-06,
      "loss": 0.89367479,
      "num_input_tokens_seen": 42350150,
      "step": 1984,
      "time_per_iteration": 2.61458158493042
    },
    {
      "auxiliary_loss_clip": 0.01177556,
      "auxiliary_loss_mlp": 0.01026957,
      "balance_loss_clip": 1.05840349,
      "balance_loss_mlp": 1.01724744,
      "epoch": 0.23868213791859555,
      "flos": 23475056912640.0,
      "grad_norm": 2.0577482234676463,
      "language_loss": 0.78511691,
      "learning_rate": 3.560466229502485e-06,
      "loss": 0.80716199,
      "num_input_tokens_seen": 42369495,
      "step": 1985,
      "time_per_iteration": 2.5246098041534424
    },
    {
      "auxiliary_loss_clip": 0.0117998,
      "auxiliary_loss_mlp": 0.00763875,
      "balance_loss_clip": 1.06073093,
      "balance_loss_mlp": 1.00109267,
      "epoch": 0.23880238080923466,
      "flos": 16617340224000.0,
      "grad_norm": 2.2642586534869737,
      "language_loss": 0.8974998,
      "learning_rate": 3.5599788730834384e-06,
      "loss": 0.91693836,
      "num_input_tokens_seen": 42387455,
      "step": 1986,
      "time_per_iteration": 2.4817707538604736
    },
    {
      "auxiliary_loss_clip": 0.01200265,
      "auxiliary_loss_mlp": 0.01029176,
      "balance_loss_clip": 1.0603857,
      "balance_loss_mlp": 1.01988411,
      "epoch": 0.23892262369987374,
      "flos": 17348409734400.0,
      "grad_norm": 2.670884152669051,
      "language_loss": 0.78372413,
      "learning_rate": 3.559491280016836e-06,
      "loss": 0.80601859,
      "num_input_tokens_seen": 42405400,
      "step": 1987,
      "time_per_iteration": 2.4722206592559814
    },
    {
      "auxiliary_loss_clip": 0.01182919,
      "auxiliary_loss_mlp": 0.01039264,
      "balance_loss_clip": 1.05933201,
      "balance_loss_mlp": 1.02882707,
      "epoch": 0.23904286659051283,
      "flos": 22309899540480.0,
      "grad_norm": 2.4691006407956957,
      "language_loss": 0.70876753,
      "learning_rate": 3.5590034503766465e-06,
      "loss": 0.73098934,
      "num_input_tokens_seen": 42425065,
      "step": 1988,
      "time_per_iteration": 2.526493549346924
    },
    {
      "auxiliary_loss_clip": 0.01210234,
      "auxiliary_loss_mlp": 0.01035053,
      "balance_loss_clip": 1.06063533,
      "balance_loss_mlp": 1.02621365,
      "epoch": 0.23916310948115194,
      "flos": 21178246579200.0,
      "grad_norm": 4.9314699812013405,
      "language_loss": 0.81078196,
      "learning_rate": 3.558515384236874e-06,
      "loss": 0.83323479,
      "num_input_tokens_seen": 42442495,
      "step": 1989,
      "time_per_iteration": 2.4772684574127197
    },
    {
      "auxiliary_loss_clip": 0.01157176,
      "auxiliary_loss_mlp": 0.00764762,
      "balance_loss_clip": 1.05485356,
      "balance_loss_mlp": 1.00118685,
      "epoch": 0.23928335237179102,
      "flos": 14137349506560.0,
      "grad_norm": 2.0335352429936515,
      "language_loss": 0.84043908,
      "learning_rate": 3.558027081671556e-06,
      "loss": 0.85965842,
      "num_input_tokens_seen": 42459480,
      "step": 1990,
      "time_per_iteration": 2.5206592082977295
    },
    {
      "auxiliary_loss_clip": 0.01200441,
      "auxiliary_loss_mlp": 0.01038769,
      "balance_loss_clip": 1.05949664,
      "balance_loss_mlp": 1.0281533,
      "epoch": 0.2394035952624301,
      "flos": 23769596436480.0,
      "grad_norm": 1.708095185118585,
      "language_loss": 0.69002587,
      "learning_rate": 3.557538542754769e-06,
      "loss": 0.71241796,
      "num_input_tokens_seen": 42479175,
      "step": 1991,
      "time_per_iteration": 3.338768243789673
    },
    {
      "auxiliary_loss_clip": 0.01212787,
      "auxiliary_loss_mlp": 0.01036353,
      "balance_loss_clip": 1.06277037,
      "balance_loss_mlp": 1.0266912,
      "epoch": 0.2395238381530692,
      "flos": 24206198250240.0,
      "grad_norm": 3.227478221003939,
      "language_loss": 0.66544425,
      "learning_rate": 3.557049767560623e-06,
      "loss": 0.68793571,
      "num_input_tokens_seen": 42498090,
      "step": 1992,
      "time_per_iteration": 2.4876692295074463
    },
    {
      "auxiliary_loss_clip": 0.01155904,
      "auxiliary_loss_mlp": 0.01032077,
      "balance_loss_clip": 1.05894113,
      "balance_loss_mlp": 1.02250493,
      "epoch": 0.2396440810437083,
      "flos": 25295763450240.0,
      "grad_norm": 2.274049790923728,
      "language_loss": 0.85965216,
      "learning_rate": 3.5565607561632655e-06,
      "loss": 0.88153195,
      "num_input_tokens_seen": 42516930,
      "step": 1993,
      "time_per_iteration": 2.623361825942993
    },
    {
      "auxiliary_loss_clip": 0.01177635,
      "auxiliary_loss_mlp": 0.01029115,
      "balance_loss_clip": 1.0570519,
      "balance_loss_mlp": 1.01871979,
      "epoch": 0.23976432393434738,
      "flos": 28543093436160.0,
      "grad_norm": 2.644441143435662,
      "language_loss": 0.79429114,
      "learning_rate": 3.5560715086368787e-06,
      "loss": 0.81635857,
      "num_input_tokens_seen": 42534800,
      "step": 1994,
      "time_per_iteration": 2.5808451175689697
    },
    {
      "auxiliary_loss_clip": 0.01175713,
      "auxiliary_loss_mlp": 0.01033302,
      "balance_loss_clip": 1.0576731,
      "balance_loss_mlp": 1.02399766,
      "epoch": 0.23988456682498646,
      "flos": 19494358945920.0,
      "grad_norm": 2.2746090547770277,
      "language_loss": 0.82297271,
      "learning_rate": 3.5555820250556816e-06,
      "loss": 0.84506285,
      "num_input_tokens_seen": 42552000,
      "step": 1995,
      "time_per_iteration": 2.533167839050293
    },
    {
      "auxiliary_loss_clip": 0.011888,
      "auxiliary_loss_mlp": 0.01035532,
      "balance_loss_clip": 1.06117821,
      "balance_loss_mlp": 1.02594709,
      "epoch": 0.24000480971562557,
      "flos": 20266331068800.0,
      "grad_norm": 2.381183328399539,
      "language_loss": 0.69149089,
      "learning_rate": 3.5550923054939278e-06,
      "loss": 0.71373427,
      "num_input_tokens_seen": 42571455,
      "step": 1996,
      "time_per_iteration": 2.5521271228790283
    },
    {
      "auxiliary_loss_clip": 0.01146469,
      "auxiliary_loss_mlp": 0.0103399,
      "balance_loss_clip": 1.05402184,
      "balance_loss_mlp": 1.02445889,
      "epoch": 0.24012505260626466,
      "flos": 25443176866560.0,
      "grad_norm": 2.0017799039157533,
      "language_loss": 0.74773115,
      "learning_rate": 3.5546023500259083e-06,
      "loss": 0.76953578,
      "num_input_tokens_seen": 42592550,
      "step": 1997,
      "time_per_iteration": 4.325376987457275
    },
    {
      "auxiliary_loss_clip": 0.0115644,
      "auxiliary_loss_mlp": 0.01033456,
      "balance_loss_clip": 1.05416584,
      "balance_loss_mlp": 1.02372873,
      "epoch": 0.24024529549690374,
      "flos": 15553342529280.0,
      "grad_norm": 4.136079711503388,
      "language_loss": 0.80801117,
      "learning_rate": 3.5541121587259477e-06,
      "loss": 0.82991016,
      "num_input_tokens_seen": 42610385,
      "step": 1998,
      "time_per_iteration": 3.39522123336792
    },
    {
      "auxiliary_loss_clip": 0.01120171,
      "auxiliary_loss_mlp": 0.01027313,
      "balance_loss_clip": 1.04854107,
      "balance_loss_mlp": 1.02526307,
      "epoch": 0.24036553838754285,
      "flos": 57122351867520.0,
      "grad_norm": 0.8263216891795258,
      "language_loss": 0.57880324,
      "learning_rate": 3.553621731668408e-06,
      "loss": 0.60027814,
      "num_input_tokens_seen": 42673595,
      "step": 1999,
      "time_per_iteration": 3.084599256515503
    },
    {
      "auxiliary_loss_clip": 0.01190031,
      "auxiliary_loss_mlp": 0.0103552,
      "balance_loss_clip": 1.05644226,
      "balance_loss_mlp": 1.02538168,
      "epoch": 0.24048578127818193,
      "flos": 24969946158720.0,
      "grad_norm": 2.173614219858502,
      "language_loss": 0.83147275,
      "learning_rate": 3.553131068927688e-06,
      "loss": 0.85372829,
      "num_input_tokens_seen": 42692000,
      "step": 2000,
      "time_per_iteration": 2.5484702587127686
    },
    {
      "auxiliary_loss_clip": 0.01168126,
      "auxiliary_loss_mlp": 0.01033877,
      "balance_loss_clip": 1.0582186,
      "balance_loss_mlp": 1.02546728,
      "epoch": 0.24060602416882101,
      "flos": 23330947547520.0,
      "grad_norm": 1.7355793923444756,
      "language_loss": 0.80381954,
      "learning_rate": 3.552640170578219e-06,
      "loss": 0.82583958,
      "num_input_tokens_seen": 42712250,
      "step": 2001,
      "time_per_iteration": 2.5578243732452393
    },
    {
      "auxiliary_loss_clip": 0.01184208,
      "auxiliary_loss_mlp": 0.01042265,
      "balance_loss_clip": 1.0605073,
      "balance_loss_mlp": 1.03318691,
      "epoch": 0.2407262670594601,
      "flos": 14173260128640.0,
      "grad_norm": 2.11599837347217,
      "language_loss": 0.77785522,
      "learning_rate": 3.5521490366944703e-06,
      "loss": 0.80012,
      "num_input_tokens_seen": 42729900,
      "step": 2002,
      "time_per_iteration": 2.518450975418091
    },
    {
      "auxiliary_loss_clip": 0.01169127,
      "auxiliary_loss_mlp": 0.0103176,
      "balance_loss_clip": 1.05627108,
      "balance_loss_mlp": 1.02252769,
      "epoch": 0.2408465099500992,
      "flos": 13663113217920.0,
      "grad_norm": 2.6119038656090794,
      "language_loss": 0.79861313,
      "learning_rate": 3.5516576673509474e-06,
      "loss": 0.82062197,
      "num_input_tokens_seen": 42747900,
      "step": 2003,
      "time_per_iteration": 2.53810977935791
    },
    {
      "auxiliary_loss_clip": 0.01215718,
      "auxiliary_loss_mlp": 0.01038508,
      "balance_loss_clip": 1.06410551,
      "balance_loss_mlp": 1.0290246,
      "epoch": 0.2409667528407383,
      "flos": 31248029076480.0,
      "grad_norm": 2.0195210994399795,
      "language_loss": 0.86382061,
      "learning_rate": 3.5511660626221896e-06,
      "loss": 0.88636291,
      "num_input_tokens_seen": 42768540,
      "step": 2004,
      "time_per_iteration": 2.553990364074707
    },
    {
      "auxiliary_loss_clip": 0.01182337,
      "auxiliary_loss_mlp": 0.00764808,
      "balance_loss_clip": 1.06022203,
      "balance_loss_mlp": 1.00137043,
      "epoch": 0.24108699573137737,
      "flos": 22199941031040.0,
      "grad_norm": 2.370699556953951,
      "language_loss": 0.89387584,
      "learning_rate": 3.5506742225827744e-06,
      "loss": 0.9133473,
      "num_input_tokens_seen": 42785395,
      "step": 2005,
      "time_per_iteration": 2.5323922634124756
    },
    {
      "auxiliary_loss_clip": 0.01170277,
      "auxiliary_loss_mlp": 0.01036954,
      "balance_loss_clip": 1.05847931,
      "balance_loss_mlp": 1.02695262,
      "epoch": 0.24120723862201648,
      "flos": 26103035664000.0,
      "grad_norm": 2.1499744919736767,
      "language_loss": 0.9029125,
      "learning_rate": 3.5501821473073116e-06,
      "loss": 0.92498481,
      "num_input_tokens_seen": 42801980,
      "step": 2006,
      "time_per_iteration": 2.579784870147705
    },
    {
      "auxiliary_loss_clip": 0.01162718,
      "auxiliary_loss_mlp": 0.01042249,
      "balance_loss_clip": 1.05691576,
      "balance_loss_mlp": 1.03118014,
      "epoch": 0.24132748151265557,
      "flos": 18624926246400.0,
      "grad_norm": 2.021978399084634,
      "language_loss": 0.86715221,
      "learning_rate": 3.54968983687045e-06,
      "loss": 0.88920188,
      "num_input_tokens_seen": 42818850,
      "step": 2007,
      "time_per_iteration": 2.555131196975708
    },
    {
      "auxiliary_loss_clip": 0.0118702,
      "auxiliary_loss_mlp": 0.0104759,
      "balance_loss_clip": 1.06143117,
      "balance_loss_mlp": 1.03646731,
      "epoch": 0.24144772440329465,
      "flos": 15267673664640.0,
      "grad_norm": 2.45496738679731,
      "language_loss": 0.88954461,
      "learning_rate": 3.549197291346872e-06,
      "loss": 0.91189069,
      "num_input_tokens_seen": 42835375,
      "step": 2008,
      "time_per_iteration": 2.5052084922790527
    },
    {
      "auxiliary_loss_clip": 0.01200798,
      "auxiliary_loss_mlp": 0.01038247,
      "balance_loss_clip": 1.06058848,
      "balance_loss_mlp": 1.02877605,
      "epoch": 0.24156796729393373,
      "flos": 24024274842240.0,
      "grad_norm": 2.2877681148079785,
      "language_loss": 0.79204863,
      "learning_rate": 3.548704510811297e-06,
      "loss": 0.81443906,
      "num_input_tokens_seen": 42854570,
      "step": 2009,
      "time_per_iteration": 2.513949155807495
    },
    {
      "auxiliary_loss_clip": 0.01159791,
      "auxiliary_loss_mlp": 0.0104702,
      "balance_loss_clip": 1.05521989,
      "balance_loss_mlp": 1.03644013,
      "epoch": 0.24168821018457284,
      "flos": 26286790665600.0,
      "grad_norm": 2.1958659209856344,
      "language_loss": 0.75249618,
      "learning_rate": 3.5482114953384787e-06,
      "loss": 0.77456427,
      "num_input_tokens_seen": 42873800,
      "step": 2010,
      "time_per_iteration": 2.5988929271698
    },
    {
      "auxiliary_loss_clip": 0.01203421,
      "auxiliary_loss_mlp": 0.01041529,
      "balance_loss_clip": 1.06248856,
      "balance_loss_mlp": 1.03088951,
      "epoch": 0.24180845307521193,
      "flos": 18223193560320.0,
      "grad_norm": 3.09569395858439,
      "language_loss": 0.84602809,
      "learning_rate": 3.5477182450032077e-06,
      "loss": 0.86847758,
      "num_input_tokens_seen": 42892400,
      "step": 2011,
      "time_per_iteration": 2.4890429973602295
    },
    {
      "auxiliary_loss_clip": 0.01196911,
      "auxiliary_loss_mlp": 0.01042508,
      "balance_loss_clip": 1.06060362,
      "balance_loss_mlp": 1.03245819,
      "epoch": 0.241928695965851,
      "flos": 20449260057600.0,
      "grad_norm": 2.4190500559330257,
      "language_loss": 0.83505142,
      "learning_rate": 3.5472247598803097e-06,
      "loss": 0.85744566,
      "num_input_tokens_seen": 42911745,
      "step": 2012,
      "time_per_iteration": 2.4939379692077637
    },
    {
      "auxiliary_loss_clip": 0.01216312,
      "auxiliary_loss_mlp": 0.01041645,
      "balance_loss_clip": 1.06358266,
      "balance_loss_mlp": 1.03129733,
      "epoch": 0.24204893885649012,
      "flos": 25556475340800.0,
      "grad_norm": 2.5065725840685147,
      "language_loss": 0.85111445,
      "learning_rate": 3.546731040044645e-06,
      "loss": 0.873694,
      "num_input_tokens_seen": 42926915,
      "step": 2013,
      "time_per_iteration": 2.5154919624328613
    },
    {
      "auxiliary_loss_clip": 0.01214434,
      "auxiliary_loss_mlp": 0.01029225,
      "balance_loss_clip": 1.06319857,
      "balance_loss_mlp": 1.01926494,
      "epoch": 0.2421691817471292,
      "flos": 30660207004800.0,
      "grad_norm": 1.8008223679215087,
      "language_loss": 0.75196111,
      "learning_rate": 3.546237085571112e-06,
      "loss": 0.77439767,
      "num_input_tokens_seen": 42945350,
      "step": 2014,
      "time_per_iteration": 2.5209712982177734
    },
    {
      "auxiliary_loss_clip": 0.01201106,
      "auxiliary_loss_mlp": 0.01034831,
      "balance_loss_clip": 1.06409693,
      "balance_loss_mlp": 1.02533054,
      "epoch": 0.24228942463776829,
      "flos": 21945011230080.0,
      "grad_norm": 6.148319701296482,
      "language_loss": 0.72269219,
      "learning_rate": 3.5457428965346425e-06,
      "loss": 0.74505156,
      "num_input_tokens_seen": 42964290,
      "step": 2015,
      "time_per_iteration": 2.5096564292907715
    },
    {
      "auxiliary_loss_clip": 0.01137212,
      "auxiliary_loss_mlp": 0.01034901,
      "balance_loss_clip": 1.05390656,
      "balance_loss_mlp": 1.02496481,
      "epoch": 0.2424096675284074,
      "flos": 33984493879680.0,
      "grad_norm": 1.5657157676524969,
      "language_loss": 0.74686772,
      "learning_rate": 3.545248473010205e-06,
      "loss": 0.7685889,
      "num_input_tokens_seen": 42987095,
      "step": 2016,
      "time_per_iteration": 2.7369117736816406
    },
    {
      "auxiliary_loss_clip": 0.01218534,
      "auxiliary_loss_mlp": 0.00765327,
      "balance_loss_clip": 1.06256843,
      "balance_loss_mlp": 1.00142455,
      "epoch": 0.24252991041904648,
      "flos": 21653416621440.0,
      "grad_norm": 1.6743571227131462,
      "language_loss": 0.87753975,
      "learning_rate": 3.544753815072802e-06,
      "loss": 0.89737833,
      "num_input_tokens_seen": 43005750,
      "step": 2017,
      "time_per_iteration": 2.492324113845825
    },
    {
      "auxiliary_loss_clip": 0.01114044,
      "auxiliary_loss_mlp": 0.01033153,
      "balance_loss_clip": 1.04609799,
      "balance_loss_mlp": 1.02315176,
      "epoch": 0.24265015330968556,
      "flos": 21870065502720.0,
      "grad_norm": 1.9727245398722877,
      "language_loss": 0.88129169,
      "learning_rate": 3.544258922797474e-06,
      "loss": 0.90276372,
      "num_input_tokens_seen": 43023870,
      "step": 2018,
      "time_per_iteration": 3.4754931926727295
    },
    {
      "auxiliary_loss_clip": 0.01213558,
      "auxiliary_loss_mlp": 0.01036534,
      "balance_loss_clip": 1.06346726,
      "balance_loss_mlp": 1.02721834,
      "epoch": 0.24277039620032465,
      "flos": 25628260671360.0,
      "grad_norm": 1.5793490770632446,
      "language_loss": 0.77898896,
      "learning_rate": 3.543763796259295e-06,
      "loss": 0.80148989,
      "num_input_tokens_seen": 43043825,
      "step": 2019,
      "time_per_iteration": 2.499797821044922
    },
    {
      "auxiliary_loss_clip": 0.0119975,
      "auxiliary_loss_mlp": 0.01038526,
      "balance_loss_clip": 1.06118834,
      "balance_loss_mlp": 1.02820837,
      "epoch": 0.24289063909096376,
      "flos": 26286575184000.0,
      "grad_norm": 1.7661350007665524,
      "language_loss": 0.90859336,
      "learning_rate": 3.5432684355333754e-06,
      "loss": 0.93097615,
      "num_input_tokens_seen": 43062480,
      "step": 2020,
      "time_per_iteration": 2.6600663661956787
    },
    {
      "auxiliary_loss_clip": 0.01199005,
      "auxiliary_loss_mlp": 0.0103918,
      "balance_loss_clip": 1.05935287,
      "balance_loss_mlp": 1.02935088,
      "epoch": 0.24301088198160284,
      "flos": 25075056332160.0,
      "grad_norm": 4.423105025546754,
      "language_loss": 0.76537555,
      "learning_rate": 3.5427728406948613e-06,
      "loss": 0.7877574,
      "num_input_tokens_seen": 43081595,
      "step": 2021,
      "time_per_iteration": 2.519862651824951
    },
    {
      "auxiliary_loss_clip": 0.01106161,
      "auxiliary_loss_mlp": 0.01016923,
      "balance_loss_clip": 1.04079425,
      "balance_loss_mlp": 1.01457453,
      "epoch": 0.24313112487224192,
      "flos": 69900948673920.0,
      "grad_norm": 0.7546676576833723,
      "language_loss": 0.57935798,
      "learning_rate": 3.542277011818934e-06,
      "loss": 0.60058886,
      "num_input_tokens_seen": 43145430,
      "step": 2022,
      "time_per_iteration": 3.404602289199829
    },
    {
      "auxiliary_loss_clip": 0.01186284,
      "auxiliary_loss_mlp": 0.01035743,
      "balance_loss_clip": 1.06215978,
      "balance_loss_mlp": 1.02656412,
      "epoch": 0.24325136776288103,
      "flos": 40662334235520.0,
      "grad_norm": 2.093529315317973,
      "language_loss": 0.74164367,
      "learning_rate": 3.5417809489808104e-06,
      "loss": 0.76386392,
      "num_input_tokens_seen": 43167040,
      "step": 2023,
      "time_per_iteration": 3.550360679626465
    },
    {
      "auxiliary_loss_clip": 0.01201141,
      "auxiliary_loss_mlp": 0.01035041,
      "balance_loss_clip": 1.06249368,
      "balance_loss_mlp": 1.02625513,
      "epoch": 0.24337161065352012,
      "flos": 25046400257280.0,
      "grad_norm": 2.049317021195419,
      "language_loss": 0.72529352,
      "learning_rate": 3.5412846522557422e-06,
      "loss": 0.74765539,
      "num_input_tokens_seen": 43187930,
      "step": 2024,
      "time_per_iteration": 3.3872809410095215
    },
    {
      "auxiliary_loss_clip": 0.01214038,
      "auxiliary_loss_mlp": 0.01037638,
      "balance_loss_clip": 1.06300068,
      "balance_loss_mlp": 1.02782702,
      "epoch": 0.2434918535441592,
      "flos": 18661160090880.0,
      "grad_norm": 2.0911263302162486,
      "language_loss": 0.73977381,
      "learning_rate": 3.540788121719018e-06,
      "loss": 0.7622906,
      "num_input_tokens_seen": 43206350,
      "step": 2025,
      "time_per_iteration": 3.2318644523620605
    },
    {
      "auxiliary_loss_clip": 0.01161845,
      "auxiliary_loss_mlp": 0.01038611,
      "balance_loss_clip": 1.05774343,
      "balance_loss_mlp": 1.02838254,
      "epoch": 0.24361209643479828,
      "flos": 23915142345600.0,
      "grad_norm": 1.96263399866518,
      "language_loss": 0.81964219,
      "learning_rate": 3.5402913574459604e-06,
      "loss": 0.84164679,
      "num_input_tokens_seen": 43226255,
      "step": 2026,
      "time_per_iteration": 2.556619882583618
    },
    {
      "auxiliary_loss_clip": 0.01131489,
      "auxiliary_loss_mlp": 0.01037101,
      "balance_loss_clip": 1.0484376,
      "balance_loss_mlp": 1.02795196,
      "epoch": 0.2437323393254374,
      "flos": 28657505232000.0,
      "grad_norm": 1.5884391313971205,
      "language_loss": 0.86278164,
      "learning_rate": 3.5397943595119297e-06,
      "loss": 0.8844676,
      "num_input_tokens_seen": 43247675,
      "step": 2027,
      "time_per_iteration": 2.677842378616333
    },
    {
      "auxiliary_loss_clip": 0.01177597,
      "auxiliary_loss_mlp": 0.01037646,
      "balance_loss_clip": 1.05822158,
      "balance_loss_mlp": 1.02702475,
      "epoch": 0.24385258221607647,
      "flos": 23550325862400.0,
      "grad_norm": 2.9476236891476213,
      "language_loss": 0.77395499,
      "learning_rate": 3.5392971279923177e-06,
      "loss": 0.79610747,
      "num_input_tokens_seen": 43265895,
      "step": 2028,
      "time_per_iteration": 2.5468316078186035
    },
    {
      "auxiliary_loss_clip": 0.01161203,
      "auxiliary_loss_mlp": 0.01033426,
      "balance_loss_clip": 1.05382395,
      "balance_loss_mlp": 1.02217269,
      "epoch": 0.24397282510671556,
      "flos": 25336091445120.0,
      "grad_norm": 2.155010711190702,
      "language_loss": 0.82711738,
      "learning_rate": 3.5387996629625557e-06,
      "loss": 0.84906363,
      "num_input_tokens_seen": 43283485,
      "step": 2029,
      "time_per_iteration": 2.5875539779663086
    },
    {
      "auxiliary_loss_clip": 0.01136193,
      "auxiliary_loss_mlp": 0.01001549,
      "balance_loss_clip": 1.04746079,
      "balance_loss_mlp": 0.99948645,
      "epoch": 0.24409306799735467,
      "flos": 65187421430400.0,
      "grad_norm": 0.810109331667588,
      "language_loss": 0.55020881,
      "learning_rate": 3.5383019644981083e-06,
      "loss": 0.57158619,
      "num_input_tokens_seen": 43347180,
      "step": 2030,
      "time_per_iteration": 3.0763237476348877
    },
    {
      "auxiliary_loss_clip": 0.01182364,
      "auxiliary_loss_mlp": 0.01027815,
      "balance_loss_clip": 1.05807924,
      "balance_loss_mlp": 1.01784921,
      "epoch": 0.24421331088799375,
      "flos": 19537093152000.0,
      "grad_norm": 2.361511047805532,
      "language_loss": 0.72730899,
      "learning_rate": 3.5378040326744763e-06,
      "loss": 0.74941087,
      "num_input_tokens_seen": 43366665,
      "step": 2031,
      "time_per_iteration": 2.498952865600586
    },
    {
      "auxiliary_loss_clip": 0.01170289,
      "auxiliary_loss_mlp": 0.01033252,
      "balance_loss_clip": 1.05861604,
      "balance_loss_mlp": 1.02385211,
      "epoch": 0.24433355377863283,
      "flos": 21068575378560.0,
      "grad_norm": 2.4674421774485764,
      "language_loss": 0.85472143,
      "learning_rate": 3.5373058675671946e-06,
      "loss": 0.87675679,
      "num_input_tokens_seen": 43384670,
      "step": 2032,
      "time_per_iteration": 2.543724536895752
    },
    {
      "auxiliary_loss_clip": 0.0114541,
      "auxiliary_loss_mlp": 0.01033524,
      "balance_loss_clip": 1.05329907,
      "balance_loss_mlp": 1.02300346,
      "epoch": 0.24445379666927192,
      "flos": 22637189289600.0,
      "grad_norm": 2.7095208164151057,
      "language_loss": 0.72379386,
      "learning_rate": 3.536807469251836e-06,
      "loss": 0.74558318,
      "num_input_tokens_seen": 43403825,
      "step": 2033,
      "time_per_iteration": 2.586548328399658
    },
    {
      "auxiliary_loss_clip": 0.01172252,
      "auxiliary_loss_mlp": 0.01031929,
      "balance_loss_clip": 1.05458188,
      "balance_loss_mlp": 1.0223093,
      "epoch": 0.24457403955991103,
      "flos": 21251612108160.0,
      "grad_norm": 3.4754204057747735,
      "language_loss": 0.82476127,
      "learning_rate": 3.5363088378040055e-06,
      "loss": 0.84680307,
      "num_input_tokens_seen": 43422715,
      "step": 2034,
      "time_per_iteration": 2.580615997314453
    },
    {
      "auxiliary_loss_clip": 0.01133405,
      "auxiliary_loss_mlp": 0.00754292,
      "balance_loss_clip": 1.04519033,
      "balance_loss_mlp": 1.0009644,
      "epoch": 0.2446942824505501,
      "flos": 66997820764800.0,
      "grad_norm": 0.7546160432939819,
      "language_loss": 0.64419895,
      "learning_rate": 3.5358099732993463e-06,
      "loss": 0.66307592,
      "num_input_tokens_seen": 43481825,
      "step": 2035,
      "time_per_iteration": 2.9549691677093506
    },
    {
      "auxiliary_loss_clip": 0.01188764,
      "auxiliary_loss_mlp": 0.01033974,
      "balance_loss_clip": 1.05916369,
      "balance_loss_mlp": 1.02451491,
      "epoch": 0.2448145253411892,
      "flos": 20411122792320.0,
      "grad_norm": 4.0376159961893725,
      "language_loss": 0.89357877,
      "learning_rate": 3.535310875813535e-06,
      "loss": 0.91580617,
      "num_input_tokens_seen": 43500220,
      "step": 2036,
      "time_per_iteration": 2.54103422164917
    },
    {
      "auxiliary_loss_clip": 0.01195911,
      "auxiliary_loss_mlp": 0.01032595,
      "balance_loss_clip": 1.05919862,
      "balance_loss_mlp": 1.02315331,
      "epoch": 0.2449347682318283,
      "flos": 28804739080320.0,
      "grad_norm": 2.136723515496185,
      "language_loss": 0.81929231,
      "learning_rate": 3.5348115454222843e-06,
      "loss": 0.84157741,
      "num_input_tokens_seen": 43522805,
      "step": 2037,
      "time_per_iteration": 2.5451924800872803
    },
    {
      "auxiliary_loss_clip": 0.01178184,
      "auxiliary_loss_mlp": 0.01046084,
      "balance_loss_clip": 1.05475402,
      "balance_loss_mlp": 1.03624344,
      "epoch": 0.2450550111224674,
      "flos": 22528990546560.0,
      "grad_norm": 2.1410221616501515,
      "language_loss": 0.85668814,
      "learning_rate": 3.5343119822013425e-06,
      "loss": 0.87893081,
      "num_input_tokens_seen": 43541915,
      "step": 2038,
      "time_per_iteration": 2.524336338043213
    },
    {
      "auxiliary_loss_clip": 0.01204193,
      "auxiliary_loss_mlp": 0.01043001,
      "balance_loss_clip": 1.060987,
      "balance_loss_mlp": 1.03256416,
      "epoch": 0.24517525401310647,
      "flos": 21759137326080.0,
      "grad_norm": 3.1350782773695625,
      "language_loss": 0.77415615,
      "learning_rate": 3.533812186226493e-06,
      "loss": 0.79662812,
      "num_input_tokens_seen": 43562625,
      "step": 2039,
      "time_per_iteration": 2.4927306175231934
    },
    {
      "auxiliary_loss_clip": 0.01207834,
      "auxiliary_loss_mlp": 0.01031357,
      "balance_loss_clip": 1.05925858,
      "balance_loss_mlp": 1.0221417,
      "epoch": 0.24529549690374555,
      "flos": 25043311687680.0,
      "grad_norm": 1.976807956764818,
      "language_loss": 0.7574625,
      "learning_rate": 3.5333121575735545e-06,
      "loss": 0.77985442,
      "num_input_tokens_seen": 43582265,
      "step": 2040,
      "time_per_iteration": 2.467381715774536
    },
    {
      "auxiliary_loss_clip": 0.01179493,
      "auxiliary_loss_mlp": 0.01033534,
      "balance_loss_clip": 1.05795443,
      "balance_loss_mlp": 1.02340174,
      "epoch": 0.24541573979438466,
      "flos": 32123638915200.0,
      "grad_norm": 2.876660177603598,
      "language_loss": 0.75867587,
      "learning_rate": 3.532811896318381e-06,
      "loss": 0.78080618,
      "num_input_tokens_seen": 43604335,
      "step": 2041,
      "time_per_iteration": 2.621657609939575
    },
    {
      "auxiliary_loss_clip": 0.01170441,
      "auxiliary_loss_mlp": 0.01029322,
      "balance_loss_clip": 1.05528152,
      "balance_loss_mlp": 1.01926112,
      "epoch": 0.24553598268502375,
      "flos": 31357556622720.0,
      "grad_norm": 2.0825100759679556,
      "language_loss": 0.81932783,
      "learning_rate": 3.5323114025368615e-06,
      "loss": 0.84132552,
      "num_input_tokens_seen": 43619400,
      "step": 2042,
      "time_per_iteration": 2.6108646392822266
    },
    {
      "auxiliary_loss_clip": 0.01189871,
      "auxiliary_loss_mlp": 0.01030265,
      "balance_loss_clip": 1.05483294,
      "balance_loss_mlp": 1.02075195,
      "epoch": 0.24565622557566283,
      "flos": 14027462824320.0,
      "grad_norm": 2.184648304099261,
      "language_loss": 0.8168236,
      "learning_rate": 3.53181067630492e-06,
      "loss": 0.8390249,
      "num_input_tokens_seen": 43636870,
      "step": 2043,
      "time_per_iteration": 2.488929510116577
    },
    {
      "auxiliary_loss_clip": 0.01172107,
      "auxiliary_loss_mlp": 0.01039677,
      "balance_loss_clip": 1.0547843,
      "balance_loss_mlp": 1.03005052,
      "epoch": 0.24577646846630194,
      "flos": 16581465515520.0,
      "grad_norm": 1.8214350373296109,
      "language_loss": 0.76064312,
      "learning_rate": 3.5313097176985175e-06,
      "loss": 0.78276092,
      "num_input_tokens_seen": 43655180,
      "step": 2044,
      "time_per_iteration": 2.494130849838257
    },
    {
      "auxiliary_loss_clip": 0.01195588,
      "auxiliary_loss_mlp": 0.0103033,
      "balance_loss_clip": 1.05939984,
      "balance_loss_mlp": 1.02100754,
      "epoch": 0.24589671135694102,
      "flos": 18807424272000.0,
      "grad_norm": 1.8690621047120717,
      "language_loss": 0.8122102,
      "learning_rate": 3.5308085267936482e-06,
      "loss": 0.83446932,
      "num_input_tokens_seen": 43672895,
      "step": 2045,
      "time_per_iteration": 3.2762341499328613
    },
    {
      "auxiliary_loss_clip": 0.01135645,
      "auxiliary_loss_mlp": 0.00763297,
      "balance_loss_clip": 1.05259132,
      "balance_loss_mlp": 1.00127506,
      "epoch": 0.2460169542475801,
      "flos": 19938538529280.0,
      "grad_norm": 1.7637586934780578,
      "language_loss": 0.89747202,
      "learning_rate": 3.530307103666342e-06,
      "loss": 0.91646147,
      "num_input_tokens_seen": 43691975,
      "step": 2046,
      "time_per_iteration": 2.6238620281219482
    },
    {
      "auxiliary_loss_clip": 0.01172407,
      "auxiliary_loss_mlp": 0.01029838,
      "balance_loss_clip": 1.05668688,
      "balance_loss_mlp": 1.02040827,
      "epoch": 0.24613719713821922,
      "flos": 24171221381760.0,
      "grad_norm": 1.918258230136605,
      "language_loss": 0.80333829,
      "learning_rate": 3.5298054483926658e-06,
      "loss": 0.82536077,
      "num_input_tokens_seen": 43712670,
      "step": 2047,
      "time_per_iteration": 2.558009386062622
    },
    {
      "auxiliary_loss_clip": 0.01204154,
      "auxiliary_loss_mlp": 0.01034855,
      "balance_loss_clip": 1.06022596,
      "balance_loss_mlp": 1.02528846,
      "epoch": 0.2462574400288583,
      "flos": 30221055325440.0,
      "grad_norm": 1.946700773247387,
      "language_loss": 0.83041006,
      "learning_rate": 3.5293035610487187e-06,
      "loss": 0.85280013,
      "num_input_tokens_seen": 43732035,
      "step": 2048,
      "time_per_iteration": 2.5432701110839844
    },
    {
      "auxiliary_loss_clip": 0.01094238,
      "auxiliary_loss_mlp": 0.0100003,
      "balance_loss_clip": 1.03473091,
      "balance_loss_mlp": 0.99782497,
      "epoch": 0.24637768291949738,
      "flos": 68943030819840.0,
      "grad_norm": 0.7355315456194561,
      "language_loss": 0.62013125,
      "learning_rate": 3.5288014417106374e-06,
      "loss": 0.64107394,
      "num_input_tokens_seen": 43798055,
      "step": 2049,
      "time_per_iteration": 3.1210954189300537
    },
    {
      "auxiliary_loss_clip": 0.01164488,
      "auxiliary_loss_mlp": 0.01034967,
      "balance_loss_clip": 1.05463457,
      "balance_loss_mlp": 1.02562737,
      "epoch": 0.24649792581013646,
      "flos": 34383999922560.0,
      "grad_norm": 1.8033945803821139,
      "language_loss": 0.75461221,
      "learning_rate": 3.528299090454593e-06,
      "loss": 0.77660674,
      "num_input_tokens_seen": 43818590,
      "step": 2050,
      "time_per_iteration": 3.580585241317749
    },
    {
      "auxiliary_loss_clip": 0.01198203,
      "auxiliary_loss_mlp": 0.01031496,
      "balance_loss_clip": 1.05714953,
      "balance_loss_mlp": 1.02179265,
      "epoch": 0.24661816870077558,
      "flos": 19680448331520.0,
      "grad_norm": 2.4760690033290906,
      "language_loss": 0.8288486,
      "learning_rate": 3.527796507356792e-06,
      "loss": 0.85114563,
      "num_input_tokens_seen": 43832480,
      "step": 2051,
      "time_per_iteration": 3.2736921310424805
    },
    {
      "auxiliary_loss_clip": 0.01198782,
      "auxiliary_loss_mlp": 0.01035215,
      "balance_loss_clip": 1.05769539,
      "balance_loss_mlp": 1.02580333,
      "epoch": 0.24673841159141466,
      "flos": 20002279213440.0,
      "grad_norm": 2.8511891206930087,
      "language_loss": 0.90640426,
      "learning_rate": 3.527293692493475e-06,
      "loss": 0.9287442,
      "num_input_tokens_seen": 43848345,
      "step": 2052,
      "time_per_iteration": 3.175246000289917
    },
    {
      "auxiliary_loss_clip": 0.01197318,
      "auxiliary_loss_mlp": 0.01034074,
      "balance_loss_clip": 1.05705571,
      "balance_loss_mlp": 1.02398872,
      "epoch": 0.24685865448205374,
      "flos": 21646593037440.0,
      "grad_norm": 4.698295329890442,
      "language_loss": 0.73035491,
      "learning_rate": 3.52679064594092e-06,
      "loss": 0.75266886,
      "num_input_tokens_seen": 43865685,
      "step": 2053,
      "time_per_iteration": 2.487556219100952
    },
    {
      "auxiliary_loss_clip": 0.01135895,
      "auxiliary_loss_mlp": 0.01034459,
      "balance_loss_clip": 1.04321694,
      "balance_loss_mlp": 1.02548802,
      "epoch": 0.24697889737269285,
      "flos": 17960470508160.0,
      "grad_norm": 2.366456713382833,
      "language_loss": 0.74841893,
      "learning_rate": 3.5262873677754375e-06,
      "loss": 0.77012247,
      "num_input_tokens_seen": 43883690,
      "step": 2054,
      "time_per_iteration": 2.579397439956665
    },
    {
      "auxiliary_loss_clip": 0.01205296,
      "auxiliary_loss_mlp": 0.01033834,
      "balance_loss_clip": 1.05769897,
      "balance_loss_mlp": 1.02456522,
      "epoch": 0.24709914026333193,
      "flos": 27344611221120.0,
      "grad_norm": 1.5794420879481634,
      "language_loss": 0.80695426,
      "learning_rate": 3.5257838580733745e-06,
      "loss": 0.82934558,
      "num_input_tokens_seen": 43903295,
      "step": 2055,
      "time_per_iteration": 2.513293743133545
    },
    {
      "auxiliary_loss_clip": 0.01198354,
      "auxiliary_loss_mlp": 0.01034304,
      "balance_loss_clip": 1.05797529,
      "balance_loss_mlp": 1.02474952,
      "epoch": 0.24721938315397102,
      "flos": 19275519335040.0,
      "grad_norm": 2.2452749283039934,
      "language_loss": 0.87138367,
      "learning_rate": 3.5252801169111138e-06,
      "loss": 0.89371026,
      "num_input_tokens_seen": 43920960,
      "step": 2056,
      "time_per_iteration": 2.4872045516967773
    },
    {
      "auxiliary_loss_clip": 0.01176834,
      "auxiliary_loss_mlp": 0.01033213,
      "balance_loss_clip": 1.05707264,
      "balance_loss_mlp": 1.02405167,
      "epoch": 0.2473396260446101,
      "flos": 23185796688000.0,
      "grad_norm": 1.7779788533585361,
      "language_loss": 0.79825628,
      "learning_rate": 3.524776144365072e-06,
      "loss": 0.82035679,
      "num_input_tokens_seen": 43939415,
      "step": 2057,
      "time_per_iteration": 2.544431209564209
    },
    {
      "auxiliary_loss_clip": 0.01172434,
      "auxiliary_loss_mlp": 0.01037819,
      "balance_loss_clip": 1.05627775,
      "balance_loss_mlp": 1.02811563,
      "epoch": 0.2474598689352492,
      "flos": 21142443697920.0,
      "grad_norm": 1.747216171799763,
      "language_loss": 0.79203486,
      "learning_rate": 3.5242719405117016e-06,
      "loss": 0.8141374,
      "num_input_tokens_seen": 43959220,
      "step": 2058,
      "time_per_iteration": 2.553508996963501
    },
    {
      "auxiliary_loss_clip": 0.01183398,
      "auxiliary_loss_mlp": 0.00764941,
      "balance_loss_clip": 1.05795598,
      "balance_loss_mlp": 1.00136673,
      "epoch": 0.2475801118258883,
      "flos": 21648352803840.0,
      "grad_norm": 2.4071583400354455,
      "language_loss": 0.75201744,
      "learning_rate": 3.5237675054274893e-06,
      "loss": 0.77150083,
      "num_input_tokens_seen": 43978420,
      "step": 2059,
      "time_per_iteration": 2.564394235610962
    },
    {
      "auxiliary_loss_clip": 0.01194892,
      "auxiliary_loss_mlp": 0.01036708,
      "balance_loss_clip": 1.05726445,
      "balance_loss_mlp": 1.02686167,
      "epoch": 0.24770035471652738,
      "flos": 22674500542080.0,
      "grad_norm": 1.9041185073414162,
      "language_loss": 0.80239081,
      "learning_rate": 3.5232628391889584e-06,
      "loss": 0.82470685,
      "num_input_tokens_seen": 43996710,
      "step": 2060,
      "time_per_iteration": 2.5405423641204834
    },
    {
      "auxiliary_loss_clip": 0.01147297,
      "auxiliary_loss_mlp": 0.01029758,
      "balance_loss_clip": 1.05293083,
      "balance_loss_mlp": 1.02075791,
      "epoch": 0.2478205976071665,
      "flos": 22163814927360.0,
      "grad_norm": 2.552710114290554,
      "language_loss": 0.64309239,
      "learning_rate": 3.522757941872666e-06,
      "loss": 0.66486293,
      "num_input_tokens_seen": 44014865,
      "step": 2061,
      "time_per_iteration": 2.6115903854370117
    },
    {
      "auxiliary_loss_clip": 0.0121193,
      "auxiliary_loss_mlp": 0.00764735,
      "balance_loss_clip": 1.06299806,
      "balance_loss_mlp": 1.00151336,
      "epoch": 0.24794084049780557,
      "flos": 24973106555520.0,
      "grad_norm": 1.9459748099389635,
      "language_loss": 0.82584316,
      "learning_rate": 3.5222528135552042e-06,
      "loss": 0.84560978,
      "num_input_tokens_seen": 44036325,
      "step": 2062,
      "time_per_iteration": 2.4948155879974365
    },
    {
      "auxiliary_loss_clip": 0.01193669,
      "auxiliary_loss_mlp": 0.01038304,
      "balance_loss_clip": 1.05974042,
      "balance_loss_mlp": 1.02858233,
      "epoch": 0.24806108338844465,
      "flos": 18296379521280.0,
      "grad_norm": 2.0349655137334497,
      "language_loss": 0.80289996,
      "learning_rate": 3.521747454313201e-06,
      "loss": 0.82521963,
      "num_input_tokens_seen": 44055005,
      "step": 2063,
      "time_per_iteration": 2.5140719413757324
    },
    {
      "auxiliary_loss_clip": 0.01154758,
      "auxiliary_loss_mlp": 0.01030696,
      "balance_loss_clip": 1.04835868,
      "balance_loss_mlp": 1.02133822,
      "epoch": 0.24818132627908374,
      "flos": 19282163351040.0,
      "grad_norm": 1.8442731566255053,
      "language_loss": 0.66645384,
      "learning_rate": 3.521241864223319e-06,
      "loss": 0.68830836,
      "num_input_tokens_seen": 44073965,
      "step": 2064,
      "time_per_iteration": 2.591148853302002
    },
    {
      "auxiliary_loss_clip": 0.01099617,
      "auxiliary_loss_mlp": 0.01012281,
      "balance_loss_clip": 1.03159118,
      "balance_loss_mlp": 1.00978971,
      "epoch": 0.24830156916972285,
      "flos": 70285837881600.0,
      "grad_norm": 0.7902384694965982,
      "language_loss": 0.62013566,
      "learning_rate": 3.5207360433622552e-06,
      "loss": 0.64125466,
      "num_input_tokens_seen": 44135965,
      "step": 2065,
      "time_per_iteration": 3.106696128845215
    },
    {
      "auxiliary_loss_clip": 0.0117789,
      "auxiliary_loss_mlp": 0.01036297,
      "balance_loss_clip": 1.05839241,
      "balance_loss_mlp": 1.02717733,
      "epoch": 0.24842181206036193,
      "flos": 40409128287360.0,
      "grad_norm": 2.7841112066361813,
      "language_loss": 0.74404866,
      "learning_rate": 3.5202299918067437e-06,
      "loss": 0.76619053,
      "num_input_tokens_seen": 44159560,
      "step": 2066,
      "time_per_iteration": 2.7093608379364014
    },
    {
      "auxiliary_loss_clip": 0.01191923,
      "auxiliary_loss_mlp": 0.01029238,
      "balance_loss_clip": 1.0574069,
      "balance_loss_mlp": 1.02025592,
      "epoch": 0.248542054951001,
      "flos": 20082432412800.0,
      "grad_norm": 2.520681619220884,
      "language_loss": 0.69171578,
      "learning_rate": 3.519723709633551e-06,
      "loss": 0.71392739,
      "num_input_tokens_seen": 44178320,
      "step": 2067,
      "time_per_iteration": 2.506783962249756
    },
    {
      "auxiliary_loss_clip": 0.01176716,
      "auxiliary_loss_mlp": 0.01034366,
      "balance_loss_clip": 1.05890632,
      "balance_loss_mlp": 1.02412593,
      "epoch": 0.24866229784164012,
      "flos": 23513948363520.0,
      "grad_norm": 2.2137245662377234,
      "language_loss": 0.83609009,
      "learning_rate": 3.519217196919479e-06,
      "loss": 0.85820091,
      "num_input_tokens_seen": 44197305,
      "step": 2068,
      "time_per_iteration": 2.6375486850738525
    },
    {
      "auxiliary_loss_clip": 0.01187801,
      "auxiliary_loss_mlp": 0.01036683,
      "balance_loss_clip": 1.06230497,
      "balance_loss_mlp": 1.0276525,
      "epoch": 0.2487825407322792,
      "flos": 19865101173120.0,
      "grad_norm": 1.7368374066876753,
      "language_loss": 0.73603821,
      "learning_rate": 3.518710453741367e-06,
      "loss": 0.75828302,
      "num_input_tokens_seen": 44216505,
      "step": 2069,
      "time_per_iteration": 2.523452043533325
    },
    {
      "auxiliary_loss_clip": 0.01171708,
      "auxiliary_loss_mlp": 0.0076475,
      "balance_loss_clip": 1.0539875,
      "balance_loss_mlp": 1.00146341,
      "epoch": 0.2489027836229183,
      "flos": 22017622573440.0,
      "grad_norm": 2.8041739847238407,
      "language_loss": 0.67833257,
      "learning_rate": 3.518203480176086e-06,
      "loss": 0.69769704,
      "num_input_tokens_seen": 44235435,
      "step": 2070,
      "time_per_iteration": 2.554170846939087
    },
    {
      "auxiliary_loss_clip": 0.01118645,
      "auxiliary_loss_mlp": 0.01041243,
      "balance_loss_clip": 1.04649854,
      "balance_loss_mlp": 1.03166485,
      "epoch": 0.2490230265135574,
      "flos": 23294354567040.0,
      "grad_norm": 1.7876822009766662,
      "language_loss": 0.80530941,
      "learning_rate": 3.517696276300545e-06,
      "loss": 0.82690835,
      "num_input_tokens_seen": 44256975,
      "step": 2071,
      "time_per_iteration": 2.6718549728393555
    },
    {
      "auxiliary_loss_clip": 0.01201026,
      "auxiliary_loss_mlp": 0.0103944,
      "balance_loss_clip": 1.06533027,
      "balance_loss_mlp": 1.02928329,
      "epoch": 0.24914326940419648,
      "flos": 19826784339840.0,
      "grad_norm": 2.690801197211992,
      "language_loss": 0.6918785,
      "learning_rate": 3.517188842191685e-06,
      "loss": 0.71428317,
      "num_input_tokens_seen": 44275125,
      "step": 2072,
      "time_per_iteration": 3.311549425125122
    },
    {
      "auxiliary_loss_clip": 0.01193792,
      "auxiliary_loss_mlp": 0.01030596,
      "balance_loss_clip": 1.05785728,
      "balance_loss_mlp": 1.02098203,
      "epoch": 0.24926351229483557,
      "flos": 20229271211520.0,
      "grad_norm": 1.6511159642335216,
      "language_loss": 0.73940581,
      "learning_rate": 3.5166811779264837e-06,
      "loss": 0.76164967,
      "num_input_tokens_seen": 44295445,
      "step": 2073,
      "time_per_iteration": 2.5805375576019287
    },
    {
      "auxiliary_loss_clip": 0.01209578,
      "auxiliary_loss_mlp": 0.01032976,
      "balance_loss_clip": 1.0592593,
      "balance_loss_mlp": 1.02302814,
      "epoch": 0.24938375518547465,
      "flos": 23294570048640.0,
      "grad_norm": 1.7831994857452753,
      "language_loss": 0.77649891,
      "learning_rate": 3.5161732835819545e-06,
      "loss": 0.79892445,
      "num_input_tokens_seen": 44314755,
      "step": 2074,
      "time_per_iteration": 2.480890989303589
    },
    {
      "auxiliary_loss_clip": 0.01213042,
      "auxiliary_loss_mlp": 0.01029485,
      "balance_loss_clip": 1.06322157,
      "balance_loss_mlp": 1.02025867,
      "epoch": 0.24950399807611376,
      "flos": 17311673099520.0,
      "grad_norm": 1.7685037465490576,
      "language_loss": 0.83107197,
      "learning_rate": 3.515665159235143e-06,
      "loss": 0.85349727,
      "num_input_tokens_seen": 44333640,
      "step": 2075,
      "time_per_iteration": 2.4774997234344482
    },
    {
      "auxiliary_loss_clip": 0.01172965,
      "auxiliary_loss_mlp": 0.01026145,
      "balance_loss_clip": 1.05004966,
      "balance_loss_mlp": 1.01793146,
      "epoch": 0.24962424096675284,
      "flos": 19024863252480.0,
      "grad_norm": 1.5841661672194396,
      "language_loss": 0.74937916,
      "learning_rate": 3.5151568049631318e-06,
      "loss": 0.77137029,
      "num_input_tokens_seen": 44352355,
      "step": 2076,
      "time_per_iteration": 2.553252696990967
    },
    {
      "auxiliary_loss_clip": 0.01209839,
      "auxiliary_loss_mlp": 0.01031618,
      "balance_loss_clip": 1.05929852,
      "balance_loss_mlp": 1.02175391,
      "epoch": 0.24974448385739192,
      "flos": 33398790710400.0,
      "grad_norm": 1.7465785249407617,
      "language_loss": 0.80122817,
      "learning_rate": 3.5146482208430385e-06,
      "loss": 0.82364279,
      "num_input_tokens_seen": 44374185,
      "step": 2077,
      "time_per_iteration": 3.498643159866333
    },
    {
      "auxiliary_loss_clip": 0.01127088,
      "auxiliary_loss_mlp": 0.0103391,
      "balance_loss_clip": 1.04837871,
      "balance_loss_mlp": 1.02331853,
      "epoch": 0.24986472674803104,
      "flos": 30007279532160.0,
      "grad_norm": 1.8325132885915663,
      "language_loss": 0.67639148,
      "learning_rate": 3.514139406952014e-06,
      "loss": 0.69800144,
      "num_input_tokens_seen": 44396210,
      "step": 2078,
      "time_per_iteration": 3.4114110469818115
    },
    {
      "auxiliary_loss_clip": 0.01195756,
      "auxiliary_loss_mlp": 0.01031752,
      "balance_loss_clip": 1.06046069,
      "balance_loss_mlp": 1.02257836,
      "epoch": 0.24998496963867012,
      "flos": 26613074833920.0,
      "grad_norm": 1.767743304935125,
      "language_loss": 0.83352041,
      "learning_rate": 3.5136303633672454e-06,
      "loss": 0.85579538,
      "num_input_tokens_seen": 44416340,
      "step": 2079,
      "time_per_iteration": 2.549199342727661
    },
    {
      "auxiliary_loss_clip": 0.01174866,
      "auxiliary_loss_mlp": 0.00764589,
      "balance_loss_clip": 1.05822933,
      "balance_loss_mlp": 1.00126028,
      "epoch": 0.25010521252930923,
      "flos": 23553989049600.0,
      "grad_norm": 1.763635891382621,
      "language_loss": 0.74358666,
      "learning_rate": 3.5131210901659544e-06,
      "loss": 0.76298118,
      "num_input_tokens_seen": 44438095,
      "step": 2080,
      "time_per_iteration": 2.6323556900024414
    },
    {
      "auxiliary_loss_clip": 0.01156598,
      "auxiliary_loss_mlp": 0.01032041,
      "balance_loss_clip": 1.0512861,
      "balance_loss_mlp": 1.02250409,
      "epoch": 0.2502254554199483,
      "flos": 23441193365760.0,
      "grad_norm": 2.245331530783253,
      "language_loss": 0.81737947,
      "learning_rate": 3.5126115874253967e-06,
      "loss": 0.83926582,
      "num_input_tokens_seen": 44457650,
      "step": 2081,
      "time_per_iteration": 2.5791807174682617
    },
    {
      "auxiliary_loss_clip": 0.01166006,
      "auxiliary_loss_mlp": 0.0103492,
      "balance_loss_clip": 1.05755448,
      "balance_loss_mlp": 1.02504325,
      "epoch": 0.2503456983105874,
      "flos": 28761681651840.0,
      "grad_norm": 2.027573687189531,
      "language_loss": 0.80711162,
      "learning_rate": 3.5121018552228644e-06,
      "loss": 0.82912087,
      "num_input_tokens_seen": 44476155,
      "step": 2082,
      "time_per_iteration": 2.6169235706329346
    },
    {
      "auxiliary_loss_clip": 0.01166833,
      "auxiliary_loss_mlp": 0.01030079,
      "balance_loss_clip": 1.05506468,
      "balance_loss_mlp": 1.02041078,
      "epoch": 0.2504659412012265,
      "flos": 18770256673920.0,
      "grad_norm": 2.029972118376554,
      "language_loss": 0.76470101,
      "learning_rate": 3.5115918936356827e-06,
      "loss": 0.78667009,
      "num_input_tokens_seen": 44492910,
      "step": 2083,
      "time_per_iteration": 2.552333354949951
    },
    {
      "auxiliary_loss_clip": 0.01147884,
      "auxiliary_loss_mlp": 0.01036163,
      "balance_loss_clip": 1.0536449,
      "balance_loss_mlp": 1.02694845,
      "epoch": 0.25058618409186556,
      "flos": 16873383346560.0,
      "grad_norm": 2.370308756378618,
      "language_loss": 0.78755164,
      "learning_rate": 3.5110817027412123e-06,
      "loss": 0.80939209,
      "num_input_tokens_seen": 44512000,
      "step": 2084,
      "time_per_iteration": 2.5458931922912598
    },
    {
      "auxiliary_loss_clip": 0.01157185,
      "auxiliary_loss_mlp": 0.01030422,
      "balance_loss_clip": 1.05020928,
      "balance_loss_mlp": 1.02140415,
      "epoch": 0.25070642698250467,
      "flos": 24425540651520.0,
      "grad_norm": 2.8180417453343094,
      "language_loss": 0.68780911,
      "learning_rate": 3.5105712826168493e-06,
      "loss": 0.70968521,
      "num_input_tokens_seen": 44531650,
      "step": 2085,
      "time_per_iteration": 2.596689224243164
    },
    {
      "auxiliary_loss_clip": 0.01192986,
      "auxiliary_loss_mlp": 0.00763298,
      "balance_loss_clip": 1.05675495,
      "balance_loss_mlp": 1.00116539,
      "epoch": 0.2508266698731437,
      "flos": 20260944028800.0,
      "grad_norm": 1.9214801655449527,
      "language_loss": 0.7067641,
      "learning_rate": 3.5100606333400235e-06,
      "loss": 0.72632694,
      "num_input_tokens_seen": 44548785,
      "step": 2086,
      "time_per_iteration": 2.499660015106201
    },
    {
      "auxiliary_loss_clip": 0.01191625,
      "auxiliary_loss_mlp": 0.0103215,
      "balance_loss_clip": 1.05773067,
      "balance_loss_mlp": 1.02131391,
      "epoch": 0.25094691276378284,
      "flos": 19245318975360.0,
      "grad_norm": 2.9429969861002228,
      "language_loss": 0.77205259,
      "learning_rate": 3.5095497549882006e-06,
      "loss": 0.7942903,
      "num_input_tokens_seen": 44567230,
      "step": 2087,
      "time_per_iteration": 2.5072739124298096
    },
    {
      "auxiliary_loss_clip": 0.01200347,
      "auxiliary_loss_mlp": 0.01029978,
      "balance_loss_clip": 1.06318438,
      "balance_loss_mlp": 1.02018487,
      "epoch": 0.25106715565442195,
      "flos": 26943237671040.0,
      "grad_norm": 2.52724869028484,
      "language_loss": 0.7207886,
      "learning_rate": 3.50903864763888e-06,
      "loss": 0.74309188,
      "num_input_tokens_seen": 44588020,
      "step": 2088,
      "time_per_iteration": 2.5900368690490723
    },
    {
      "auxiliary_loss_clip": 0.01200584,
      "auxiliary_loss_mlp": 0.01031707,
      "balance_loss_clip": 1.05949688,
      "balance_loss_mlp": 1.02236116,
      "epoch": 0.251187398545061,
      "flos": 48359570572800.0,
      "grad_norm": 1.9646511295915956,
      "language_loss": 0.75727779,
      "learning_rate": 3.5085273113695965e-06,
      "loss": 0.77960068,
      "num_input_tokens_seen": 44612590,
      "step": 2089,
      "time_per_iteration": 2.723231077194214
    },
    {
      "auxiliary_loss_clip": 0.01210739,
      "auxiliary_loss_mlp": 0.01034052,
      "balance_loss_clip": 1.06029487,
      "balance_loss_mlp": 1.02427125,
      "epoch": 0.2513076414357001,
      "flos": 27016100409600.0,
      "grad_norm": 1.789578014643017,
      "language_loss": 0.78525037,
      "learning_rate": 3.508015746257919e-06,
      "loss": 0.80769825,
      "num_input_tokens_seen": 44631630,
      "step": 2090,
      "time_per_iteration": 2.5166873931884766
    },
    {
      "auxiliary_loss_clip": 0.0116848,
      "auxiliary_loss_mlp": 0.01034886,
      "balance_loss_clip": 1.05636525,
      "balance_loss_mlp": 1.02509272,
      "epoch": 0.2514278843263392,
      "flos": 19463619882240.0,
      "grad_norm": 2.0204404147046375,
      "language_loss": 0.83061755,
      "learning_rate": 3.5075039523814518e-06,
      "loss": 0.85265118,
      "num_input_tokens_seen": 44650820,
      "step": 2091,
      "time_per_iteration": 2.564723491668701
    },
    {
      "auxiliary_loss_clip": 0.01200617,
      "auxiliary_loss_mlp": 0.01031737,
      "balance_loss_clip": 1.05775988,
      "balance_loss_mlp": 1.0213418,
      "epoch": 0.2515481272169783,
      "flos": 16866092885760.0,
      "grad_norm": 2.188227869645875,
      "language_loss": 0.81560993,
      "learning_rate": 3.506991929817834e-06,
      "loss": 0.83793342,
      "num_input_tokens_seen": 44667540,
      "step": 2092,
      "time_per_iteration": 2.5234711170196533
    },
    {
      "auxiliary_loss_clip": 0.01206147,
      "auxiliary_loss_mlp": 0.0102999,
      "balance_loss_clip": 1.06064868,
      "balance_loss_mlp": 1.02123988,
      "epoch": 0.2516683701076174,
      "flos": 23732464752000.0,
      "grad_norm": 1.777410296667334,
      "language_loss": 0.82636571,
      "learning_rate": 3.506479678644738e-06,
      "loss": 0.84872711,
      "num_input_tokens_seen": 44687935,
      "step": 2093,
      "time_per_iteration": 2.471876621246338
    },
    {
      "auxiliary_loss_clip": 0.0114097,
      "auxiliary_loss_mlp": 0.01029315,
      "balance_loss_clip": 1.05137801,
      "balance_loss_mlp": 1.02036881,
      "epoch": 0.2517886129982565,
      "flos": 27635954434560.0,
      "grad_norm": 2.463330777144152,
      "language_loss": 0.73889148,
      "learning_rate": 3.505967198939873e-06,
      "loss": 0.76059437,
      "num_input_tokens_seen": 44704975,
      "step": 2094,
      "time_per_iteration": 2.62829852104187
    },
    {
      "auxiliary_loss_clip": 0.01174546,
      "auxiliary_loss_mlp": 0.01028558,
      "balance_loss_clip": 1.05231702,
      "balance_loss_mlp": 1.01917052,
      "epoch": 0.25190885588889556,
      "flos": 38104596529920.0,
      "grad_norm": 1.9723883693400794,
      "language_loss": 0.77958047,
      "learning_rate": 3.5054544907809813e-06,
      "loss": 0.80161142,
      "num_input_tokens_seen": 44725475,
      "step": 2095,
      "time_per_iteration": 2.6456923484802246
    },
    {
      "auxiliary_loss_clip": 0.01177222,
      "auxiliary_loss_mlp": 0.00764623,
      "balance_loss_clip": 1.05789363,
      "balance_loss_mlp": 1.00136232,
      "epoch": 0.25202909877953467,
      "flos": 22269894768000.0,
      "grad_norm": 2.785930989601767,
      "language_loss": 0.80698156,
      "learning_rate": 3.50494155424584e-06,
      "loss": 0.82640004,
      "num_input_tokens_seen": 44744380,
      "step": 2096,
      "time_per_iteration": 2.55389404296875
    },
    {
      "auxiliary_loss_clip": 0.01198467,
      "auxiliary_loss_mlp": 0.01032096,
      "balance_loss_clip": 1.05908585,
      "balance_loss_mlp": 1.02257681,
      "epoch": 0.2521493416701738,
      "flos": 21761759018880.0,
      "grad_norm": 1.623382079299628,
      "language_loss": 0.83419585,
      "learning_rate": 3.504428389412262e-06,
      "loss": 0.85650146,
      "num_input_tokens_seen": 44765190,
      "step": 2097,
      "time_per_iteration": 2.52713942527771
    },
    {
      "auxiliary_loss_clip": 0.01190967,
      "auxiliary_loss_mlp": 0.01033106,
      "balance_loss_clip": 1.05650091,
      "balance_loss_mlp": 1.02393866,
      "epoch": 0.25226958456081283,
      "flos": 27746738956800.0,
      "grad_norm": 2.2312354328634574,
      "language_loss": 0.72665489,
      "learning_rate": 3.5039149963580927e-06,
      "loss": 0.74889559,
      "num_input_tokens_seen": 44785210,
      "step": 2098,
      "time_per_iteration": 3.3202974796295166
    },
    {
      "auxiliary_loss_clip": 0.01174854,
      "auxiliary_loss_mlp": 0.01033435,
      "balance_loss_clip": 1.05871892,
      "balance_loss_mlp": 1.02432728,
      "epoch": 0.25238982745145194,
      "flos": 30732171903360.0,
      "grad_norm": 2.281896727527845,
      "language_loss": 0.70450383,
      "learning_rate": 3.503401375161215e-06,
      "loss": 0.72658676,
      "num_input_tokens_seen": 44804955,
      "step": 2099,
      "time_per_iteration": 2.578850269317627
    },
    {
      "auxiliary_loss_clip": 0.0120508,
      "auxiliary_loss_mlp": 0.01030092,
      "balance_loss_clip": 1.05811071,
      "balance_loss_mlp": 1.0211755,
      "epoch": 0.252510070342091,
      "flos": 20266331068800.0,
      "grad_norm": 1.7796924308867659,
      "language_loss": 0.83597136,
      "learning_rate": 3.502887525899544e-06,
      "loss": 0.85832304,
      "num_input_tokens_seen": 44823935,
      "step": 2100,
      "time_per_iteration": 2.46354341506958
    },
    {
      "auxiliary_loss_clip": 0.01180265,
      "auxiliary_loss_mlp": 0.01029197,
      "balance_loss_clip": 1.05734277,
      "balance_loss_mlp": 1.01944029,
      "epoch": 0.2526303132327301,
      "flos": 22747399194240.0,
      "grad_norm": 1.681258631645701,
      "language_loss": 0.82519603,
      "learning_rate": 3.50237344865103e-06,
      "loss": 0.84729064,
      "num_input_tokens_seen": 44844935,
      "step": 2101,
      "time_per_iteration": 2.5621910095214844
    },
    {
      "auxiliary_loss_clip": 0.01210722,
      "auxiliary_loss_mlp": 0.0103736,
      "balance_loss_clip": 1.06085944,
      "balance_loss_mlp": 1.02814507,
      "epoch": 0.2527505561233692,
      "flos": 30263466309120.0,
      "grad_norm": 3.0287617960629722,
      "language_loss": 0.76394808,
      "learning_rate": 3.501859143493658e-06,
      "loss": 0.78642887,
      "num_input_tokens_seen": 44865565,
      "step": 2102,
      "time_per_iteration": 2.5238852500915527
    },
    {
      "auxiliary_loss_clip": 0.01127548,
      "auxiliary_loss_mlp": 0.01004128,
      "balance_loss_clip": 1.04237247,
      "balance_loss_mlp": 1.00212514,
      "epoch": 0.2528707990140083,
      "flos": 58492917164160.0,
      "grad_norm": 0.9198774542697352,
      "language_loss": 0.60590291,
      "learning_rate": 3.5013446105054488e-06,
      "loss": 0.62721968,
      "num_input_tokens_seen": 44918485,
      "step": 2103,
      "time_per_iteration": 2.8064167499542236
    },
    {
      "auxiliary_loss_clip": 0.01149734,
      "auxiliary_loss_mlp": 0.01036195,
      "balance_loss_clip": 1.053478,
      "balance_loss_mlp": 1.02671242,
      "epoch": 0.2529910419046474,
      "flos": 24645134448000.0,
      "grad_norm": 2.689947959380527,
      "language_loss": 0.74874812,
      "learning_rate": 3.5008298497644555e-06,
      "loss": 0.77060741,
      "num_input_tokens_seen": 44937530,
      "step": 2104,
      "time_per_iteration": 3.510887384414673
    },
    {
      "auxiliary_loss_clip": 0.01168663,
      "auxiliary_loss_mlp": 0.01034801,
      "balance_loss_clip": 1.05809307,
      "balance_loss_mlp": 1.02478135,
      "epoch": 0.2531112847952865,
      "flos": 23842135952640.0,
      "grad_norm": 1.620817927311642,
      "language_loss": 0.88023871,
      "learning_rate": 3.500314861348767e-06,
      "loss": 0.90227336,
      "num_input_tokens_seen": 44958165,
      "step": 2105,
      "time_per_iteration": 3.3672237396240234
    },
    {
      "auxiliary_loss_clip": 0.01157978,
      "auxiliary_loss_mlp": 0.01036351,
      "balance_loss_clip": 1.05668521,
      "balance_loss_mlp": 1.02749956,
      "epoch": 0.25323152768592555,
      "flos": 16143822207360.0,
      "grad_norm": 1.7870348900076634,
      "language_loss": 0.76794493,
      "learning_rate": 3.499799645336507e-06,
      "loss": 0.78988814,
      "num_input_tokens_seen": 44975060,
      "step": 2106,
      "time_per_iteration": 2.5176327228546143
    },
    {
      "auxiliary_loss_clip": 0.01198163,
      "auxiliary_loss_mlp": 0.01029664,
      "balance_loss_clip": 1.06277585,
      "balance_loss_mlp": 1.02108669,
      "epoch": 0.25335177057656466,
      "flos": 28405161210240.0,
      "grad_norm": 1.560940088177471,
      "language_loss": 0.87272751,
      "learning_rate": 3.4992842018058336e-06,
      "loss": 0.89500576,
      "num_input_tokens_seen": 44997960,
      "step": 2107,
      "time_per_iteration": 2.5629451274871826
    },
    {
      "auxiliary_loss_clip": 0.01170272,
      "auxiliary_loss_mlp": 0.01029399,
      "balance_loss_clip": 1.05606508,
      "balance_loss_mlp": 1.02040446,
      "epoch": 0.25347201346720377,
      "flos": 18799666934400.0,
      "grad_norm": 2.247012906264391,
      "language_loss": 0.88612056,
      "learning_rate": 3.4987685308349384e-06,
      "loss": 0.90811729,
      "num_input_tokens_seen": 45015690,
      "step": 2108,
      "time_per_iteration": 2.551663875579834
    },
    {
      "auxiliary_loss_clip": 0.01162662,
      "auxiliary_loss_mlp": 0.01036231,
      "balance_loss_clip": 1.05203533,
      "balance_loss_mlp": 1.02670062,
      "epoch": 0.2535922563578428,
      "flos": 15815490963840.0,
      "grad_norm": 2.1902859430760744,
      "language_loss": 0.61364943,
      "learning_rate": 3.4982526325020497e-06,
      "loss": 0.63563836,
      "num_input_tokens_seen": 45032660,
      "step": 2109,
      "time_per_iteration": 2.5311169624328613
    },
    {
      "auxiliary_loss_clip": 0.01184227,
      "auxiliary_loss_mlp": 0.01031697,
      "balance_loss_clip": 1.06005657,
      "balance_loss_mlp": 1.02185011,
      "epoch": 0.25371249924848194,
      "flos": 16318922031360.0,
      "grad_norm": 2.5398507535148274,
      "language_loss": 0.8234089,
      "learning_rate": 3.4977365068854273e-06,
      "loss": 0.84556818,
      "num_input_tokens_seen": 45048280,
      "step": 2110,
      "time_per_iteration": 2.4695026874542236
    },
    {
      "auxiliary_loss_clip": 0.01174229,
      "auxiliary_loss_mlp": 0.01033969,
      "balance_loss_clip": 1.05703449,
      "balance_loss_mlp": 1.02419376,
      "epoch": 0.25383274213912105,
      "flos": 21761615364480.0,
      "grad_norm": 1.7644066351984797,
      "language_loss": 0.73649764,
      "learning_rate": 3.4972201540633676e-06,
      "loss": 0.75857961,
      "num_input_tokens_seen": 45067635,
      "step": 2111,
      "time_per_iteration": 2.526698350906372
    },
    {
      "auxiliary_loss_clip": 0.01169871,
      "auxiliary_loss_mlp": 0.01033634,
      "balance_loss_clip": 1.05546629,
      "balance_loss_mlp": 1.02351904,
      "epoch": 0.2539529850297601,
      "flos": 21396870708480.0,
      "grad_norm": 1.7543738315816078,
      "language_loss": 0.85379744,
      "learning_rate": 3.4967035741142008e-06,
      "loss": 0.87583244,
      "num_input_tokens_seen": 45086455,
      "step": 2112,
      "time_per_iteration": 2.527505874633789
    },
    {
      "auxiliary_loss_clip": 0.01171342,
      "auxiliary_loss_mlp": 0.01033705,
      "balance_loss_clip": 1.06256962,
      "balance_loss_mlp": 1.02473426,
      "epoch": 0.2540732279203992,
      "flos": 25228467319680.0,
      "grad_norm": 1.823329256966325,
      "language_loss": 0.81955135,
      "learning_rate": 3.4961867671162917e-06,
      "loss": 0.84160179,
      "num_input_tokens_seen": 45106385,
      "step": 2113,
      "time_per_iteration": 2.5422539710998535
    },
    {
      "auxiliary_loss_clip": 0.0121323,
      "auxiliary_loss_mlp": 0.01032302,
      "balance_loss_clip": 1.06257617,
      "balance_loss_mlp": 1.02240777,
      "epoch": 0.2541934708110383,
      "flos": 19427386037760.0,
      "grad_norm": 2.452121420849565,
      "language_loss": 0.77036297,
      "learning_rate": 3.4956697331480402e-06,
      "loss": 0.79281831,
      "num_input_tokens_seen": 45124955,
      "step": 2114,
      "time_per_iteration": 2.428863525390625
    },
    {
      "auxiliary_loss_clip": 0.01171696,
      "auxiliary_loss_mlp": 0.01032802,
      "balance_loss_clip": 1.05643868,
      "balance_loss_mlp": 1.02305675,
      "epoch": 0.2543137137016774,
      "flos": 23949436855680.0,
      "grad_norm": 1.7019061444726813,
      "language_loss": 0.79976726,
      "learning_rate": 3.495152472287879e-06,
      "loss": 0.82181227,
      "num_input_tokens_seen": 45145665,
      "step": 2115,
      "time_per_iteration": 2.6080031394958496
    },
    {
      "auxiliary_loss_clip": 0.01165409,
      "auxiliary_loss_mlp": 0.01030051,
      "balance_loss_clip": 1.05810928,
      "balance_loss_mlp": 1.02141404,
      "epoch": 0.2544339565923165,
      "flos": 25593283802880.0,
      "grad_norm": 1.8134489290409792,
      "language_loss": 0.73761773,
      "learning_rate": 3.4946349846142766e-06,
      "loss": 0.75957233,
      "num_input_tokens_seen": 45164805,
      "step": 2116,
      "time_per_iteration": 2.5908701419830322
    },
    {
      "auxiliary_loss_clip": 0.01210647,
      "auxiliary_loss_mlp": 0.01039099,
      "balance_loss_clip": 1.0630033,
      "balance_loss_mlp": 1.02992618,
      "epoch": 0.25455419948295555,
      "flos": 21689470897920.0,
      "grad_norm": 2.714984083260144,
      "language_loss": 0.7545352,
      "learning_rate": 3.4941172702057353e-06,
      "loss": 0.77703261,
      "num_input_tokens_seen": 45184865,
      "step": 2117,
      "time_per_iteration": 2.5367674827575684
    },
    {
      "auxiliary_loss_clip": 0.01179643,
      "auxiliary_loss_mlp": 0.01033511,
      "balance_loss_clip": 1.06030905,
      "balance_loss_mlp": 1.02409899,
      "epoch": 0.25467444237359466,
      "flos": 26250341339520.0,
      "grad_norm": 1.8224707180842552,
      "language_loss": 0.80697495,
      "learning_rate": 3.4935993291407924e-06,
      "loss": 0.82910645,
      "num_input_tokens_seen": 45203690,
      "step": 2118,
      "time_per_iteration": 2.5747146606445312
    },
    {
      "auxiliary_loss_clip": 0.01177731,
      "auxiliary_loss_mlp": 0.01033375,
      "balance_loss_clip": 1.05787659,
      "balance_loss_mlp": 1.02348089,
      "epoch": 0.25479468526423377,
      "flos": 26979686997120.0,
      "grad_norm": 2.3822660401443474,
      "language_loss": 0.70912182,
      "learning_rate": 3.4930811614980183e-06,
      "loss": 0.73123294,
      "num_input_tokens_seen": 45225385,
      "step": 2119,
      "time_per_iteration": 2.5754342079162598
    },
    {
      "auxiliary_loss_clip": 0.01188256,
      "auxiliary_loss_mlp": 0.01035929,
      "balance_loss_clip": 1.05858421,
      "balance_loss_mlp": 1.02682686,
      "epoch": 0.2549149281548728,
      "flos": 23475811098240.0,
      "grad_norm": 1.8970964193664563,
      "language_loss": 0.79338145,
      "learning_rate": 3.4925627673560198e-06,
      "loss": 0.81562328,
      "num_input_tokens_seen": 45246045,
      "step": 2120,
      "time_per_iteration": 2.522830009460449
    },
    {
      "auxiliary_loss_clip": 0.01165668,
      "auxiliary_loss_mlp": 0.01038438,
      "balance_loss_clip": 1.05677044,
      "balance_loss_mlp": 1.02977741,
      "epoch": 0.25503517104551193,
      "flos": 25812302981760.0,
      "grad_norm": 2.661072957673536,
      "language_loss": 0.88428557,
      "learning_rate": 3.4920441467934357e-06,
      "loss": 0.90632659,
      "num_input_tokens_seen": 45266560,
      "step": 2121,
      "time_per_iteration": 2.587238311767578
    },
    {
      "auxiliary_loss_clip": 0.01157619,
      "auxiliary_loss_mlp": 0.01036816,
      "balance_loss_clip": 1.05562901,
      "balance_loss_mlp": 1.02777994,
      "epoch": 0.25515541393615104,
      "flos": 26645106787200.0,
      "grad_norm": 1.966157032080573,
      "language_loss": 0.82839191,
      "learning_rate": 3.491525299888941e-06,
      "loss": 0.85033631,
      "num_input_tokens_seen": 45285405,
      "step": 2122,
      "time_per_iteration": 2.6034395694732666
    },
    {
      "auxiliary_loss_clip": 0.01094853,
      "auxiliary_loss_mlp": 0.00754878,
      "balance_loss_clip": 1.04400086,
      "balance_loss_mlp": 1.00095141,
      "epoch": 0.2552756568267901,
      "flos": 65955945847680.0,
      "grad_norm": 0.8849975689910463,
      "language_loss": 0.62691009,
      "learning_rate": 3.491006226721244e-06,
      "loss": 0.64540744,
      "num_input_tokens_seen": 45349615,
      "step": 2123,
      "time_per_iteration": 3.1456305980682373
    },
    {
      "auxiliary_loss_clip": 0.01186754,
      "auxiliary_loss_mlp": 0.007643,
      "balance_loss_clip": 1.06344199,
      "balance_loss_mlp": 1.0012691,
      "epoch": 0.2553958997174292,
      "flos": 17931096161280.0,
      "grad_norm": 1.8908626252248844,
      "language_loss": 0.77408051,
      "learning_rate": 3.4904869273690882e-06,
      "loss": 0.79359102,
      "num_input_tokens_seen": 45367505,
      "step": 2124,
      "time_per_iteration": 2.534048318862915
    },
    {
      "auxiliary_loss_clip": 0.01198786,
      "auxiliary_loss_mlp": 0.01026167,
      "balance_loss_clip": 1.06181324,
      "balance_loss_mlp": 1.01729155,
      "epoch": 0.2555161426080683,
      "flos": 23367791923200.0,
      "grad_norm": 2.7644258258860352,
      "language_loss": 0.88696647,
      "learning_rate": 3.489967401911251e-06,
      "loss": 0.90921593,
      "num_input_tokens_seen": 45386805,
      "step": 2125,
      "time_per_iteration": 3.295128583908081
    },
    {
      "auxiliary_loss_clip": 0.0121717,
      "auxiliary_loss_mlp": 0.01033972,
      "balance_loss_clip": 1.06608415,
      "balance_loss_mlp": 1.02309442,
      "epoch": 0.2556363854987074,
      "flos": 40625130723840.0,
      "grad_norm": 1.8508022066010383,
      "language_loss": 0.6953373,
      "learning_rate": 3.4894476504265428e-06,
      "loss": 0.71784872,
      "num_input_tokens_seen": 45411045,
      "step": 2126,
      "time_per_iteration": 2.6433393955230713
    },
    {
      "auxiliary_loss_clip": 0.01115047,
      "auxiliary_loss_mlp": 0.01004231,
      "balance_loss_clip": 1.04365349,
      "balance_loss_mlp": 1.00169134,
      "epoch": 0.2557566283893465,
      "flos": 68019443389440.0,
      "grad_norm": 0.7405179474258614,
      "language_loss": 0.5443002,
      "learning_rate": 3.4889276729938104e-06,
      "loss": 0.56549299,
      "num_input_tokens_seen": 45469575,
      "step": 2127,
      "time_per_iteration": 2.9664998054504395
    },
    {
      "auxiliary_loss_clip": 0.01174716,
      "auxiliary_loss_mlp": 0.01027755,
      "balance_loss_clip": 1.05700588,
      "balance_loss_mlp": 1.01780152,
      "epoch": 0.2558768712799856,
      "flos": 22635645004800.0,
      "grad_norm": 1.9136250464817475,
      "language_loss": 0.80596977,
      "learning_rate": 3.488407469691934e-06,
      "loss": 0.82799447,
      "num_input_tokens_seen": 45490270,
      "step": 2128,
      "time_per_iteration": 2.522911787033081
    },
    {
      "auxiliary_loss_clip": 0.01179273,
      "auxiliary_loss_mlp": 0.01032516,
      "balance_loss_clip": 1.05745125,
      "balance_loss_mlp": 1.02284241,
      "epoch": 0.25599711417062465,
      "flos": 26396354125440.0,
      "grad_norm": 2.349254643827396,
      "language_loss": 0.80623996,
      "learning_rate": 3.487887040599828e-06,
      "loss": 0.82835782,
      "num_input_tokens_seen": 45510070,
      "step": 2129,
      "time_per_iteration": 2.5704989433288574
    },
    {
      "auxiliary_loss_clip": 0.01213838,
      "auxiliary_loss_mlp": 0.0103564,
      "balance_loss_clip": 1.06481028,
      "balance_loss_mlp": 1.02570391,
      "epoch": 0.25611735706126376,
      "flos": 22852042490880.0,
      "grad_norm": 2.2105955077988324,
      "language_loss": 0.7597664,
      "learning_rate": 3.4873663857964407e-06,
      "loss": 0.78226113,
      "num_input_tokens_seen": 45527285,
      "step": 2130,
      "time_per_iteration": 3.4133925437927246
    },
    {
      "auxiliary_loss_clip": 0.011489,
      "auxiliary_loss_mlp": 0.01035092,
      "balance_loss_clip": 1.05491734,
      "balance_loss_mlp": 1.02495933,
      "epoch": 0.2562375999519028,
      "flos": 23367863750400.0,
      "grad_norm": 1.7716620352230272,
      "language_loss": 0.66543591,
      "learning_rate": 3.4868455053607556e-06,
      "loss": 0.68727589,
      "num_input_tokens_seen": 45546900,
      "step": 2131,
      "time_per_iteration": 3.3764991760253906
    },
    {
      "auxiliary_loss_clip": 0.01199258,
      "auxiliary_loss_mlp": 0.0103568,
      "balance_loss_clip": 1.05922198,
      "balance_loss_mlp": 1.02603614,
      "epoch": 0.2563578428425419,
      "flos": 22856962654080.0,
      "grad_norm": 3.2073863762838473,
      "language_loss": 0.72177035,
      "learning_rate": 3.486324399371789e-06,
      "loss": 0.74411976,
      "num_input_tokens_seen": 45566200,
      "step": 2132,
      "time_per_iteration": 2.498868942260742
    },
    {
      "auxiliary_loss_clip": 0.01161479,
      "auxiliary_loss_mlp": 0.01038006,
      "balance_loss_clip": 1.05672216,
      "balance_loss_mlp": 1.02896965,
      "epoch": 0.25647808573318104,
      "flos": 21653883498240.0,
      "grad_norm": 1.9541045187352488,
      "language_loss": 0.78494954,
      "learning_rate": 3.485803067908593e-06,
      "loss": 0.80694437,
      "num_input_tokens_seen": 45585710,
      "step": 2133,
      "time_per_iteration": 2.546391487121582
    },
    {
      "auxiliary_loss_clip": 0.01110771,
      "auxiliary_loss_mlp": 0.0103468,
      "balance_loss_clip": 1.04474068,
      "balance_loss_mlp": 1.02492917,
      "epoch": 0.2565983286238201,
      "flos": 33730569659520.0,
      "grad_norm": 1.8062645234911745,
      "language_loss": 0.79570651,
      "learning_rate": 3.485281511050253e-06,
      "loss": 0.81716096,
      "num_input_tokens_seen": 45607845,
      "step": 2134,
      "time_per_iteration": 2.7235312461853027
    },
    {
      "auxiliary_loss_clip": 0.01199729,
      "auxiliary_loss_mlp": 0.01034719,
      "balance_loss_clip": 1.06069183,
      "balance_loss_mlp": 1.02504539,
      "epoch": 0.2567185715144592,
      "flos": 16216002587520.0,
      "grad_norm": 3.6119305721229447,
      "language_loss": 0.89920598,
      "learning_rate": 3.484759728875889e-06,
      "loss": 0.92155039,
      "num_input_tokens_seen": 45623210,
      "step": 2135,
      "time_per_iteration": 2.4774370193481445
    },
    {
      "auxiliary_loss_clip": 0.01136779,
      "auxiliary_loss_mlp": 0.0103751,
      "balance_loss_clip": 1.05229926,
      "balance_loss_mlp": 1.0283016,
      "epoch": 0.2568388144050983,
      "flos": 17458475984640.0,
      "grad_norm": 1.7588320643640283,
      "language_loss": 0.80961126,
      "learning_rate": 3.4842377214646543e-06,
      "loss": 0.83135414,
      "num_input_tokens_seen": 45641505,
      "step": 2136,
      "time_per_iteration": 2.571418046951294
    },
    {
      "auxiliary_loss_clip": 0.01209165,
      "auxiliary_loss_mlp": 0.01035252,
      "balance_loss_clip": 1.0621078,
      "balance_loss_mlp": 1.02619827,
      "epoch": 0.25695905729573737,
      "flos": 20887442069760.0,
      "grad_norm": 1.7296693757860617,
      "language_loss": 0.66624212,
      "learning_rate": 3.483715488895737e-06,
      "loss": 0.68868637,
      "num_input_tokens_seen": 45661835,
      "step": 2137,
      "time_per_iteration": 2.4665029048919678
    },
    {
      "auxiliary_loss_clip": 0.01147459,
      "auxiliary_loss_mlp": 0.01029968,
      "balance_loss_clip": 1.05019176,
      "balance_loss_mlp": 1.02044892,
      "epoch": 0.2570793001863765,
      "flos": 24717278914560.0,
      "grad_norm": 2.1795694889333523,
      "language_loss": 0.78334129,
      "learning_rate": 3.48319303124836e-06,
      "loss": 0.80511558,
      "num_input_tokens_seen": 45682215,
      "step": 2138,
      "time_per_iteration": 2.625080108642578
    },
    {
      "auxiliary_loss_clip": 0.0117557,
      "auxiliary_loss_mlp": 0.01031237,
      "balance_loss_clip": 1.05925131,
      "balance_loss_mlp": 1.0218612,
      "epoch": 0.2571995430770156,
      "flos": 26906896085760.0,
      "grad_norm": 2.085022932892146,
      "language_loss": 0.66648912,
      "learning_rate": 3.4826703486017798e-06,
      "loss": 0.68855721,
      "num_input_tokens_seen": 45701840,
      "step": 2139,
      "time_per_iteration": 2.5612897872924805
    },
    {
      "auxiliary_loss_clip": 0.01192519,
      "auxiliary_loss_mlp": 0.01030485,
      "balance_loss_clip": 1.06067789,
      "balance_loss_mlp": 1.02156854,
      "epoch": 0.25731978596765465,
      "flos": 19792561656960.0,
      "grad_norm": 1.6318146761712238,
      "language_loss": 0.76885098,
      "learning_rate": 3.4821474410352867e-06,
      "loss": 0.79108107,
      "num_input_tokens_seen": 45720500,
      "step": 2140,
      "time_per_iteration": 2.556918144226074
    },
    {
      "auxiliary_loss_clip": 0.01084958,
      "auxiliary_loss_mlp": 0.01005569,
      "balance_loss_clip": 1.03921795,
      "balance_loss_mlp": 1.00239778,
      "epoch": 0.25744002885829376,
      "flos": 70564970471040.0,
      "grad_norm": 0.8999623557472874,
      "language_loss": 0.62621933,
      "learning_rate": 3.481624308628205e-06,
      "loss": 0.64712459,
      "num_input_tokens_seen": 45781870,
      "step": 2141,
      "time_per_iteration": 3.2933452129364014
    },
    {
      "auxiliary_loss_clip": 0.01177322,
      "auxiliary_loss_mlp": 0.01033661,
      "balance_loss_clip": 1.05670762,
      "balance_loss_mlp": 1.02414823,
      "epoch": 0.25756027174893287,
      "flos": 18038181582720.0,
      "grad_norm": 3.212182069857203,
      "language_loss": 1.00315654,
      "learning_rate": 3.481100951459893e-06,
      "loss": 1.02526641,
      "num_input_tokens_seen": 45794890,
      "step": 2142,
      "time_per_iteration": 2.534937620162964
    },
    {
      "auxiliary_loss_clip": 0.01191474,
      "auxiliary_loss_mlp": 0.01031178,
      "balance_loss_clip": 1.058568,
      "balance_loss_mlp": 1.0217663,
      "epoch": 0.2576805146395719,
      "flos": 22674069578880.0,
      "grad_norm": 1.5899786862721246,
      "language_loss": 0.78834361,
      "learning_rate": 3.4805773696097453e-06,
      "loss": 0.81057012,
      "num_input_tokens_seen": 45815780,
      "step": 2143,
      "time_per_iteration": 2.525907039642334
    },
    {
      "auxiliary_loss_clip": 0.01174866,
      "auxiliary_loss_mlp": 0.01029233,
      "balance_loss_clip": 1.05991483,
      "balance_loss_mlp": 1.01961243,
      "epoch": 0.25780075753021103,
      "flos": 16472225278080.0,
      "grad_norm": 1.9527030131637773,
      "language_loss": 0.87907928,
      "learning_rate": 3.4800535631571874e-06,
      "loss": 0.90112025,
      "num_input_tokens_seen": 45831310,
      "step": 2144,
      "time_per_iteration": 2.5082976818084717
    },
    {
      "auxiliary_loss_clip": 0.01183989,
      "auxiliary_loss_mlp": 0.01037565,
      "balance_loss_clip": 1.05801511,
      "balance_loss_mlp": 1.0278194,
      "epoch": 0.25792100042085014,
      "flos": 22820297846400.0,
      "grad_norm": 2.493707492825981,
      "language_loss": 0.76478982,
      "learning_rate": 3.4795295321816804e-06,
      "loss": 0.78700531,
      "num_input_tokens_seen": 45850135,
      "step": 2145,
      "time_per_iteration": 2.529543876647949
    },
    {
      "auxiliary_loss_clip": 0.01168256,
      "auxiliary_loss_mlp": 0.01037876,
      "balance_loss_clip": 1.05690563,
      "balance_loss_mlp": 1.02841663,
      "epoch": 0.2580412433114892,
      "flos": 18697286194560.0,
      "grad_norm": 3.393185209914697,
      "language_loss": 0.91164839,
      "learning_rate": 3.47900527676272e-06,
      "loss": 0.93370974,
      "num_input_tokens_seen": 45868470,
      "step": 2146,
      "time_per_iteration": 2.5117640495300293
    },
    {
      "auxiliary_loss_clip": 0.01211428,
      "auxiliary_loss_mlp": 0.01035204,
      "balance_loss_clip": 1.06449318,
      "balance_loss_mlp": 1.02572691,
      "epoch": 0.2581614862021283,
      "flos": 14283146810880.0,
      "grad_norm": 1.9667345444962852,
      "language_loss": 0.8832534,
      "learning_rate": 3.478480796979835e-06,
      "loss": 0.90571964,
      "num_input_tokens_seen": 45886355,
      "step": 2147,
      "time_per_iteration": 2.446657657623291
    },
    {
      "auxiliary_loss_clip": 0.01174209,
      "auxiliary_loss_mlp": 0.01028127,
      "balance_loss_clip": 1.05717802,
      "balance_loss_mlp": 1.01895952,
      "epoch": 0.25828172909276736,
      "flos": 29498281856640.0,
      "grad_norm": 1.4975642819999468,
      "language_loss": 0.7757293,
      "learning_rate": 3.4779560929125894e-06,
      "loss": 0.79775262,
      "num_input_tokens_seen": 45907900,
      "step": 2148,
      "time_per_iteration": 2.6038734912872314
    },
    {
      "auxiliary_loss_clip": 0.01089575,
      "auxiliary_loss_mlp": 0.01005444,
      "balance_loss_clip": 1.04155564,
      "balance_loss_mlp": 1.00286889,
      "epoch": 0.2584019719834065,
      "flos": 67114387376640.0,
      "grad_norm": 0.6799228745901635,
      "language_loss": 0.56979376,
      "learning_rate": 3.4774311646405783e-06,
      "loss": 0.5907439,
      "num_input_tokens_seen": 45977805,
      "step": 2149,
      "time_per_iteration": 3.242149591445923
    },
    {
      "auxiliary_loss_clip": 0.0115538,
      "auxiliary_loss_mlp": 0.01031984,
      "balance_loss_clip": 1.05454135,
      "balance_loss_mlp": 1.02326965,
      "epoch": 0.2585222148740456,
      "flos": 22893555634560.0,
      "grad_norm": 2.2430496661563573,
      "language_loss": 0.83480787,
      "learning_rate": 3.476906012243435e-06,
      "loss": 0.85668153,
      "num_input_tokens_seen": 45996715,
      "step": 2150,
      "time_per_iteration": 2.583371877670288
    },
    {
      "auxiliary_loss_clip": 0.01184199,
      "auxiliary_loss_mlp": 0.0102884,
      "balance_loss_clip": 1.06057811,
      "balance_loss_mlp": 1.01930368,
      "epoch": 0.25864245776468464,
      "flos": 28909202808960.0,
      "grad_norm": 1.6246665045476154,
      "language_loss": 0.81386536,
      "learning_rate": 3.476380635800824e-06,
      "loss": 0.83599567,
      "num_input_tokens_seen": 46017915,
      "step": 2151,
      "time_per_iteration": 2.6141316890716553
    },
    {
      "auxiliary_loss_clip": 0.011777,
      "auxiliary_loss_mlp": 0.01031166,
      "balance_loss_clip": 1.05898774,
      "balance_loss_mlp": 1.02232063,
      "epoch": 0.25876270065532375,
      "flos": 14793185980800.0,
      "grad_norm": 2.0830984784151703,
      "language_loss": 0.86036265,
      "learning_rate": 3.475855035392444e-06,
      "loss": 0.88245124,
      "num_input_tokens_seen": 46033235,
      "step": 2152,
      "time_per_iteration": 3.8390533924102783
    },
    {
      "auxiliary_loss_clip": 0.01131826,
      "auxiliary_loss_mlp": 0.01029817,
      "balance_loss_clip": 1.05437434,
      "balance_loss_mlp": 1.02089381,
      "epoch": 0.25888294354596286,
      "flos": 60467821810560.0,
      "grad_norm": 2.08306049660786,
      "language_loss": 0.71569443,
      "learning_rate": 3.475329211098029e-06,
      "loss": 0.73731083,
      "num_input_tokens_seen": 46056390,
      "step": 2153,
      "time_per_iteration": 2.9688358306884766
    },
    {
      "auxiliary_loss_clip": 0.01152508,
      "auxiliary_loss_mlp": 0.01028557,
      "balance_loss_clip": 1.05670452,
      "balance_loss_mlp": 1.01941359,
      "epoch": 0.2590031864366019,
      "flos": 27851166771840.0,
      "grad_norm": 1.5582507213354646,
      "language_loss": 0.82287574,
      "learning_rate": 3.4748031629973453e-06,
      "loss": 0.84468639,
      "num_input_tokens_seen": 46077120,
      "step": 2154,
      "time_per_iteration": 2.6567227840423584
    },
    {
      "auxiliary_loss_clip": 0.01073611,
      "auxiliary_loss_mlp": 0.01003855,
      "balance_loss_clip": 1.04009473,
      "balance_loss_mlp": 1.00209069,
      "epoch": 0.25912342932724103,
      "flos": 62422444206720.0,
      "grad_norm": 0.9153383801395144,
      "language_loss": 0.56557369,
      "learning_rate": 3.4742768911701944e-06,
      "loss": 0.58634841,
      "num_input_tokens_seen": 46139815,
      "step": 2155,
      "time_per_iteration": 3.23642897605896
    },
    {
      "auxiliary_loss_clip": 0.0120126,
      "auxiliary_loss_mlp": 0.01040047,
      "balance_loss_clip": 1.06386042,
      "balance_loss_mlp": 1.0291332,
      "epoch": 0.25924367221788014,
      "flos": 12378839368320.0,
      "grad_norm": 3.071590148759398,
      "language_loss": 0.69918042,
      "learning_rate": 3.4737503956964113e-06,
      "loss": 0.72159344,
      "num_input_tokens_seen": 46152120,
      "step": 2156,
      "time_per_iteration": 4.235851764678955
    },
    {
      "auxiliary_loss_clip": 0.01172114,
      "auxiliary_loss_mlp": 0.01038798,
      "balance_loss_clip": 1.05689108,
      "balance_loss_mlp": 1.02848077,
      "epoch": 0.2593639151085192,
      "flos": 14575208296320.0,
      "grad_norm": 3.177376680075476,
      "language_loss": 0.67230749,
      "learning_rate": 3.473223676655865e-06,
      "loss": 0.69441658,
      "num_input_tokens_seen": 46170120,
      "step": 2157,
      "time_per_iteration": 3.321575164794922
    },
    {
      "auxiliary_loss_clip": 0.01171035,
      "auxiliary_loss_mlp": 0.01034157,
      "balance_loss_clip": 1.05484676,
      "balance_loss_mlp": 1.02372575,
      "epoch": 0.2594841579991583,
      "flos": 15230937029760.0,
      "grad_norm": 1.986048264940477,
      "language_loss": 0.79678667,
      "learning_rate": 3.472696734128459e-06,
      "loss": 0.8188386,
      "num_input_tokens_seen": 46187985,
      "step": 2158,
      "time_per_iteration": 2.5242321491241455
    },
    {
      "auxiliary_loss_clip": 0.0119552,
      "auxiliary_loss_mlp": 0.01031364,
      "balance_loss_clip": 1.06124496,
      "balance_loss_mlp": 1.02183914,
      "epoch": 0.2596044008897974,
      "flos": 23623583650560.0,
      "grad_norm": 2.572027662561668,
      "language_loss": 0.75841039,
      "learning_rate": 3.4721695681941286e-06,
      "loss": 0.78067923,
      "num_input_tokens_seen": 46207025,
      "step": 2159,
      "time_per_iteration": 2.5904955863952637
    },
    {
      "auxiliary_loss_clip": 0.01175774,
      "auxiliary_loss_mlp": 0.00764758,
      "balance_loss_clip": 1.05667567,
      "balance_loss_mlp": 1.00088191,
      "epoch": 0.25972464378043647,
      "flos": 13772281628160.0,
      "grad_norm": 3.556448341155961,
      "language_loss": 0.82570046,
      "learning_rate": 3.471642178932845e-06,
      "loss": 0.84510577,
      "num_input_tokens_seen": 46225670,
      "step": 2160,
      "time_per_iteration": 2.5687167644500732
    },
    {
      "auxiliary_loss_clip": 0.01180046,
      "auxiliary_loss_mlp": 0.01030961,
      "balance_loss_clip": 1.05669641,
      "balance_loss_mlp": 1.02163255,
      "epoch": 0.2598448866710756,
      "flos": 19573578391680.0,
      "grad_norm": 2.3503553334861516,
      "language_loss": 0.89278805,
      "learning_rate": 3.471114566424613e-06,
      "loss": 0.91489816,
      "num_input_tokens_seen": 46244130,
      "step": 2161,
      "time_per_iteration": 2.524306297302246
    },
    {
      "auxiliary_loss_clip": 0.01178501,
      "auxiliary_loss_mlp": 0.01030836,
      "balance_loss_clip": 1.06000483,
      "balance_loss_mlp": 1.02081084,
      "epoch": 0.25996512956171464,
      "flos": 21653237053440.0,
      "grad_norm": 1.868592188369467,
      "language_loss": 0.76046926,
      "learning_rate": 3.4705867307494715e-06,
      "loss": 0.78256261,
      "num_input_tokens_seen": 46263200,
      "step": 2162,
      "time_per_iteration": 2.550487756729126
    },
    {
      "auxiliary_loss_clip": 0.01197357,
      "auxiliary_loss_mlp": 0.01030685,
      "balance_loss_clip": 1.06076562,
      "balance_loss_mlp": 1.02163744,
      "epoch": 0.26008537245235375,
      "flos": 18223480869120.0,
      "grad_norm": 2.1920378979209243,
      "language_loss": 0.84468186,
      "learning_rate": 3.470058671987492e-06,
      "loss": 0.86696231,
      "num_input_tokens_seen": 46281465,
      "step": 2163,
      "time_per_iteration": 2.4827804565429688
    },
    {
      "auxiliary_loss_clip": 0.01198144,
      "auxiliary_loss_mlp": 0.01036858,
      "balance_loss_clip": 1.05964136,
      "balance_loss_mlp": 1.02676725,
      "epoch": 0.26020561534299286,
      "flos": 24645385843200.0,
      "grad_norm": 1.9838805671554751,
      "language_loss": 0.84219944,
      "learning_rate": 3.4695303902187805e-06,
      "loss": 0.86454946,
      "num_input_tokens_seen": 46301020,
      "step": 2164,
      "time_per_iteration": 2.576317310333252
    },
    {
      "auxiliary_loss_clip": 0.01159739,
      "auxiliary_loss_mlp": 0.01037567,
      "balance_loss_clip": 1.05413914,
      "balance_loss_mlp": 1.02752924,
      "epoch": 0.2603258582336319,
      "flos": 25773662926080.0,
      "grad_norm": 2.0365316758181566,
      "language_loss": 0.78617871,
      "learning_rate": 3.469001885523478e-06,
      "loss": 0.80815184,
      "num_input_tokens_seen": 46321740,
      "step": 2165,
      "time_per_iteration": 2.676100254058838
    },
    {
      "auxiliary_loss_clip": 0.01206667,
      "auxiliary_loss_mlp": 0.01041361,
      "balance_loss_clip": 1.06028962,
      "balance_loss_mlp": 1.03142488,
      "epoch": 0.260446101124271,
      "flos": 28766314506240.0,
      "grad_norm": 1.6983852959484727,
      "language_loss": 0.81188893,
      "learning_rate": 3.4684731579817568e-06,
      "loss": 0.83436918,
      "num_input_tokens_seen": 46342730,
      "step": 2166,
      "time_per_iteration": 2.53603458404541
    },
    {
      "auxiliary_loss_clip": 0.01132057,
      "auxiliary_loss_mlp": 0.0103698,
      "balance_loss_clip": 1.0555687,
      "balance_loss_mlp": 1.02784848,
      "epoch": 0.26056634401491013,
      "flos": 25666757072640.0,
      "grad_norm": 1.5626522140039851,
      "language_loss": 0.76280355,
      "learning_rate": 3.4679442076738247e-06,
      "loss": 0.78449392,
      "num_input_tokens_seen": 46362445,
      "step": 2167,
      "time_per_iteration": 2.657926082611084
    },
    {
      "auxiliary_loss_clip": 0.01212013,
      "auxiliary_loss_mlp": 0.01033837,
      "balance_loss_clip": 1.06358683,
      "balance_loss_mlp": 1.02368653,
      "epoch": 0.2606865869055492,
      "flos": 27052765217280.0,
      "grad_norm": 2.0148815023432887,
      "language_loss": 0.83762741,
      "learning_rate": 3.4674150346799245e-06,
      "loss": 0.86008584,
      "num_input_tokens_seen": 46382145,
      "step": 2168,
      "time_per_iteration": 2.4923644065856934
    },
    {
      "auxiliary_loss_clip": 0.01177443,
      "auxiliary_loss_mlp": 0.01030746,
      "balance_loss_clip": 1.05903304,
      "balance_loss_mlp": 1.0215435,
      "epoch": 0.2608068297961883,
      "flos": 17712615686400.0,
      "grad_norm": 2.297595910295783,
      "language_loss": 0.79767686,
      "learning_rate": 3.4668856390803295e-06,
      "loss": 0.81975877,
      "num_input_tokens_seen": 46400025,
      "step": 2169,
      "time_per_iteration": 2.5130422115325928
    },
    {
      "auxiliary_loss_clip": 0.01183175,
      "auxiliary_loss_mlp": 0.01030862,
      "balance_loss_clip": 1.060076,
      "balance_loss_mlp": 1.02166498,
      "epoch": 0.2609270726868274,
      "flos": 18551632544640.0,
      "grad_norm": 1.8567906303704276,
      "language_loss": 0.89915407,
      "learning_rate": 3.4663560209553495e-06,
      "loss": 0.92129445,
      "num_input_tokens_seen": 46418090,
      "step": 2170,
      "time_per_iteration": 2.4837265014648438
    },
    {
      "auxiliary_loss_clip": 0.01168994,
      "auxiliary_loss_mlp": 0.01032734,
      "balance_loss_clip": 1.05658078,
      "balance_loss_mlp": 1.023579,
      "epoch": 0.26104731557746647,
      "flos": 21835699165440.0,
      "grad_norm": 1.6486232169764274,
      "language_loss": 0.79310489,
      "learning_rate": 3.4658261803853267e-06,
      "loss": 0.81512219,
      "num_input_tokens_seen": 46436015,
      "step": 2171,
      "time_per_iteration": 2.5603272914886475
    },
    {
      "auxiliary_loss_clip": 0.01175017,
      "auxiliary_loss_mlp": 0.01033053,
      "balance_loss_clip": 1.05923963,
      "balance_loss_mlp": 1.02346206,
      "epoch": 0.2611675584681056,
      "flos": 21689650465920.0,
      "grad_norm": 7.403395318207266,
      "language_loss": 0.80895674,
      "learning_rate": 3.4652961174506383e-06,
      "loss": 0.83103752,
      "num_input_tokens_seen": 46455885,
      "step": 2172,
      "time_per_iteration": 2.526472330093384
    },
    {
      "auxiliary_loss_clip": 0.01120973,
      "auxiliary_loss_mlp": 0.01001811,
      "balance_loss_clip": 1.05513811,
      "balance_loss_mlp": 1.00016582,
      "epoch": 0.2612878013587447,
      "flos": 71862101389440.0,
      "grad_norm": 0.9712397895893216,
      "language_loss": 0.58154279,
      "learning_rate": 3.464765832231694e-06,
      "loss": 0.60277063,
      "num_input_tokens_seen": 46510050,
      "step": 2173,
      "time_per_iteration": 3.0684995651245117
    },
    {
      "auxiliary_loss_clip": 0.01196592,
      "auxiliary_loss_mlp": 0.01028922,
      "balance_loss_clip": 1.06342673,
      "balance_loss_mlp": 1.01979637,
      "epoch": 0.26140804424938374,
      "flos": 20227511445120.0,
      "grad_norm": 1.870311859634478,
      "language_loss": 0.7064085,
      "learning_rate": 3.4642353248089373e-06,
      "loss": 0.72866356,
      "num_input_tokens_seen": 46528810,
      "step": 2174,
      "time_per_iteration": 2.4820303916931152
    },
    {
      "auxiliary_loss_clip": 0.01173217,
      "auxiliary_loss_mlp": 0.0102929,
      "balance_loss_clip": 1.05699348,
      "balance_loss_mlp": 1.01954424,
      "epoch": 0.26152828714002285,
      "flos": 25557085872000.0,
      "grad_norm": 1.629799740219285,
      "language_loss": 0.80035865,
      "learning_rate": 3.463704595262846e-06,
      "loss": 0.82238376,
      "num_input_tokens_seen": 46549690,
      "step": 2175,
      "time_per_iteration": 2.5791454315185547
    },
    {
      "auxiliary_loss_clip": 0.01160216,
      "auxiliary_loss_mlp": 0.0103619,
      "balance_loss_clip": 1.05637741,
      "balance_loss_mlp": 1.02717173,
      "epoch": 0.26164853003066196,
      "flos": 25446516831360.0,
      "grad_norm": 2.072702741122209,
      "language_loss": 0.70460141,
      "learning_rate": 3.463173643673931e-06,
      "loss": 0.72656536,
      "num_input_tokens_seen": 46572215,
      "step": 2176,
      "time_per_iteration": 2.6131815910339355
    },
    {
      "auxiliary_loss_clip": 0.01128957,
      "auxiliary_loss_mlp": 0.0100988,
      "balance_loss_clip": 1.05613852,
      "balance_loss_mlp": 1.00804412,
      "epoch": 0.261768772921301,
      "flos": 53944580568960.0,
      "grad_norm": 0.9016404712133905,
      "language_loss": 0.63482016,
      "learning_rate": 3.4626424701227387e-06,
      "loss": 0.65620857,
      "num_input_tokens_seen": 46627275,
      "step": 2177,
      "time_per_iteration": 3.012408971786499
    },
    {
      "auxiliary_loss_clip": 0.01138466,
      "auxiliary_loss_mlp": 0.0100753,
      "balance_loss_clip": 1.05523539,
      "balance_loss_mlp": 1.00567079,
      "epoch": 0.26188901581194013,
      "flos": 70687606481280.0,
      "grad_norm": 0.8200781995043724,
      "language_loss": 0.55837572,
      "learning_rate": 3.4621110746898452e-06,
      "loss": 0.57983571,
      "num_input_tokens_seen": 46695135,
      "step": 2178,
      "time_per_iteration": 4.358171701431274
    },
    {
      "auxiliary_loss_clip": 0.01197398,
      "auxiliary_loss_mlp": 0.01032189,
      "balance_loss_clip": 1.06295514,
      "balance_loss_mlp": 1.02277136,
      "epoch": 0.2620092587025792,
      "flos": 21069580959360.0,
      "grad_norm": 1.6563492262859696,
      "language_loss": 0.74328589,
      "learning_rate": 3.4615794574558654e-06,
      "loss": 0.76558173,
      "num_input_tokens_seen": 46714145,
      "step": 2179,
      "time_per_iteration": 2.561577796936035
    },
    {
      "auxiliary_loss_clip": 0.01178204,
      "auxiliary_loss_mlp": 0.01025749,
      "balance_loss_clip": 1.05941629,
      "balance_loss_mlp": 1.01761293,
      "epoch": 0.2621295015932183,
      "flos": 18369601395840.0,
      "grad_norm": 3.124222071143197,
      "language_loss": 0.83916706,
      "learning_rate": 3.4610476185014436e-06,
      "loss": 0.86120659,
      "num_input_tokens_seen": 46731405,
      "step": 2180,
      "time_per_iteration": 2.5642709732055664
    },
    {
      "auxiliary_loss_clip": 0.01208678,
      "auxiliary_loss_mlp": 0.01031623,
      "balance_loss_clip": 1.06085014,
      "balance_loss_mlp": 1.02179396,
      "epoch": 0.2622497444838574,
      "flos": 23659997063040.0,
      "grad_norm": 1.8866342374167722,
      "language_loss": 0.79257935,
      "learning_rate": 3.4605155579072597e-06,
      "loss": 0.81498241,
      "num_input_tokens_seen": 46751260,
      "step": 2181,
      "time_per_iteration": 2.545687437057495
    },
    {
      "auxiliary_loss_clip": 0.01138863,
      "auxiliary_loss_mlp": 0.0102935,
      "balance_loss_clip": 1.05171895,
      "balance_loss_mlp": 1.02037311,
      "epoch": 0.26236998737449646,
      "flos": 22123810154880.0,
      "grad_norm": 1.7193284635237824,
      "language_loss": 0.71557009,
      "learning_rate": 3.459983275754027e-06,
      "loss": 0.73725224,
      "num_input_tokens_seen": 46770155,
      "step": 2182,
      "time_per_iteration": 2.701566219329834
    },
    {
      "auxiliary_loss_clip": 0.01208017,
      "auxiliary_loss_mlp": 0.01029567,
      "balance_loss_clip": 1.06258309,
      "balance_loss_mlp": 1.02051926,
      "epoch": 0.26249023026513557,
      "flos": 17895185539200.0,
      "grad_norm": 2.709506957094928,
      "language_loss": 0.80074733,
      "learning_rate": 3.4594507721224918e-06,
      "loss": 0.82312316,
      "num_input_tokens_seen": 46788805,
      "step": 2183,
      "time_per_iteration": 3.352287530899048
    },
    {
      "auxiliary_loss_clip": 0.01178987,
      "auxiliary_loss_mlp": 0.01041921,
      "balance_loss_clip": 1.05708432,
      "balance_loss_mlp": 1.03295064,
      "epoch": 0.2626104731557747,
      "flos": 18332936588160.0,
      "grad_norm": 1.636558240169881,
      "language_loss": 0.82141447,
      "learning_rate": 3.4589180470934353e-06,
      "loss": 0.84362358,
      "num_input_tokens_seen": 46808670,
      "step": 2184,
      "time_per_iteration": 4.191047191619873
    },
    {
      "auxiliary_loss_clip": 0.01199236,
      "auxiliary_loss_mlp": 0.0103517,
      "balance_loss_clip": 1.05877709,
      "balance_loss_mlp": 1.0253948,
      "epoch": 0.26273071604641374,
      "flos": 19317714837120.0,
      "grad_norm": 1.7760122200612447,
      "language_loss": 0.76692963,
      "learning_rate": 3.4583851007476713e-06,
      "loss": 0.78927362,
      "num_input_tokens_seen": 46827140,
      "step": 2185,
      "time_per_iteration": 2.5613741874694824
    },
    {
      "auxiliary_loss_clip": 0.01165308,
      "auxiliary_loss_mlp": 0.01031117,
      "balance_loss_clip": 1.05647099,
      "balance_loss_mlp": 1.02132988,
      "epoch": 0.26285095893705285,
      "flos": 18327477720960.0,
      "grad_norm": 2.0776210310879475,
      "language_loss": 0.6891613,
      "learning_rate": 3.4578519331660464e-06,
      "loss": 0.71112555,
      "num_input_tokens_seen": 46844135,
      "step": 2186,
      "time_per_iteration": 2.7397818565368652
    },
    {
      "auxiliary_loss_clip": 0.01190579,
      "auxiliary_loss_mlp": 0.01032679,
      "balance_loss_clip": 1.06217432,
      "balance_loss_mlp": 1.02412593,
      "epoch": 0.26297120182769196,
      "flos": 20193827466240.0,
      "grad_norm": 2.096911355778989,
      "language_loss": 0.81944418,
      "learning_rate": 3.4573185444294426e-06,
      "loss": 0.84167671,
      "num_input_tokens_seen": 46862500,
      "step": 2187,
      "time_per_iteration": 2.6239640712738037
    },
    {
      "auxiliary_loss_clip": 0.01175456,
      "auxiliary_loss_mlp": 0.00763908,
      "balance_loss_clip": 1.05786455,
      "balance_loss_mlp": 1.00085139,
      "epoch": 0.263091444718331,
      "flos": 22418421505920.0,
      "grad_norm": 1.7986511251009665,
      "language_loss": 0.79012066,
      "learning_rate": 3.456784934618774e-06,
      "loss": 0.80951428,
      "num_input_tokens_seen": 46883665,
      "step": 2188,
      "time_per_iteration": 2.6499650478363037
    },
    {
      "auxiliary_loss_clip": 0.01177985,
      "auxiliary_loss_mlp": 0.01031499,
      "balance_loss_clip": 1.05916893,
      "balance_loss_mlp": 1.02302361,
      "epoch": 0.2632116876089701,
      "flos": 19024827338880.0,
      "grad_norm": 1.9286659506983903,
      "language_loss": 0.80325353,
      "learning_rate": 3.4562511038149897e-06,
      "loss": 0.82534832,
      "num_input_tokens_seen": 46899160,
      "step": 2189,
      "time_per_iteration": 2.610973834991455
    },
    {
      "auxiliary_loss_clip": 0.0107272,
      "auxiliary_loss_mlp": 0.0100414,
      "balance_loss_clip": 1.04186225,
      "balance_loss_mlp": 1.00206554,
      "epoch": 0.26333193049960923,
      "flos": 67308054531840.0,
      "grad_norm": 0.8596719810699128,
      "language_loss": 0.57779741,
      "learning_rate": 3.4557170520990705e-06,
      "loss": 0.59856594,
      "num_input_tokens_seen": 46959835,
      "step": 2190,
      "time_per_iteration": 3.254857301712036
    },
    {
      "auxiliary_loss_clip": 0.01184648,
      "auxiliary_loss_mlp": 0.01033131,
      "balance_loss_clip": 1.05729985,
      "balance_loss_mlp": 1.02413034,
      "epoch": 0.2634521733902483,
      "flos": 25048806468480.0,
      "grad_norm": 2.596168456299673,
      "language_loss": 0.86207569,
      "learning_rate": 3.4551827795520324e-06,
      "loss": 0.88425344,
      "num_input_tokens_seen": 46982720,
      "step": 2191,
      "time_per_iteration": 2.6148581504821777
    },
    {
      "auxiliary_loss_clip": 0.01192949,
      "auxiliary_loss_mlp": 0.010263,
      "balance_loss_clip": 1.05866671,
      "balance_loss_mlp": 1.01775312,
      "epoch": 0.2635724162808874,
      "flos": 20594985534720.0,
      "grad_norm": 1.6603366512539401,
      "language_loss": 0.85139942,
      "learning_rate": 3.4546482862549226e-06,
      "loss": 0.8735919,
      "num_input_tokens_seen": 47003035,
      "step": 2192,
      "time_per_iteration": 2.573833703994751
    },
    {
      "auxiliary_loss_clip": 0.0115699,
      "auxiliary_loss_mlp": 0.01038709,
      "balance_loss_clip": 1.05532324,
      "balance_loss_mlp": 1.02886784,
      "epoch": 0.2636926591715265,
      "flos": 19244636616960.0,
      "grad_norm": 2.106359142695926,
      "language_loss": 0.78791392,
      "learning_rate": 3.4541135722888253e-06,
      "loss": 0.80987096,
      "num_input_tokens_seen": 47019625,
      "step": 2193,
      "time_per_iteration": 2.6386377811431885
    },
    {
      "auxiliary_loss_clip": 0.01203023,
      "auxiliary_loss_mlp": 0.01029189,
      "balance_loss_clip": 1.05889523,
      "balance_loss_mlp": 1.01966405,
      "epoch": 0.26381290206216557,
      "flos": 28804882734720.0,
      "grad_norm": 1.9076585989765682,
      "language_loss": 0.80558097,
      "learning_rate": 3.453578637734854e-06,
      "loss": 0.82790315,
      "num_input_tokens_seen": 47040815,
      "step": 2194,
      "time_per_iteration": 2.6013710498809814
    },
    {
      "auxiliary_loss_clip": 0.01209885,
      "auxiliary_loss_mlp": 0.01034202,
      "balance_loss_clip": 1.06535769,
      "balance_loss_mlp": 1.02486193,
      "epoch": 0.2639331449528047,
      "flos": 25008909436800.0,
      "grad_norm": 1.6391315992143196,
      "language_loss": 0.78201735,
      "learning_rate": 3.4530434826741605e-06,
      "loss": 0.80445826,
      "num_input_tokens_seen": 47061755,
      "step": 2195,
      "time_per_iteration": 2.553014039993286
    },
    {
      "auxiliary_loss_clip": 0.01173584,
      "auxiliary_loss_mlp": 0.01033158,
      "balance_loss_clip": 1.05922675,
      "balance_loss_mlp": 1.02456856,
      "epoch": 0.26405338784344373,
      "flos": 46535775465600.0,
      "grad_norm": 2.0830058878795175,
      "language_loss": 0.69017345,
      "learning_rate": 3.452508107187926e-06,
      "loss": 0.71224082,
      "num_input_tokens_seen": 47085130,
      "step": 2196,
      "time_per_iteration": 2.774874687194824
    },
    {
      "auxiliary_loss_clip": 0.01130644,
      "auxiliary_loss_mlp": 0.0102902,
      "balance_loss_clip": 1.04896259,
      "balance_loss_mlp": 1.01932192,
      "epoch": 0.26417363073408284,
      "flos": 21179467641600.0,
      "grad_norm": 2.514524947463671,
      "language_loss": 0.77544224,
      "learning_rate": 3.451972511357366e-06,
      "loss": 0.79703885,
      "num_input_tokens_seen": 47104675,
      "step": 2197,
      "time_per_iteration": 2.6837775707244873
    },
    {
      "auxiliary_loss_clip": 0.01189841,
      "auxiliary_loss_mlp": 0.01030978,
      "balance_loss_clip": 1.06157112,
      "balance_loss_mlp": 1.02299118,
      "epoch": 0.26429387362472195,
      "flos": 22674751937280.0,
      "grad_norm": 1.6631402666991062,
      "language_loss": 0.85201919,
      "learning_rate": 3.45143669526373e-06,
      "loss": 0.87422729,
      "num_input_tokens_seen": 47124435,
      "step": 2198,
      "time_per_iteration": 2.550856590270996
    },
    {
      "auxiliary_loss_clip": 0.01112672,
      "auxiliary_loss_mlp": 0.0100712,
      "balance_loss_clip": 1.05072165,
      "balance_loss_mlp": 1.00540304,
      "epoch": 0.264414116515361,
      "flos": 67180534272000.0,
      "grad_norm": 0.7883361354841523,
      "language_loss": 0.63230938,
      "learning_rate": 3.450900658988302e-06,
      "loss": 0.65350729,
      "num_input_tokens_seen": 47185985,
      "step": 2199,
      "time_per_iteration": 3.0846142768859863
    },
    {
      "auxiliary_loss_clip": 0.01167213,
      "auxiliary_loss_mlp": 0.01036646,
      "balance_loss_clip": 1.05756617,
      "balance_loss_mlp": 1.02717495,
      "epoch": 0.2645343594060001,
      "flos": 25664709997440.0,
      "grad_norm": 1.8785795321841914,
      "language_loss": 0.77414012,
      "learning_rate": 3.450364402612397e-06,
      "loss": 0.79617876,
      "num_input_tokens_seen": 47203140,
      "step": 2200,
      "time_per_iteration": 2.60007905960083
    },
    {
      "auxiliary_loss_clip": 0.0117046,
      "auxiliary_loss_mlp": 0.01032737,
      "balance_loss_clip": 1.05552101,
      "balance_loss_mlp": 1.02308655,
      "epoch": 0.26465460229663923,
      "flos": 22491822948480.0,
      "grad_norm": 1.9171533015382658,
      "language_loss": 0.83797586,
      "learning_rate": 3.449827926217366e-06,
      "loss": 0.86000788,
      "num_input_tokens_seen": 47222575,
      "step": 2201,
      "time_per_iteration": 2.5830206871032715
    },
    {
      "auxiliary_loss_clip": 0.01180158,
      "auxiliary_loss_mlp": 0.01031717,
      "balance_loss_clip": 1.05565095,
      "balance_loss_mlp": 1.0224961,
      "epoch": 0.2647748451872783,
      "flos": 29388036038400.0,
      "grad_norm": 1.8039305930145464,
      "language_loss": 0.80450892,
      "learning_rate": 3.449291229884591e-06,
      "loss": 0.82662767,
      "num_input_tokens_seen": 47243815,
      "step": 2202,
      "time_per_iteration": 2.6505749225616455
    },
    {
      "auxiliary_loss_clip": 0.01167585,
      "auxiliary_loss_mlp": 0.01029513,
      "balance_loss_clip": 1.05624235,
      "balance_loss_mlp": 1.02018523,
      "epoch": 0.2648950880779174,
      "flos": 26797799502720.0,
      "grad_norm": 3.3983243971699366,
      "language_loss": 0.86773252,
      "learning_rate": 3.4487543136954887e-06,
      "loss": 0.88970351,
      "num_input_tokens_seen": 47263435,
      "step": 2203,
      "time_per_iteration": 2.6584179401397705
    },
    {
      "auxiliary_loss_clip": 0.0116103,
      "auxiliary_loss_mlp": 0.01034696,
      "balance_loss_clip": 1.05634904,
      "balance_loss_mlp": 1.02539182,
      "epoch": 0.2650153309685565,
      "flos": 28841008838400.0,
      "grad_norm": 1.659183504880564,
      "language_loss": 0.91029167,
      "learning_rate": 3.448217177731509e-06,
      "loss": 0.93224883,
      "num_input_tokens_seen": 47283920,
      "step": 2204,
      "time_per_iteration": 2.6511173248291016
    },
    {
      "auxiliary_loss_clip": 0.01171713,
      "auxiliary_loss_mlp": 0.01032123,
      "balance_loss_clip": 1.05940711,
      "balance_loss_mlp": 1.02384424,
      "epoch": 0.26513557385919556,
      "flos": 20303247271680.0,
      "grad_norm": 2.069605312070416,
      "language_loss": 0.78080666,
      "learning_rate": 3.4476798220741348e-06,
      "loss": 0.80284506,
      "num_input_tokens_seen": 47302800,
      "step": 2205,
      "time_per_iteration": 3.443758249282837
    },
    {
      "auxiliary_loss_clip": 0.01207371,
      "auxiliary_loss_mlp": 0.01032254,
      "balance_loss_clip": 1.06392574,
      "balance_loss_mlp": 1.02402902,
      "epoch": 0.26525581674983467,
      "flos": 17676274101120.0,
      "grad_norm": 1.571396884156926,
      "language_loss": 0.78391492,
      "learning_rate": 3.4471422468048826e-06,
      "loss": 0.80631113,
      "num_input_tokens_seen": 47321525,
      "step": 2206,
      "time_per_iteration": 2.517871856689453
    },
    {
      "auxiliary_loss_clip": 0.01183666,
      "auxiliary_loss_mlp": 0.01028851,
      "balance_loss_clip": 1.06074798,
      "balance_loss_mlp": 1.0202744,
      "epoch": 0.2653760596404738,
      "flos": 26833746038400.0,
      "grad_norm": 3.022226023368953,
      "language_loss": 0.73499399,
      "learning_rate": 3.4466044520053022e-06,
      "loss": 0.75711918,
      "num_input_tokens_seen": 47340530,
      "step": 2207,
      "time_per_iteration": 2.5882511138916016
    },
    {
      "auxiliary_loss_clip": 0.01163886,
      "auxiliary_loss_mlp": 0.01032197,
      "balance_loss_clip": 1.05616009,
      "balance_loss_mlp": 1.02327967,
      "epoch": 0.26549630253111284,
      "flos": 22782160581120.0,
      "grad_norm": 1.9279362087523995,
      "language_loss": 0.60457009,
      "learning_rate": 3.446066437756977e-06,
      "loss": 0.62653089,
      "num_input_tokens_seen": 47359735,
      "step": 2208,
      "time_per_iteration": 2.584620714187622
    },
    {
      "auxiliary_loss_clip": 0.01174298,
      "auxiliary_loss_mlp": 0.01024962,
      "balance_loss_clip": 1.05758798,
      "balance_loss_mlp": 1.01624823,
      "epoch": 0.26561654542175195,
      "flos": 23550002640000.0,
      "grad_norm": 2.2992960028362135,
      "language_loss": 0.75146151,
      "learning_rate": 3.4455282041415224e-06,
      "loss": 0.77345413,
      "num_input_tokens_seen": 47378945,
      "step": 2209,
      "time_per_iteration": 2.6133382320404053
    },
    {
      "auxiliary_loss_clip": 0.01166358,
      "auxiliary_loss_mlp": 0.01028491,
      "balance_loss_clip": 1.0576725,
      "balance_loss_mlp": 1.01975298,
      "epoch": 0.265736788312391,
      "flos": 26906680604160.0,
      "grad_norm": 2.506035992973833,
      "language_loss": 0.86997288,
      "learning_rate": 3.4449897512405894e-06,
      "loss": 0.8919214,
      "num_input_tokens_seen": 47398095,
      "step": 2210,
      "time_per_iteration": 4.433354139328003
    },
    {
      "auxiliary_loss_clip": 0.01124917,
      "auxiliary_loss_mlp": 0.00763939,
      "balance_loss_clip": 1.05033374,
      "balance_loss_mlp": 1.00095773,
      "epoch": 0.2658570312030301,
      "flos": 23477139901440.0,
      "grad_norm": 2.5691505657724,
      "language_loss": 0.7501049,
      "learning_rate": 3.444451079135859e-06,
      "loss": 0.76899344,
      "num_input_tokens_seen": 47417605,
      "step": 2211,
      "time_per_iteration": 2.7216885089874268
    },
    {
      "auxiliary_loss_clip": 0.01135794,
      "auxiliary_loss_mlp": 0.00764254,
      "balance_loss_clip": 1.05027723,
      "balance_loss_mlp": 1.00095475,
      "epoch": 0.2659772740936692,
      "flos": 21866402315520.0,
      "grad_norm": 4.164591014971873,
      "language_loss": 0.74083984,
      "learning_rate": 3.4439121879090493e-06,
      "loss": 0.75984025,
      "num_input_tokens_seen": 47435385,
      "step": 2212,
      "time_per_iteration": 2.6728148460388184
    },
    {
      "auxiliary_loss_clip": 0.01180487,
      "auxiliary_loss_mlp": 0.0103204,
      "balance_loss_clip": 1.05928302,
      "balance_loss_mlp": 1.02285492,
      "epoch": 0.2660975169843083,
      "flos": 19793100360960.0,
      "grad_norm": 1.8156408081871496,
      "language_loss": 0.83416331,
      "learning_rate": 3.4433730776419082e-06,
      "loss": 0.85628855,
      "num_input_tokens_seen": 47454310,
      "step": 2213,
      "time_per_iteration": 2.5893635749816895
    },
    {
      "auxiliary_loss_clip": 0.01194827,
      "auxiliary_loss_mlp": 0.00764103,
      "balance_loss_clip": 1.05907667,
      "balance_loss_mlp": 1.00104117,
      "epoch": 0.2662177598749474,
      "flos": 29018981750400.0,
      "grad_norm": 2.88617253080466,
      "language_loss": 0.8063392,
      "learning_rate": 3.4428337484162183e-06,
      "loss": 0.82592845,
      "num_input_tokens_seen": 47475120,
      "step": 2214,
      "time_per_iteration": 2.6042234897613525
    },
    {
      "auxiliary_loss_clip": 0.0117323,
      "auxiliary_loss_mlp": 0.0103082,
      "balance_loss_clip": 1.05683851,
      "balance_loss_mlp": 1.02192712,
      "epoch": 0.2663380027655865,
      "flos": 21762549118080.0,
      "grad_norm": 1.8702691314145343,
      "language_loss": 0.84157419,
      "learning_rate": 3.442294200313797e-06,
      "loss": 0.86361468,
      "num_input_tokens_seen": 47493150,
      "step": 2215,
      "time_per_iteration": 2.5609002113342285
    },
    {
      "auxiliary_loss_clip": 0.01129059,
      "auxiliary_loss_mlp": 0.0100152,
      "balance_loss_clip": 1.04827094,
      "balance_loss_mlp": 1.00000596,
      "epoch": 0.26645824565622556,
      "flos": 66980333819520.0,
      "grad_norm": 0.766375761484475,
      "language_loss": 0.52656794,
      "learning_rate": 3.4417544334164916e-06,
      "loss": 0.54787374,
      "num_input_tokens_seen": 47557295,
      "step": 2216,
      "time_per_iteration": 3.1085009574890137
    },
    {
      "auxiliary_loss_clip": 0.0113574,
      "auxiliary_loss_mlp": 0.01031027,
      "balance_loss_clip": 1.05086994,
      "balance_loss_mlp": 1.02155602,
      "epoch": 0.26657848854686467,
      "flos": 25264198373760.0,
      "grad_norm": 1.7237241397883767,
      "language_loss": 0.77284902,
      "learning_rate": 3.4412144478061854e-06,
      "loss": 0.79451668,
      "num_input_tokens_seen": 47579705,
      "step": 2217,
      "time_per_iteration": 2.688756227493286
    },
    {
      "auxiliary_loss_clip": 0.01117476,
      "auxiliary_loss_mlp": 0.01030548,
      "balance_loss_clip": 1.0491854,
      "balance_loss_mlp": 1.0210824,
      "epoch": 0.2666987314375038,
      "flos": 23696769611520.0,
      "grad_norm": 1.8510502611519608,
      "language_loss": 0.75081944,
      "learning_rate": 3.4406742435647925e-06,
      "loss": 0.77229965,
      "num_input_tokens_seen": 47599770,
      "step": 2218,
      "time_per_iteration": 2.677963972091675
    },
    {
      "auxiliary_loss_clip": 0.01187372,
      "auxiliary_loss_mlp": 0.01032668,
      "balance_loss_clip": 1.06072783,
      "balance_loss_mlp": 1.02397799,
      "epoch": 0.26681897432814283,
      "flos": 27048958375680.0,
      "grad_norm": 2.0988809630433156,
      "language_loss": 0.78762227,
      "learning_rate": 3.440133820774263e-06,
      "loss": 0.80982268,
      "num_input_tokens_seen": 47619580,
      "step": 2219,
      "time_per_iteration": 2.612597703933716
    },
    {
      "auxiliary_loss_clip": 0.01177929,
      "auxiliary_loss_mlp": 0.01038852,
      "balance_loss_clip": 1.05692625,
      "balance_loss_mlp": 1.02911866,
      "epoch": 0.26693921721878194,
      "flos": 28985944216320.0,
      "grad_norm": 1.9719463049090495,
      "language_loss": 0.8199681,
      "learning_rate": 3.439593179516578e-06,
      "loss": 0.84213591,
      "num_input_tokens_seen": 47639490,
      "step": 2220,
      "time_per_iteration": 2.597015142440796
    },
    {
      "auxiliary_loss_clip": 0.01180935,
      "auxiliary_loss_mlp": 0.0103203,
      "balance_loss_clip": 1.05918884,
      "balance_loss_mlp": 1.02271366,
      "epoch": 0.26705946010942105,
      "flos": 21507834798720.0,
      "grad_norm": 1.7801340039728282,
      "language_loss": 0.81058663,
      "learning_rate": 3.4390523198737524e-06,
      "loss": 0.83271623,
      "num_input_tokens_seen": 47658650,
      "step": 2221,
      "time_per_iteration": 2.5597517490386963
    },
    {
      "auxiliary_loss_clip": 0.01205964,
      "auxiliary_loss_mlp": 0.00764025,
      "balance_loss_clip": 1.06115246,
      "balance_loss_mlp": 1.00104702,
      "epoch": 0.2671797030000601,
      "flos": 21471277731840.0,
      "grad_norm": 1.5964224359378898,
      "language_loss": 0.73500913,
      "learning_rate": 3.4385112419278333e-06,
      "loss": 0.75470906,
      "num_input_tokens_seen": 47679875,
      "step": 2222,
      "time_per_iteration": 2.51297664642334
    },
    {
      "auxiliary_loss_clip": 0.01119354,
      "auxiliary_loss_mlp": 0.0100569,
      "balance_loss_clip": 1.04763269,
      "balance_loss_mlp": 1.00414073,
      "epoch": 0.2672999458906992,
      "flos": 64189929767040.0,
      "grad_norm": 0.7976912376205909,
      "language_loss": 0.64793193,
      "learning_rate": 3.4379699457609033e-06,
      "loss": 0.66918242,
      "num_input_tokens_seen": 47737700,
      "step": 2223,
      "time_per_iteration": 2.986011028289795
    },
    {
      "auxiliary_loss_clip": 0.01166102,
      "auxiliary_loss_mlp": 0.01026669,
      "balance_loss_clip": 1.0538398,
      "balance_loss_mlp": 1.01759088,
      "epoch": 0.26742018878133833,
      "flos": 16909042573440.0,
      "grad_norm": 1.9473485888576572,
      "language_loss": 0.90078104,
      "learning_rate": 3.4374284314550755e-06,
      "loss": 0.92270875,
      "num_input_tokens_seen": 47756740,
      "step": 2224,
      "time_per_iteration": 2.568167209625244
    },
    {
      "auxiliary_loss_clip": 0.01203695,
      "auxiliary_loss_mlp": 0.01026819,
      "balance_loss_clip": 1.06096828,
      "balance_loss_mlp": 1.01820922,
      "epoch": 0.2675404316719774,
      "flos": 20667560964480.0,
      "grad_norm": 2.000341276099548,
      "language_loss": 0.80752969,
      "learning_rate": 3.436886699092498e-06,
      "loss": 0.82983488,
      "num_input_tokens_seen": 47775255,
      "step": 2225,
      "time_per_iteration": 2.687025308609009
    },
    {
      "auxiliary_loss_clip": 0.01206604,
      "auxiliary_loss_mlp": 0.01033894,
      "balance_loss_clip": 1.06055307,
      "balance_loss_mlp": 1.02459025,
      "epoch": 0.2676606745626165,
      "flos": 17485013157120.0,
      "grad_norm": 2.7264236370420205,
      "language_loss": 0.71757275,
      "learning_rate": 3.4363447487553502e-06,
      "loss": 0.73997772,
      "num_input_tokens_seen": 47788570,
      "step": 2226,
      "time_per_iteration": 2.472900152206421
    },
    {
      "auxiliary_loss_clip": 0.01170698,
      "auxiliary_loss_mlp": 0.01032031,
      "balance_loss_clip": 1.05710268,
      "balance_loss_mlp": 1.02229166,
      "epoch": 0.26778091745325555,
      "flos": 27852675143040.0,
      "grad_norm": 1.8891480480563503,
      "language_loss": 0.77584475,
      "learning_rate": 3.4358025805258455e-06,
      "loss": 0.79787207,
      "num_input_tokens_seen": 47808275,
      "step": 2227,
      "time_per_iteration": 2.6246864795684814
    },
    {
      "auxiliary_loss_clip": 0.01151102,
      "auxiliary_loss_mlp": 0.01024657,
      "balance_loss_clip": 1.05271149,
      "balance_loss_mlp": 1.01568055,
      "epoch": 0.26790116034389466,
      "flos": 20955995176320.0,
      "grad_norm": 1.6981420800543432,
      "language_loss": 0.83317292,
      "learning_rate": 3.435260194486232e-06,
      "loss": 0.85493058,
      "num_input_tokens_seen": 47826245,
      "step": 2228,
      "time_per_iteration": 2.6458258628845215
    },
    {
      "auxiliary_loss_clip": 0.01174902,
      "auxiliary_loss_mlp": 0.01029675,
      "balance_loss_clip": 1.05742013,
      "balance_loss_mlp": 1.02044272,
      "epoch": 0.2680214032345338,
      "flos": 18040659621120.0,
      "grad_norm": 3.3687843102912742,
      "language_loss": 0.82139403,
      "learning_rate": 3.4347175907187875e-06,
      "loss": 0.84343982,
      "num_input_tokens_seen": 47843235,
      "step": 2229,
      "time_per_iteration": 2.531506061553955
    },
    {
      "auxiliary_loss_clip": 0.01178841,
      "auxiliary_loss_mlp": 0.01036322,
      "balance_loss_clip": 1.05836511,
      "balance_loss_mlp": 1.02801943,
      "epoch": 0.26814164612517283,
      "flos": 22419427086720.0,
      "grad_norm": 1.7159714206015537,
      "language_loss": 0.879282,
      "learning_rate": 3.4341747693058254e-06,
      "loss": 0.90143371,
      "num_input_tokens_seen": 47861710,
      "step": 2230,
      "time_per_iteration": 2.546628952026367
    },
    {
      "auxiliary_loss_clip": 0.01106508,
      "auxiliary_loss_mlp": 0.01029378,
      "balance_loss_clip": 1.05124855,
      "balance_loss_mlp": 1.02104545,
      "epoch": 0.26826188901581194,
      "flos": 35627371159680.0,
      "grad_norm": 1.9615332249089135,
      "language_loss": 0.77418643,
      "learning_rate": 3.4336317303296916e-06,
      "loss": 0.79554534,
      "num_input_tokens_seen": 47882685,
      "step": 2231,
      "time_per_iteration": 3.6975531578063965
    },
    {
      "auxiliary_loss_clip": 0.01184785,
      "auxiliary_loss_mlp": 0.01026561,
      "balance_loss_clip": 1.05721736,
      "balance_loss_mlp": 1.01787663,
      "epoch": 0.26838213190645105,
      "flos": 17639788861440.0,
      "grad_norm": 2.0245650612079857,
      "language_loss": 0.75145757,
      "learning_rate": 3.4330884738727635e-06,
      "loss": 0.77357101,
      "num_input_tokens_seen": 47900860,
      "step": 2232,
      "time_per_iteration": 2.540956974029541
    },
    {
      "auxiliary_loss_clip": 0.01136939,
      "auxiliary_loss_mlp": 0.01028713,
      "balance_loss_clip": 1.05197644,
      "balance_loss_mlp": 1.01986146,
      "epoch": 0.2685023747970901,
      "flos": 22674823764480.0,
      "grad_norm": 1.850396606133055,
      "language_loss": 0.70930791,
      "learning_rate": 3.4325450000174535e-06,
      "loss": 0.73096442,
      "num_input_tokens_seen": 47917500,
      "step": 2233,
      "time_per_iteration": 2.6546518802642822
    },
    {
      "auxiliary_loss_clip": 0.0113734,
      "auxiliary_loss_mlp": 0.01034148,
      "balance_loss_clip": 1.05256629,
      "balance_loss_mlp": 1.02443218,
      "epoch": 0.2686226176877292,
      "flos": 20120533764480.0,
      "grad_norm": 1.7274205688248698,
      "language_loss": 0.74145687,
      "learning_rate": 3.4320013088462067e-06,
      "loss": 0.76317173,
      "num_input_tokens_seen": 47934860,
      "step": 2234,
      "time_per_iteration": 2.616955280303955
    },
    {
      "auxiliary_loss_clip": 0.01163611,
      "auxiliary_loss_mlp": 0.01029775,
      "balance_loss_clip": 1.05412567,
      "balance_loss_mlp": 1.02103734,
      "epoch": 0.2687428605783683,
      "flos": 21872040750720.0,
      "grad_norm": 1.4676258164373803,
      "language_loss": 0.8179189,
      "learning_rate": 3.431457400441499e-06,
      "loss": 0.83985281,
      "num_input_tokens_seen": 47955255,
      "step": 2235,
      "time_per_iteration": 2.6273481845855713
    },
    {
      "auxiliary_loss_clip": 0.01050567,
      "auxiliary_loss_mlp": 0.01007846,
      "balance_loss_clip": 1.03402674,
      "balance_loss_mlp": 1.00623703,
      "epoch": 0.2688631034690074,
      "flos": 69943320766080.0,
      "grad_norm": 0.9279174900054146,
      "language_loss": 0.60790086,
      "learning_rate": 3.4309132748858424e-06,
      "loss": 0.62848496,
      "num_input_tokens_seen": 48016245,
      "step": 2236,
      "time_per_iteration": 4.947690963745117
    },
    {
      "auxiliary_loss_clip": 0.01184347,
      "auxiliary_loss_mlp": 0.01032172,
      "balance_loss_clip": 1.0592854,
      "balance_loss_mlp": 1.02332115,
      "epoch": 0.2689833463596465,
      "flos": 22856639431680.0,
      "grad_norm": 2.0357490634703126,
      "language_loss": 0.83803993,
      "learning_rate": 3.430368932261779e-06,
      "loss": 0.86020511,
      "num_input_tokens_seen": 48036600,
      "step": 2237,
      "time_per_iteration": 3.3660030364990234
    },
    {
      "auxiliary_loss_clip": 0.0117048,
      "auxiliary_loss_mlp": 0.01029827,
      "balance_loss_clip": 1.05655789,
      "balance_loss_mlp": 1.02069557,
      "epoch": 0.2691035892502856,
      "flos": 17200242132480.0,
      "grad_norm": 1.9003080193379176,
      "language_loss": 0.75151074,
      "learning_rate": 3.429824372651886e-06,
      "loss": 0.77351379,
      "num_input_tokens_seen": 48054750,
      "step": 2238,
      "time_per_iteration": 2.5235631465911865
    },
    {
      "auxiliary_loss_clip": 0.01150177,
      "auxiliary_loss_mlp": 0.01035213,
      "balance_loss_clip": 1.05424917,
      "balance_loss_mlp": 1.02627242,
      "epoch": 0.26922383214092466,
      "flos": 17747484814080.0,
      "grad_norm": 1.9934701546489355,
      "language_loss": 0.8366797,
      "learning_rate": 3.4292795961387732e-06,
      "loss": 0.85853362,
      "num_input_tokens_seen": 48072650,
      "step": 2239,
      "time_per_iteration": 2.696112871170044
    },
    {
      "auxiliary_loss_clip": 0.01202859,
      "auxiliary_loss_mlp": 0.01032386,
      "balance_loss_clip": 1.05973649,
      "balance_loss_mlp": 1.02382112,
      "epoch": 0.26934407503156377,
      "flos": 16173376122240.0,
      "grad_norm": 2.1702667712905153,
      "language_loss": 0.87873799,
      "learning_rate": 3.4287346028050818e-06,
      "loss": 0.90109044,
      "num_input_tokens_seen": 48088720,
      "step": 2240,
      "time_per_iteration": 2.4811339378356934
    },
    {
      "auxiliary_loss_clip": 0.01169416,
      "auxiliary_loss_mlp": 0.01027507,
      "balance_loss_clip": 1.05507183,
      "balance_loss_mlp": 1.01900136,
      "epoch": 0.2694643179222028,
      "flos": 23732895715200.0,
      "grad_norm": 1.7180874399626538,
      "language_loss": 0.7955277,
      "learning_rate": 3.4281893927334866e-06,
      "loss": 0.81749696,
      "num_input_tokens_seen": 48108630,
      "step": 2241,
      "time_per_iteration": 2.637162923812866
    },
    {
      "auxiliary_loss_clip": 0.01188866,
      "auxiliary_loss_mlp": 0.01028182,
      "balance_loss_clip": 1.06009424,
      "balance_loss_mlp": 1.01996827,
      "epoch": 0.26958456081284193,
      "flos": 24718140840960.0,
      "grad_norm": 2.030111149495139,
      "language_loss": 0.75194442,
      "learning_rate": 3.4276439660066963e-06,
      "loss": 0.77411485,
      "num_input_tokens_seen": 48128330,
      "step": 2242,
      "time_per_iteration": 2.587635040283203
    },
    {
      "auxiliary_loss_clip": 0.01199403,
      "auxiliary_loss_mlp": 0.01031044,
      "balance_loss_clip": 1.05970287,
      "balance_loss_mlp": 1.02209735,
      "epoch": 0.26970480370348104,
      "flos": 18112588606080.0,
      "grad_norm": 2.0028487577953835,
      "language_loss": 0.84297961,
      "learning_rate": 3.427098322707452e-06,
      "loss": 0.86528409,
      "num_input_tokens_seen": 48144295,
      "step": 2243,
      "time_per_iteration": 2.4801080226898193
    },
    {
      "auxiliary_loss_clip": 0.0118927,
      "auxiliary_loss_mlp": 0.0103394,
      "balance_loss_clip": 1.06410575,
      "balance_loss_mlp": 1.02439713,
      "epoch": 0.2698250465941201,
      "flos": 10816546250880.0,
      "grad_norm": 1.9671372307385728,
      "language_loss": 0.89197153,
      "learning_rate": 3.426552462918526e-06,
      "loss": 0.91420358,
      "num_input_tokens_seen": 48162230,
      "step": 2244,
      "time_per_iteration": 2.593916654586792
    },
    {
      "auxiliary_loss_clip": 0.01202744,
      "auxiliary_loss_mlp": 0.0102988,
      "balance_loss_clip": 1.06268883,
      "balance_loss_mlp": 1.02182698,
      "epoch": 0.2699452894847592,
      "flos": 17308117653120.0,
      "grad_norm": 2.292324163488663,
      "language_loss": 0.73412573,
      "learning_rate": 3.426006386722726e-06,
      "loss": 0.75645196,
      "num_input_tokens_seen": 48180290,
      "step": 2245,
      "time_per_iteration": 2.4957385063171387
    },
    {
      "auxiliary_loss_clip": 0.01160077,
      "auxiliary_loss_mlp": 0.01037226,
      "balance_loss_clip": 1.05859447,
      "balance_loss_mlp": 1.02848828,
      "epoch": 0.2700655323753983,
      "flos": 18078150441600.0,
      "grad_norm": 1.9090827293427153,
      "language_loss": 0.92414212,
      "learning_rate": 3.4254600942028914e-06,
      "loss": 0.94611514,
      "num_input_tokens_seen": 48198165,
      "step": 2246,
      "time_per_iteration": 2.623569965362549
    },
    {
      "auxiliary_loss_clip": 0.01168978,
      "auxiliary_loss_mlp": 0.01031802,
      "balance_loss_clip": 1.05709541,
      "balance_loss_mlp": 1.02337432,
      "epoch": 0.2701857752660374,
      "flos": 18186636493440.0,
      "grad_norm": 10.830067433011967,
      "language_loss": 0.82037365,
      "learning_rate": 3.424913585441893e-06,
      "loss": 0.84238148,
      "num_input_tokens_seen": 48216000,
      "step": 2247,
      "time_per_iteration": 2.5540707111358643
    },
    {
      "auxiliary_loss_clip": 0.01183803,
      "auxiliary_loss_mlp": 0.01030401,
      "balance_loss_clip": 1.0598104,
      "balance_loss_mlp": 1.02165675,
      "epoch": 0.2703060181566765,
      "flos": 16319496648960.0,
      "grad_norm": 1.9830612677353634,
      "language_loss": 0.87202406,
      "learning_rate": 3.4243668605226374e-06,
      "loss": 0.89416611,
      "num_input_tokens_seen": 48233025,
      "step": 2248,
      "time_per_iteration": 2.516472101211548
    },
    {
      "auxiliary_loss_clip": 0.01157501,
      "auxiliary_loss_mlp": 0.01033893,
      "balance_loss_clip": 1.05503964,
      "balance_loss_mlp": 1.02526188,
      "epoch": 0.2704262610473156,
      "flos": 19572357329280.0,
      "grad_norm": 2.227013617892131,
      "language_loss": 0.82716298,
      "learning_rate": 3.423819919528061e-06,
      "loss": 0.84907687,
      "num_input_tokens_seen": 48251110,
      "step": 2249,
      "time_per_iteration": 2.5871667861938477
    },
    {
      "auxiliary_loss_clip": 0.01149035,
      "auxiliary_loss_mlp": 0.01029623,
      "balance_loss_clip": 1.05148458,
      "balance_loss_mlp": 1.02018166,
      "epoch": 0.27054650393795465,
      "flos": 20740746925440.0,
      "grad_norm": 1.7484871819685195,
      "language_loss": 0.77721608,
      "learning_rate": 3.4232727625411355e-06,
      "loss": 0.79900265,
      "num_input_tokens_seen": 48270215,
      "step": 2250,
      "time_per_iteration": 2.6701366901397705
    },
    {
      "auxiliary_loss_clip": 0.01118308,
      "auxiliary_loss_mlp": 0.01024662,
      "balance_loss_clip": 1.04822075,
      "balance_loss_mlp": 1.0167346,
      "epoch": 0.27066674682859376,
      "flos": 18658322916480.0,
      "grad_norm": 1.7222932146497025,
      "language_loss": 0.86081409,
      "learning_rate": 3.4227253896448626e-06,
      "loss": 0.88224375,
      "num_input_tokens_seen": 48288075,
      "step": 2251,
      "time_per_iteration": 2.643840789794922
    },
    {
      "auxiliary_loss_clip": 0.01202671,
      "auxiliary_loss_mlp": 0.010317,
      "balance_loss_clip": 1.06053793,
      "balance_loss_mlp": 1.02314711,
      "epoch": 0.2707869897192329,
      "flos": 23002759958400.0,
      "grad_norm": 3.7543008295539306,
      "language_loss": 0.82018793,
      "learning_rate": 3.42217780092228e-06,
      "loss": 0.84253162,
      "num_input_tokens_seen": 48306415,
      "step": 2252,
      "time_per_iteration": 2.5433032512664795
    },
    {
      "auxiliary_loss_clip": 0.01091971,
      "auxiliary_loss_mlp": 0.01003335,
      "balance_loss_clip": 1.04173684,
      "balance_loss_mlp": 1.00177336,
      "epoch": 0.27090723260987193,
      "flos": 58323240293760.0,
      "grad_norm": 0.7879095360478465,
      "language_loss": 0.60360694,
      "learning_rate": 3.421629996456456e-06,
      "loss": 0.62456,
      "num_input_tokens_seen": 48365035,
      "step": 2253,
      "time_per_iteration": 3.0753674507141113
    },
    {
      "auxiliary_loss_clip": 0.01185307,
      "auxiliary_loss_mlp": 0.01032312,
      "balance_loss_clip": 1.05666614,
      "balance_loss_mlp": 1.02256703,
      "epoch": 0.27102747550051104,
      "flos": 11984540797440.0,
      "grad_norm": 2.0454470776182334,
      "language_loss": 0.82128537,
      "learning_rate": 3.421081976330491e-06,
      "loss": 0.84346157,
      "num_input_tokens_seen": 48383550,
      "step": 2254,
      "time_per_iteration": 2.5213615894317627
    },
    {
      "auxiliary_loss_clip": 0.01166618,
      "auxiliary_loss_mlp": 0.010338,
      "balance_loss_clip": 1.05304694,
      "balance_loss_mlp": 1.02470458,
      "epoch": 0.27114771839115015,
      "flos": 19900401264000.0,
      "grad_norm": 1.8520133018526586,
      "language_loss": 0.87950951,
      "learning_rate": 3.4205337406275207e-06,
      "loss": 0.9015137,
      "num_input_tokens_seen": 48403670,
      "step": 2255,
      "time_per_iteration": 2.5624430179595947
    },
    {
      "auxiliary_loss_clip": 0.01199932,
      "auxiliary_loss_mlp": 0.01027433,
      "balance_loss_clip": 1.05889618,
      "balance_loss_mlp": 1.01900458,
      "epoch": 0.2712679612817892,
      "flos": 18331966920960.0,
      "grad_norm": 2.2961845989349463,
      "language_loss": 0.75398445,
      "learning_rate": 3.4199852894307114e-06,
      "loss": 0.77625811,
      "num_input_tokens_seen": 48420420,
      "step": 2256,
      "time_per_iteration": 2.5139122009277344
    },
    {
      "auxiliary_loss_clip": 0.01131478,
      "auxiliary_loss_mlp": 0.01033429,
      "balance_loss_clip": 1.05274034,
      "balance_loss_mlp": 1.02495885,
      "epoch": 0.2713882041724283,
      "flos": 24460302038400.0,
      "grad_norm": 28.465531865149284,
      "language_loss": 0.78757352,
      "learning_rate": 3.419436622823262e-06,
      "loss": 0.80922258,
      "num_input_tokens_seen": 48441140,
      "step": 2257,
      "time_per_iteration": 2.691471815109253
    },
    {
      "auxiliary_loss_clip": 0.01172077,
      "auxiliary_loss_mlp": 0.01030746,
      "balance_loss_clip": 1.05842304,
      "balance_loss_mlp": 1.02228808,
      "epoch": 0.27150844706306737,
      "flos": 23039317025280.0,
      "grad_norm": 1.7342160261801571,
      "language_loss": 0.74526668,
      "learning_rate": 3.4188877408884063e-06,
      "loss": 0.76729488,
      "num_input_tokens_seen": 48461845,
      "step": 2258,
      "time_per_iteration": 3.4508912563323975
    },
    {
      "auxiliary_loss_clip": 0.01170132,
      "auxiliary_loss_mlp": 0.01038157,
      "balance_loss_clip": 1.05688334,
      "balance_loss_mlp": 1.0286262,
      "epoch": 0.2716286899537065,
      "flos": 22563644192640.0,
      "grad_norm": 2.5550507144329897,
      "language_loss": 0.65720934,
      "learning_rate": 3.4183386437094088e-06,
      "loss": 0.6792922,
      "num_input_tokens_seen": 48478510,
      "step": 2259,
      "time_per_iteration": 2.5705761909484863
    },
    {
      "auxiliary_loss_clip": 0.01173578,
      "auxiliary_loss_mlp": 0.01026785,
      "balance_loss_clip": 1.05497336,
      "balance_loss_mlp": 1.01840472,
      "epoch": 0.2717489328443456,
      "flos": 13115044523520.0,
      "grad_norm": 2.746135128200953,
      "language_loss": 0.81819636,
      "learning_rate": 3.417789331369565e-06,
      "loss": 0.84020001,
      "num_input_tokens_seen": 48494300,
      "step": 2260,
      "time_per_iteration": 2.6228742599487305
    },
    {
      "auxiliary_loss_clip": 0.01205123,
      "auxiliary_loss_mlp": 0.01032751,
      "balance_loss_clip": 1.06133878,
      "balance_loss_mlp": 1.02354169,
      "epoch": 0.27186917573498465,
      "flos": 29278688060160.0,
      "grad_norm": 2.565695001007483,
      "language_loss": 0.90917522,
      "learning_rate": 3.4172398039522088e-06,
      "loss": 0.93155396,
      "num_input_tokens_seen": 48515585,
      "step": 2261,
      "time_per_iteration": 2.5726802349090576
    },
    {
      "auxiliary_loss_clip": 0.01186335,
      "auxiliary_loss_mlp": 0.01025465,
      "balance_loss_clip": 1.05688477,
      "balance_loss_mlp": 1.01679265,
      "epoch": 0.27198941862562376,
      "flos": 26032220000640.0,
      "grad_norm": 1.7100227764244185,
      "language_loss": 0.80055898,
      "learning_rate": 3.4166900615407e-06,
      "loss": 0.82267702,
      "num_input_tokens_seen": 48533500,
      "step": 2262,
      "time_per_iteration": 2.561304807662964
    },
    {
      "auxiliary_loss_clip": 0.01185469,
      "auxiliary_loss_mlp": 0.0102862,
      "balance_loss_clip": 1.05736971,
      "balance_loss_mlp": 1.01979208,
      "epoch": 0.27210966151626287,
      "flos": 32780983760640.0,
      "grad_norm": 2.514317314131098,
      "language_loss": 0.75424999,
      "learning_rate": 3.416140104218436e-06,
      "loss": 0.77639091,
      "num_input_tokens_seen": 48552865,
      "step": 2263,
      "time_per_iteration": 5.107882499694824
    },
    {
      "auxiliary_loss_clip": 0.01086179,
      "auxiliary_loss_mlp": 0.00754045,
      "balance_loss_clip": 1.03053486,
      "balance_loss_mlp": 1.00092733,
      "epoch": 0.2722299044069019,
      "flos": 65471043219840.0,
      "grad_norm": 0.842769029650522,
      "language_loss": 0.69652462,
      "learning_rate": 3.4155899320688437e-06,
      "loss": 0.71492684,
      "num_input_tokens_seen": 48618940,
      "step": 2264,
      "time_per_iteration": 3.1810214519500732
    },
    {
      "auxiliary_loss_clip": 0.0113024,
      "auxiliary_loss_mlp": 0.01027132,
      "balance_loss_clip": 1.05238247,
      "balance_loss_mlp": 1.01750028,
      "epoch": 0.27235014729754103,
      "flos": 15334143782400.0,
      "grad_norm": 2.05843568013521,
      "language_loss": 0.73922384,
      "learning_rate": 3.415039545175384e-06,
      "loss": 0.76079756,
      "num_input_tokens_seen": 48634665,
      "step": 2265,
      "time_per_iteration": 2.638352870941162
    },
    {
      "auxiliary_loss_clip": 0.01187859,
      "auxiliary_loss_mlp": 0.01029153,
      "balance_loss_clip": 1.05834961,
      "balance_loss_mlp": 1.02055252,
      "epoch": 0.27247039018818014,
      "flos": 21872363973120.0,
      "grad_norm": 2.0995664252394306,
      "language_loss": 0.65288234,
      "learning_rate": 3.414488943621551e-06,
      "loss": 0.67505252,
      "num_input_tokens_seen": 48653330,
      "step": 2266,
      "time_per_iteration": 2.574570894241333
    },
    {
      "auxiliary_loss_clip": 0.01182684,
      "auxiliary_loss_mlp": 0.01031598,
      "balance_loss_clip": 1.05724061,
      "balance_loss_mlp": 1.02276421,
      "epoch": 0.2725906330788192,
      "flos": 18695490514560.0,
      "grad_norm": 1.7483211306663822,
      "language_loss": 0.73786974,
      "learning_rate": 3.41393812749087e-06,
      "loss": 0.76001251,
      "num_input_tokens_seen": 48671375,
      "step": 2267,
      "time_per_iteration": 2.517695903778076
    },
    {
      "auxiliary_loss_clip": 0.01168562,
      "auxiliary_loss_mlp": 0.01030763,
      "balance_loss_clip": 1.05664778,
      "balance_loss_mlp": 1.02169752,
      "epoch": 0.2727108759694583,
      "flos": 17886099398400.0,
      "grad_norm": 2.3245715058488194,
      "language_loss": 0.71956003,
      "learning_rate": 3.4133870968668984e-06,
      "loss": 0.74155325,
      "num_input_tokens_seen": 48686175,
      "step": 2268,
      "time_per_iteration": 2.5397541522979736
    },
    {
      "auxiliary_loss_clip": 0.01174338,
      "auxiliary_loss_mlp": 0.01029815,
      "balance_loss_clip": 1.05736113,
      "balance_loss_mlp": 1.02110076,
      "epoch": 0.2728311188600974,
      "flos": 24461666755200.0,
      "grad_norm": 1.6418234099784983,
      "language_loss": 0.78672612,
      "learning_rate": 3.412835851833229e-06,
      "loss": 0.80876768,
      "num_input_tokens_seen": 48708370,
      "step": 2269,
      "time_per_iteration": 2.7194840908050537
    },
    {
      "auxiliary_loss_clip": 0.01184135,
      "auxiliary_loss_mlp": 0.0103047,
      "balance_loss_clip": 1.06002295,
      "balance_loss_mlp": 1.02146947,
      "epoch": 0.2729513617507365,
      "flos": 30993314757120.0,
      "grad_norm": 1.7233135631284064,
      "language_loss": 0.78083861,
      "learning_rate": 3.4122843924734834e-06,
      "loss": 0.8029846,
      "num_input_tokens_seen": 48730670,
      "step": 2270,
      "time_per_iteration": 2.667440891265869
    },
    {
      "auxiliary_loss_clip": 0.01167166,
      "auxiliary_loss_mlp": 0.01032901,
      "balance_loss_clip": 1.05530655,
      "balance_loss_mlp": 1.02380538,
      "epoch": 0.2730716046413756,
      "flos": 19094637421440.0,
      "grad_norm": 1.8046767893824014,
      "language_loss": 0.87522066,
      "learning_rate": 3.411732718871319e-06,
      "loss": 0.89722133,
      "num_input_tokens_seen": 48746510,
      "step": 2271,
      "time_per_iteration": 2.5938398838043213
    },
    {
      "auxiliary_loss_clip": 0.01198006,
      "auxiliary_loss_mlp": 0.01030446,
      "balance_loss_clip": 1.06020331,
      "balance_loss_mlp": 1.02235723,
      "epoch": 0.27319184753201464,
      "flos": 26944566474240.0,
      "grad_norm": 1.6100913658973959,
      "language_loss": 0.78764844,
      "learning_rate": 3.4111808311104227e-06,
      "loss": 0.80993295,
      "num_input_tokens_seen": 48768825,
      "step": 2272,
      "time_per_iteration": 2.6117727756500244
    },
    {
      "auxiliary_loss_clip": 0.01176248,
      "auxiliary_loss_mlp": 0.01030141,
      "balance_loss_clip": 1.05379152,
      "balance_loss_mlp": 1.02077699,
      "epoch": 0.27331209042265375,
      "flos": 31759828012800.0,
      "grad_norm": 2.075599805796371,
      "language_loss": 0.69473076,
      "learning_rate": 3.410628729274517e-06,
      "loss": 0.71679461,
      "num_input_tokens_seen": 48790345,
      "step": 2273,
      "time_per_iteration": 2.641658306121826
    },
    {
      "auxiliary_loss_clip": 0.01165904,
      "auxiliary_loss_mlp": 0.00763933,
      "balance_loss_clip": 1.05490375,
      "balance_loss_mlp": 1.00111079,
      "epoch": 0.27343233331329286,
      "flos": 25739081107200.0,
      "grad_norm": 1.8633737130588401,
      "language_loss": 0.8258971,
      "learning_rate": 3.4100764134473546e-06,
      "loss": 0.84519553,
      "num_input_tokens_seen": 48809630,
      "step": 2274,
      "time_per_iteration": 2.6776325702667236
    },
    {
      "auxiliary_loss_clip": 0.01200349,
      "auxiliary_loss_mlp": 0.01031464,
      "balance_loss_clip": 1.06106305,
      "balance_loss_mlp": 1.02310181,
      "epoch": 0.2735525762039319,
      "flos": 24389414547840.0,
      "grad_norm": 2.1252814747568247,
      "language_loss": 0.84735936,
      "learning_rate": 3.4095238837127215e-06,
      "loss": 0.86967742,
      "num_input_tokens_seen": 48828770,
      "step": 2275,
      "time_per_iteration": 2.5401053428649902
    },
    {
      "auxiliary_loss_clip": 0.01152907,
      "auxiliary_loss_mlp": 0.01024967,
      "balance_loss_clip": 1.05246913,
      "balance_loss_mlp": 1.0162468,
      "epoch": 0.27367281909457103,
      "flos": 14465357527680.0,
      "grad_norm": 1.8405068988020015,
      "language_loss": 0.7926175,
      "learning_rate": 3.4089711401544355e-06,
      "loss": 0.81439626,
      "num_input_tokens_seen": 48846365,
      "step": 2276,
      "time_per_iteration": 2.635746717453003
    },
    {
      "auxiliary_loss_clip": 0.01182115,
      "auxiliary_loss_mlp": 0.01024031,
      "balance_loss_clip": 1.05336463,
      "balance_loss_mlp": 1.01560259,
      "epoch": 0.27379306198521014,
      "flos": 23476996247040.0,
      "grad_norm": 2.5395313500270675,
      "language_loss": 0.67487431,
      "learning_rate": 3.4084181828563486e-06,
      "loss": 0.69693577,
      "num_input_tokens_seen": 48863085,
      "step": 2277,
      "time_per_iteration": 2.5924232006073
    },
    {
      "auxiliary_loss_clip": 0.01142037,
      "auxiliary_loss_mlp": 0.01026488,
      "balance_loss_clip": 1.05180907,
      "balance_loss_mlp": 1.01777411,
      "epoch": 0.2739133048758492,
      "flos": 17458152762240.0,
      "grad_norm": 1.63872546872183,
      "language_loss": 0.70435667,
      "learning_rate": 3.4078650119023428e-06,
      "loss": 0.72604191,
      "num_input_tokens_seen": 48881400,
      "step": 2278,
      "time_per_iteration": 2.6369099617004395
    },
    {
      "auxiliary_loss_clip": 0.01129631,
      "auxiliary_loss_mlp": 0.01033129,
      "balance_loss_clip": 1.04789996,
      "balance_loss_mlp": 1.02339005,
      "epoch": 0.2740335477664883,
      "flos": 19273113123840.0,
      "grad_norm": 2.3591697923356945,
      "language_loss": 0.74473888,
      "learning_rate": 3.4073116273763337e-06,
      "loss": 0.76636648,
      "num_input_tokens_seen": 48895845,
      "step": 2279,
      "time_per_iteration": 2.6410837173461914
    },
    {
      "auxiliary_loss_clip": 0.01175502,
      "auxiliary_loss_mlp": 0.01033138,
      "balance_loss_clip": 1.05449617,
      "balance_loss_mlp": 1.02364254,
      "epoch": 0.2741537906571274,
      "flos": 26104723603200.0,
      "grad_norm": 1.7409045269684267,
      "language_loss": 0.8121416,
      "learning_rate": 3.40675802936227e-06,
      "loss": 0.83422792,
      "num_input_tokens_seen": 48916630,
      "step": 2280,
      "time_per_iteration": 2.618314743041992
    },
    {
      "auxiliary_loss_clip": 0.01166766,
      "auxiliary_loss_mlp": 0.01037257,
      "balance_loss_clip": 1.0579977,
      "balance_loss_mlp": 1.02746391,
      "epoch": 0.27427403354776647,
      "flos": 34164190644480.0,
      "grad_norm": 2.076216025342256,
      "language_loss": 0.7152608,
      "learning_rate": 3.4062042179441318e-06,
      "loss": 0.73730099,
      "num_input_tokens_seen": 48937100,
      "step": 2281,
      "time_per_iteration": 2.716334581375122
    },
    {
      "auxiliary_loss_clip": 0.01182867,
      "auxiliary_loss_mlp": 0.0102598,
      "balance_loss_clip": 1.0586555,
      "balance_loss_mlp": 1.01770687,
      "epoch": 0.2743942764384056,
      "flos": 18766988536320.0,
      "grad_norm": 2.4691139178697217,
      "language_loss": 0.80563724,
      "learning_rate": 3.4056501932059314e-06,
      "loss": 0.82772565,
      "num_input_tokens_seen": 48955175,
      "step": 2282,
      "time_per_iteration": 2.580759048461914
    },
    {
      "auxiliary_loss_clip": 0.01110148,
      "auxiliary_loss_mlp": 0.01003243,
      "balance_loss_clip": 1.03339624,
      "balance_loss_mlp": 1.00190163,
      "epoch": 0.2745145193290447,
      "flos": 64904048058240.0,
      "grad_norm": 0.7666347693249996,
      "language_loss": 0.58153772,
      "learning_rate": 3.405095955231715e-06,
      "loss": 0.60267162,
      "num_input_tokens_seen": 49006830,
      "step": 2283,
      "time_per_iteration": 3.0404679775238037
    },
    {
      "auxiliary_loss_clip": 0.01189729,
      "auxiliary_loss_mlp": 0.01027791,
      "balance_loss_clip": 1.05763578,
      "balance_loss_mlp": 1.01886785,
      "epoch": 0.27463476221968375,
      "flos": 16136926796160.0,
      "grad_norm": 2.523924182004298,
      "language_loss": 0.94218481,
      "learning_rate": 3.4045415041055585e-06,
      "loss": 0.96436006,
      "num_input_tokens_seen": 49022470,
      "step": 2284,
      "time_per_iteration": 2.5761613845825195
    },
    {
      "auxiliary_loss_clip": 0.01176194,
      "auxiliary_loss_mlp": 0.010301,
      "balance_loss_clip": 1.05835891,
      "balance_loss_mlp": 1.02065849,
      "epoch": 0.27475500511032286,
      "flos": 10376712213120.0,
      "grad_norm": 2.1045137554009656,
      "language_loss": 0.78368366,
      "learning_rate": 3.4039868399115728e-06,
      "loss": 0.80574656,
      "num_input_tokens_seen": 49037110,
      "step": 2285,
      "time_per_iteration": 3.416395425796509
    },
    {
      "auxiliary_loss_clip": 0.01136074,
      "auxiliary_loss_mlp": 0.01029949,
      "balance_loss_clip": 1.05695748,
      "balance_loss_mlp": 1.02121651,
      "epoch": 0.27487524800096197,
      "flos": 17311062568320.0,
      "grad_norm": 1.7549044070174171,
      "language_loss": 0.80335116,
      "learning_rate": 3.4034319627339003e-06,
      "loss": 0.82501137,
      "num_input_tokens_seen": 49053975,
      "step": 2286,
      "time_per_iteration": 2.652266025543213
    },
    {
      "auxiliary_loss_clip": 0.01175356,
      "auxiliary_loss_mlp": 0.01035137,
      "balance_loss_clip": 1.0593164,
      "balance_loss_mlp": 1.02606475,
      "epoch": 0.274995490891601,
      "flos": 27120205002240.0,
      "grad_norm": 2.7455388767119713,
      "language_loss": 0.69592637,
      "learning_rate": 3.402876872656715e-06,
      "loss": 0.71803129,
      "num_input_tokens_seen": 49072295,
      "step": 2287,
      "time_per_iteration": 2.626197099685669
    },
    {
      "auxiliary_loss_clip": 0.01171219,
      "auxiliary_loss_mlp": 0.01034554,
      "balance_loss_clip": 1.05758548,
      "balance_loss_mlp": 1.02590561,
      "epoch": 0.27511573378224013,
      "flos": 23436093634560.0,
      "grad_norm": 7.9369149199794355,
      "language_loss": 0.89517951,
      "learning_rate": 3.402321569764223e-06,
      "loss": 0.91723728,
      "num_input_tokens_seen": 49091600,
      "step": 2288,
      "time_per_iteration": 2.643428325653076
    },
    {
      "auxiliary_loss_clip": 0.01150496,
      "auxiliary_loss_mlp": 0.00764772,
      "balance_loss_clip": 1.05458713,
      "balance_loss_mlp": 1.00131893,
      "epoch": 0.2752359766728792,
      "flos": 16722019434240.0,
      "grad_norm": 3.2711832734459536,
      "language_loss": 0.83556664,
      "learning_rate": 3.4017660541406635e-06,
      "loss": 0.8547194,
      "num_input_tokens_seen": 49107665,
      "step": 2289,
      "time_per_iteration": 3.530979871749878
    },
    {
      "auxiliary_loss_clip": 0.0118377,
      "auxiliary_loss_mlp": 0.01028378,
      "balance_loss_clip": 1.05886519,
      "balance_loss_mlp": 1.01952124,
      "epoch": 0.2753562195635183,
      "flos": 25297738698240.0,
      "grad_norm": 1.7170772357239443,
      "language_loss": 0.74407005,
      "learning_rate": 3.4012103258703092e-06,
      "loss": 0.7661916,
      "num_input_tokens_seen": 49126420,
      "step": 2290,
      "time_per_iteration": 4.260874509811401
    },
    {
      "auxiliary_loss_clip": 0.01157504,
      "auxiliary_loss_mlp": 0.01023131,
      "balance_loss_clip": 1.05454397,
      "balance_loss_mlp": 1.01430392,
      "epoch": 0.2754764624541574,
      "flos": 27338972785920.0,
      "grad_norm": 2.0528263521778687,
      "language_loss": 0.82945174,
      "learning_rate": 3.4006543850374616e-06,
      "loss": 0.85125804,
      "num_input_tokens_seen": 49141470,
      "step": 2291,
      "time_per_iteration": 2.617689371109009
    },
    {
      "auxiliary_loss_clip": 0.01189104,
      "auxiliary_loss_mlp": 0.01033238,
      "balance_loss_clip": 1.05748844,
      "balance_loss_mlp": 1.02490497,
      "epoch": 0.27559670534479647,
      "flos": 17238379397760.0,
      "grad_norm": 2.0068094901254887,
      "language_loss": 0.74810296,
      "learning_rate": 3.400098231726458e-06,
      "loss": 0.77032638,
      "num_input_tokens_seen": 49158570,
      "step": 2292,
      "time_per_iteration": 2.4912991523742676
    },
    {
      "auxiliary_loss_clip": 0.01162393,
      "auxiliary_loss_mlp": 0.01036679,
      "balance_loss_clip": 1.0534339,
      "balance_loss_mlp": 1.02729678,
      "epoch": 0.2757169482354356,
      "flos": 21939085486080.0,
      "grad_norm": 1.8171196006707633,
      "language_loss": 0.87356663,
      "learning_rate": 3.3995418660216657e-06,
      "loss": 0.89555728,
      "num_input_tokens_seen": 49176025,
      "step": 2293,
      "time_per_iteration": 2.6027116775512695
    },
    {
      "auxiliary_loss_clip": 0.01208866,
      "auxiliary_loss_mlp": 0.01034035,
      "balance_loss_clip": 1.06288445,
      "balance_loss_mlp": 1.02504635,
      "epoch": 0.2758371911260747,
      "flos": 20850669521280.0,
      "grad_norm": 2.4482372769351284,
      "language_loss": 0.80623233,
      "learning_rate": 3.3989852880074848e-06,
      "loss": 0.82866138,
      "num_input_tokens_seen": 49197455,
      "step": 2294,
      "time_per_iteration": 2.487154245376587
    },
    {
      "auxiliary_loss_clip": 0.01089437,
      "auxiliary_loss_mlp": 0.01003293,
      "balance_loss_clip": 1.0398109,
      "balance_loss_mlp": 1.00113487,
      "epoch": 0.27595743401671374,
      "flos": 69269063592960.0,
      "grad_norm": 0.7411011988427074,
      "language_loss": 0.60670304,
      "learning_rate": 3.398428497768348e-06,
      "loss": 0.62763035,
      "num_input_tokens_seen": 49262625,
      "step": 2295,
      "time_per_iteration": 3.242403268814087
    },
    {
      "auxiliary_loss_clip": 0.01168106,
      "auxiliary_loss_mlp": 0.01028707,
      "balance_loss_clip": 1.05554807,
      "balance_loss_mlp": 1.01948631,
      "epoch": 0.27607767690735285,
      "flos": 21215019127680.0,
      "grad_norm": 1.7996103019163994,
      "language_loss": 0.71940064,
      "learning_rate": 3.3978714953887205e-06,
      "loss": 0.74136877,
      "num_input_tokens_seen": 49282380,
      "step": 2296,
      "time_per_iteration": 2.6048669815063477
    },
    {
      "auxiliary_loss_clip": 0.01131485,
      "auxiliary_loss_mlp": 0.01029429,
      "balance_loss_clip": 1.04829657,
      "balance_loss_mlp": 1.0204345,
      "epoch": 0.27619791979799196,
      "flos": 24825334003200.0,
      "grad_norm": 1.9130129589227036,
      "language_loss": 0.85990274,
      "learning_rate": 3.397314280953098e-06,
      "loss": 0.88151181,
      "num_input_tokens_seen": 49303205,
      "step": 2297,
      "time_per_iteration": 2.638890027999878
    },
    {
      "auxiliary_loss_clip": 0.01164224,
      "auxiliary_loss_mlp": 0.0102636,
      "balance_loss_clip": 1.05412889,
      "balance_loss_mlp": 1.01782489,
      "epoch": 0.276318162688631,
      "flos": 24753548672640.0,
      "grad_norm": 3.319790641089275,
      "language_loss": 0.8011657,
      "learning_rate": 3.3967568545460108e-06,
      "loss": 0.8230716,
      "num_input_tokens_seen": 49322745,
      "step": 2298,
      "time_per_iteration": 2.556432008743286
    },
    {
      "auxiliary_loss_clip": 0.01187012,
      "auxiliary_loss_mlp": 0.01032504,
      "balance_loss_clip": 1.06077337,
      "balance_loss_mlp": 1.02340794,
      "epoch": 0.27643840557927013,
      "flos": 18150007599360.0,
      "grad_norm": 2.2343643399360884,
      "language_loss": 0.80108303,
      "learning_rate": 3.3961992162520185e-06,
      "loss": 0.82327819,
      "num_input_tokens_seen": 49341370,
      "step": 2299,
      "time_per_iteration": 2.5123610496520996
    },
    {
      "auxiliary_loss_clip": 0.01188966,
      "auxiliary_loss_mlp": 0.01031562,
      "balance_loss_clip": 1.05956244,
      "balance_loss_mlp": 1.02217984,
      "epoch": 0.27655864846990924,
      "flos": 24823933372800.0,
      "grad_norm": 2.319248446740071,
      "language_loss": 0.71698606,
      "learning_rate": 3.3956413661557156e-06,
      "loss": 0.73919129,
      "num_input_tokens_seen": 49361545,
      "step": 2300,
      "time_per_iteration": 2.5811564922332764
    },
    {
      "auxiliary_loss_clip": 0.01165889,
      "auxiliary_loss_mlp": 0.0103373,
      "balance_loss_clip": 1.05479372,
      "balance_loss_mlp": 1.02438974,
      "epoch": 0.2766788913605483,
      "flos": 20266582464000.0,
      "grad_norm": 2.245048520138294,
      "language_loss": 0.66343451,
      "learning_rate": 3.3950833043417273e-06,
      "loss": 0.68543071,
      "num_input_tokens_seen": 49379690,
      "step": 2301,
      "time_per_iteration": 2.6058032512664795
    },
    {
      "auxiliary_loss_clip": 0.01192258,
      "auxiliary_loss_mlp": 0.01028703,
      "balance_loss_clip": 1.06261659,
      "balance_loss_mlp": 1.0188148,
      "epoch": 0.2767991342511874,
      "flos": 21470272151040.0,
      "grad_norm": 2.1459483031713105,
      "language_loss": 0.72972083,
      "learning_rate": 3.3945250308947105e-06,
      "loss": 0.75193042,
      "num_input_tokens_seen": 49395995,
      "step": 2302,
      "time_per_iteration": 2.5117101669311523
    },
    {
      "auxiliary_loss_clip": 0.01100616,
      "auxiliary_loss_mlp": 0.01010573,
      "balance_loss_clip": 1.03200841,
      "balance_loss_mlp": 1.00903499,
      "epoch": 0.2769193771418265,
      "flos": 66002627571840.0,
      "grad_norm": 1.2516432673174016,
      "language_loss": 0.68342334,
      "learning_rate": 3.3939665458993556e-06,
      "loss": 0.70453525,
      "num_input_tokens_seen": 49450415,
      "step": 2303,
      "time_per_iteration": 3.0422446727752686
    },
    {
      "auxiliary_loss_clip": 0.01163331,
      "auxiliary_loss_mlp": 0.00764806,
      "balance_loss_clip": 1.05333459,
      "balance_loss_mlp": 1.00124931,
      "epoch": 0.27703962003246557,
      "flos": 20704441253760.0,
      "grad_norm": 1.8284182175453665,
      "language_loss": 0.76726037,
      "learning_rate": 3.3934078494403843e-06,
      "loss": 0.7865417,
      "num_input_tokens_seen": 49469990,
      "step": 2304,
      "time_per_iteration": 2.6103415489196777
    },
    {
      "auxiliary_loss_clip": 0.01109891,
      "auxiliary_loss_mlp": 0.01043208,
      "balance_loss_clip": 1.04834855,
      "balance_loss_mlp": 1.03272331,
      "epoch": 0.2771598629231047,
      "flos": 22929897219840.0,
      "grad_norm": 2.4626805714158357,
      "language_loss": 0.81398809,
      "learning_rate": 3.3928489416025495e-06,
      "loss": 0.83551908,
      "num_input_tokens_seen": 49490835,
      "step": 2305,
      "time_per_iteration": 2.688297748565674
    },
    {
      "auxiliary_loss_clip": 0.01171536,
      "auxiliary_loss_mlp": 0.0104041,
      "balance_loss_clip": 1.05652678,
      "balance_loss_mlp": 1.03034282,
      "epoch": 0.27728010581374374,
      "flos": 18369457741440.0,
      "grad_norm": 2.490454316898515,
      "language_loss": 0.78673679,
      "learning_rate": 3.392289822470638e-06,
      "loss": 0.80885625,
      "num_input_tokens_seen": 49508815,
      "step": 2306,
      "time_per_iteration": 2.5390889644622803
    },
    {
      "auxiliary_loss_clip": 0.01169412,
      "auxiliary_loss_mlp": 0.0102802,
      "balance_loss_clip": 1.05469537,
      "balance_loss_mlp": 1.01853752,
      "epoch": 0.27740034870438285,
      "flos": 19427637432960.0,
      "grad_norm": 1.9863388544497083,
      "language_loss": 0.75672752,
      "learning_rate": 3.3917304921294674e-06,
      "loss": 0.77870184,
      "num_input_tokens_seen": 49526980,
      "step": 2307,
      "time_per_iteration": 2.6404366493225098
    },
    {
      "auxiliary_loss_clip": 0.0118897,
      "auxiliary_loss_mlp": 0.01033904,
      "balance_loss_clip": 1.05825448,
      "balance_loss_mlp": 1.02432537,
      "epoch": 0.27752059159502196,
      "flos": 21614776565760.0,
      "grad_norm": 1.6643399870341387,
      "language_loss": 0.8058055,
      "learning_rate": 3.3911709506638876e-06,
      "loss": 0.82803428,
      "num_input_tokens_seen": 49546290,
      "step": 2308,
      "time_per_iteration": 2.532336711883545
    },
    {
      "auxiliary_loss_clip": 0.01147685,
      "auxiliary_loss_mlp": 0.00764965,
      "balance_loss_clip": 1.0495584,
      "balance_loss_mlp": 1.00126362,
      "epoch": 0.277640834485661,
      "flos": 26608011016320.0,
      "grad_norm": 2.2848239249049964,
      "language_loss": 0.81060874,
      "learning_rate": 3.390611198158781e-06,
      "loss": 0.82973528,
      "num_input_tokens_seen": 49564165,
      "step": 2309,
      "time_per_iteration": 2.6282243728637695
    },
    {
      "auxiliary_loss_clip": 0.01207549,
      "auxiliary_loss_mlp": 0.01034021,
      "balance_loss_clip": 1.06244457,
      "balance_loss_mlp": 1.02489567,
      "epoch": 0.2777610773763001,
      "flos": 19492814661120.0,
      "grad_norm": 2.154266483990461,
      "language_loss": 0.89846921,
      "learning_rate": 3.3900512346990612e-06,
      "loss": 0.92088485,
      "num_input_tokens_seen": 49580155,
      "step": 2310,
      "time_per_iteration": 2.4862213134765625
    },
    {
      "auxiliary_loss_clip": 0.01145768,
      "auxiliary_loss_mlp": 0.01036141,
      "balance_loss_clip": 1.04983544,
      "balance_loss_mlp": 1.02585864,
      "epoch": 0.27788132026693924,
      "flos": 38290650001920.0,
      "grad_norm": 1.9653351240014445,
      "language_loss": 0.65747881,
      "learning_rate": 3.389491060369674e-06,
      "loss": 0.67929792,
      "num_input_tokens_seen": 49605830,
      "step": 2311,
      "time_per_iteration": 3.580845832824707
    },
    {
      "auxiliary_loss_clip": 0.01135299,
      "auxiliary_loss_mlp": 0.01026795,
      "balance_loss_clip": 1.05042803,
      "balance_loss_mlp": 1.01775265,
      "epoch": 0.2780015631575783,
      "flos": 22382546797440.0,
      "grad_norm": 1.975646644909417,
      "language_loss": 0.8890394,
      "learning_rate": 3.388930675255598e-06,
      "loss": 0.91066039,
      "num_input_tokens_seen": 49625680,
      "step": 2312,
      "time_per_iteration": 2.6129791736602783
    },
    {
      "auxiliary_loss_clip": 0.01180119,
      "auxiliary_loss_mlp": 0.01033888,
      "balance_loss_clip": 1.05840492,
      "balance_loss_mlp": 1.02352226,
      "epoch": 0.2781218060482174,
      "flos": 12203200840320.0,
      "grad_norm": 2.493920053750425,
      "language_loss": 0.79525542,
      "learning_rate": 3.388370079441843e-06,
      "loss": 0.81739551,
      "num_input_tokens_seen": 49641195,
      "step": 2313,
      "time_per_iteration": 2.551487684249878
    },
    {
      "auxiliary_loss_clip": 0.01162172,
      "auxiliary_loss_mlp": 0.0103543,
      "balance_loss_clip": 1.05983114,
      "balance_loss_mlp": 1.02633989,
      "epoch": 0.2782420489388565,
      "flos": 18107632529280.0,
      "grad_norm": 2.2083455780340984,
      "language_loss": 0.92839342,
      "learning_rate": 3.3878092730134505e-06,
      "loss": 0.95036948,
      "num_input_tokens_seen": 49659180,
      "step": 2314,
      "time_per_iteration": 2.5407063961029053
    },
    {
      "auxiliary_loss_clip": 0.01180894,
      "auxiliary_loss_mlp": 0.01035914,
      "balance_loss_clip": 1.05730247,
      "balance_loss_mlp": 1.02655625,
      "epoch": 0.27836229182949557,
      "flos": 18514752255360.0,
      "grad_norm": 2.309529374740936,
      "language_loss": 0.80695623,
      "learning_rate": 3.3872482560554947e-06,
      "loss": 0.82912433,
      "num_input_tokens_seen": 49677955,
      "step": 2315,
      "time_per_iteration": 2.4814672470092773
    },
    {
      "auxiliary_loss_clip": 0.0109702,
      "auxiliary_loss_mlp": 0.01001625,
      "balance_loss_clip": 1.02963758,
      "balance_loss_mlp": 1.0001471,
      "epoch": 0.2784825347201347,
      "flos": 67079230940160.0,
      "grad_norm": 0.7948957031337394,
      "language_loss": 0.57035553,
      "learning_rate": 3.386687028653082e-06,
      "loss": 0.59134197,
      "num_input_tokens_seen": 49740800,
      "step": 2316,
      "time_per_iteration": 4.028663396835327
    },
    {
      "auxiliary_loss_clip": 0.01146956,
      "auxiliary_loss_mlp": 0.01030347,
      "balance_loss_clip": 1.05576205,
      "balance_loss_mlp": 1.02076316,
      "epoch": 0.2786027776107738,
      "flos": 22631119891200.0,
      "grad_norm": 1.9196492220412786,
      "language_loss": 0.84998566,
      "learning_rate": 3.386125590891349e-06,
      "loss": 0.8717587,
      "num_input_tokens_seen": 49757675,
      "step": 2317,
      "time_per_iteration": 3.359004259109497
    },
    {
      "auxiliary_loss_clip": 0.01161774,
      "auxiliary_loss_mlp": 0.01028775,
      "balance_loss_clip": 1.05310535,
      "balance_loss_mlp": 1.01991177,
      "epoch": 0.27872302050141284,
      "flos": 15778826156160.0,
      "grad_norm": 2.513360774218176,
      "language_loss": 0.82911432,
      "learning_rate": 3.3855639428554657e-06,
      "loss": 0.85101986,
      "num_input_tokens_seen": 49775205,
      "step": 2318,
      "time_per_iteration": 2.5261754989624023
    },
    {
      "auxiliary_loss_clip": 0.01148919,
      "auxiliary_loss_mlp": 0.01028438,
      "balance_loss_clip": 1.05549622,
      "balance_loss_mlp": 1.01964068,
      "epoch": 0.27884326339205195,
      "flos": 22126970551680.0,
      "grad_norm": 1.8332929501310893,
      "language_loss": 0.80668819,
      "learning_rate": 3.385002084630635e-06,
      "loss": 0.82846177,
      "num_input_tokens_seen": 49794175,
      "step": 2319,
      "time_per_iteration": 2.569084405899048
    },
    {
      "auxiliary_loss_clip": 0.01196523,
      "auxiliary_loss_mlp": 0.01034708,
      "balance_loss_clip": 1.06220078,
      "balance_loss_mlp": 1.02451563,
      "epoch": 0.278963506282691,
      "flos": 20558715776640.0,
      "grad_norm": 2.149297710017262,
      "language_loss": 0.85028422,
      "learning_rate": 3.384440016302088e-06,
      "loss": 0.8725965,
      "num_input_tokens_seen": 49812850,
      "step": 2320,
      "time_per_iteration": 2.513162136077881
    },
    {
      "auxiliary_loss_clip": 0.01185125,
      "auxiliary_loss_mlp": 0.01034679,
      "balance_loss_clip": 1.05859601,
      "balance_loss_mlp": 1.0251838,
      "epoch": 0.2790837491733301,
      "flos": 21942928241280.0,
      "grad_norm": 2.5064229707667804,
      "language_loss": 0.62424409,
      "learning_rate": 3.3838777379550923e-06,
      "loss": 0.64644217,
      "num_input_tokens_seen": 49832295,
      "step": 2321,
      "time_per_iteration": 2.5065596103668213
    },
    {
      "auxiliary_loss_clip": 0.01179955,
      "auxiliary_loss_mlp": 0.01035653,
      "balance_loss_clip": 1.05970931,
      "balance_loss_mlp": 1.02632487,
      "epoch": 0.27920399206396923,
      "flos": 26286790665600.0,
      "grad_norm": 2.065703189611674,
      "language_loss": 0.78120005,
      "learning_rate": 3.383315249674944e-06,
      "loss": 0.80335611,
      "num_input_tokens_seen": 49850860,
      "step": 2322,
      "time_per_iteration": 2.5916476249694824
    },
    {
      "auxiliary_loss_clip": 0.01162897,
      "auxiliary_loss_mlp": 0.0103328,
      "balance_loss_clip": 1.05666399,
      "balance_loss_mlp": 1.0240953,
      "epoch": 0.2793242349546083,
      "flos": 25400981364480.0,
      "grad_norm": 2.454702743351978,
      "language_loss": 0.86042655,
      "learning_rate": 3.3827525515469715e-06,
      "loss": 0.88238835,
      "num_input_tokens_seen": 49865765,
      "step": 2323,
      "time_per_iteration": 2.6412618160247803
    },
    {
      "auxiliary_loss_clip": 0.01149503,
      "auxiliary_loss_mlp": 0.01040581,
      "balance_loss_clip": 1.05057168,
      "balance_loss_mlp": 1.03035247,
      "epoch": 0.2794444778452474,
      "flos": 20850346298880.0,
      "grad_norm": 2.6376881586176335,
      "language_loss": 0.71126264,
      "learning_rate": 3.3821896436565367e-06,
      "loss": 0.73316348,
      "num_input_tokens_seen": 49885425,
      "step": 2324,
      "time_per_iteration": 2.594083547592163
    },
    {
      "auxiliary_loss_clip": 0.01193476,
      "auxiliary_loss_mlp": 0.0103658,
      "balance_loss_clip": 1.06444263,
      "balance_loss_mlp": 1.0272342,
      "epoch": 0.2795647207358865,
      "flos": 21576244250880.0,
      "grad_norm": 1.7008364518968293,
      "language_loss": 0.7003355,
      "learning_rate": 3.381626526089032e-06,
      "loss": 0.72263604,
      "num_input_tokens_seen": 49904990,
      "step": 2325,
      "time_per_iteration": 2.5412697792053223
    },
    {
      "auxiliary_loss_clip": 0.01172224,
      "auxiliary_loss_mlp": 0.01029762,
      "balance_loss_clip": 1.0559392,
      "balance_loss_mlp": 1.020064,
      "epoch": 0.27968496362652556,
      "flos": 21471744608640.0,
      "grad_norm": 1.9893451231395027,
      "language_loss": 0.78937197,
      "learning_rate": 3.3810631989298815e-06,
      "loss": 0.81139183,
      "num_input_tokens_seen": 49924600,
      "step": 2326,
      "time_per_iteration": 2.608919858932495
    },
    {
      "auxiliary_loss_clip": 0.01156335,
      "auxiliary_loss_mlp": 0.0103312,
      "balance_loss_clip": 1.06092358,
      "balance_loss_mlp": 1.02234364,
      "epoch": 0.2798052065171647,
      "flos": 23258695340160.0,
      "grad_norm": 2.3114567806177178,
      "language_loss": 0.84328163,
      "learning_rate": 3.3804996622645423e-06,
      "loss": 0.86517608,
      "num_input_tokens_seen": 49942600,
      "step": 2327,
      "time_per_iteration": 2.665694236755371
    },
    {
      "auxiliary_loss_clip": 0.01205958,
      "auxiliary_loss_mlp": 0.0103056,
      "balance_loss_clip": 1.06268311,
      "balance_loss_mlp": 1.02151835,
      "epoch": 0.2799254494078038,
      "flos": 21539328048000.0,
      "grad_norm": 2.898145915437923,
      "language_loss": 0.89373565,
      "learning_rate": 3.3799359161785015e-06,
      "loss": 0.91610086,
      "num_input_tokens_seen": 49962250,
      "step": 2328,
      "time_per_iteration": 2.510000467300415
    },
    {
      "auxiliary_loss_clip": 0.0118616,
      "auxiliary_loss_mlp": 0.01034802,
      "balance_loss_clip": 1.05907881,
      "balance_loss_mlp": 1.02512813,
      "epoch": 0.28004569229844284,
      "flos": 26393912000640.0,
      "grad_norm": 1.6260675827277233,
      "language_loss": 0.85655361,
      "learning_rate": 3.3793719607572798e-06,
      "loss": 0.8787632,
      "num_input_tokens_seen": 49983215,
      "step": 2329,
      "time_per_iteration": 2.5816659927368164
    },
    {
      "auxiliary_loss_clip": 0.01157664,
      "auxiliary_loss_mlp": 0.0102981,
      "balance_loss_clip": 1.05366766,
      "balance_loss_mlp": 1.02076852,
      "epoch": 0.28016593518908195,
      "flos": 33547676584320.0,
      "grad_norm": 2.100938306879155,
      "language_loss": 0.77496648,
      "learning_rate": 3.378807796086428e-06,
      "loss": 0.79684114,
      "num_input_tokens_seen": 50006075,
      "step": 2330,
      "time_per_iteration": 2.6494498252868652
    },
    {
      "auxiliary_loss_clip": 0.01209438,
      "auxiliary_loss_mlp": 0.01031569,
      "balance_loss_clip": 1.06612802,
      "balance_loss_mlp": 1.02177012,
      "epoch": 0.28028617807972106,
      "flos": 15340823712000.0,
      "grad_norm": 1.9293112768998422,
      "language_loss": 0.77006054,
      "learning_rate": 3.37824342225153e-06,
      "loss": 0.79247063,
      "num_input_tokens_seen": 50022495,
      "step": 2331,
      "time_per_iteration": 2.436617136001587
    },
    {
      "auxiliary_loss_clip": 0.01148122,
      "auxiliary_loss_mlp": 0.01033854,
      "balance_loss_clip": 1.05807436,
      "balance_loss_mlp": 1.02456784,
      "epoch": 0.2804064209703601,
      "flos": 25520277409920.0,
      "grad_norm": 1.7922196358332234,
      "language_loss": 0.77682823,
      "learning_rate": 3.3776788393382006e-06,
      "loss": 0.798648,
      "num_input_tokens_seen": 50041975,
      "step": 2332,
      "time_per_iteration": 2.6401240825653076
    },
    {
      "auxiliary_loss_clip": 0.01206985,
      "auxiliary_loss_mlp": 0.0103003,
      "balance_loss_clip": 1.06424546,
      "balance_loss_mlp": 1.0206244,
      "epoch": 0.2805266638609992,
      "flos": 29351766280320.0,
      "grad_norm": 2.061127767207515,
      "language_loss": 0.76400775,
      "learning_rate": 3.3771140474320872e-06,
      "loss": 0.78637791,
      "num_input_tokens_seen": 50061925,
      "step": 2333,
      "time_per_iteration": 2.5431337356567383
    },
    {
      "auxiliary_loss_clip": 0.01169427,
      "auxiliary_loss_mlp": 0.01036684,
      "balance_loss_clip": 1.0593071,
      "balance_loss_mlp": 1.02760649,
      "epoch": 0.28064690675163834,
      "flos": 21463735875840.0,
      "grad_norm": 2.161381305127918,
      "language_loss": 0.79593223,
      "learning_rate": 3.3765490466188664e-06,
      "loss": 0.81799334,
      "num_input_tokens_seen": 50079325,
      "step": 2334,
      "time_per_iteration": 2.5762736797332764
    },
    {
      "auxiliary_loss_clip": 0.01159924,
      "auxiliary_loss_mlp": 0.01029951,
      "balance_loss_clip": 1.05718207,
      "balance_loss_mlp": 1.02011597,
      "epoch": 0.2807671496422774,
      "flos": 20995640812800.0,
      "grad_norm": 2.473230499343083,
      "language_loss": 0.7369501,
      "learning_rate": 3.3759838369842508e-06,
      "loss": 0.75884885,
      "num_input_tokens_seen": 50097400,
      "step": 2335,
      "time_per_iteration": 2.566033363342285
    },
    {
      "auxiliary_loss_clip": 0.0116503,
      "auxiliary_loss_mlp": 0.01031918,
      "balance_loss_clip": 1.06084752,
      "balance_loss_mlp": 1.02295387,
      "epoch": 0.2808873925329165,
      "flos": 21506577822720.0,
      "grad_norm": 2.2176781950601927,
      "language_loss": 0.72732151,
      "learning_rate": 3.375418418613981e-06,
      "loss": 0.749291,
      "num_input_tokens_seen": 50116425,
      "step": 2336,
      "time_per_iteration": 2.559135675430298
    },
    {
      "auxiliary_loss_clip": 0.01176915,
      "auxiliary_loss_mlp": 0.01034773,
      "balance_loss_clip": 1.06053233,
      "balance_loss_mlp": 1.02447891,
      "epoch": 0.28100763542355556,
      "flos": 16070815814400.0,
      "grad_norm": 3.5260090584103305,
      "language_loss": 0.83013093,
      "learning_rate": 3.374852791593831e-06,
      "loss": 0.85224783,
      "num_input_tokens_seen": 50132625,
      "step": 2337,
      "time_per_iteration": 3.3774821758270264
    },
    {
      "auxiliary_loss_clip": 0.01156944,
      "auxiliary_loss_mlp": 0.01033555,
      "balance_loss_clip": 1.05524826,
      "balance_loss_mlp": 1.02388132,
      "epoch": 0.28112787831419467,
      "flos": 19062605468160.0,
      "grad_norm": 3.0706785236155483,
      "language_loss": 0.53758192,
      "learning_rate": 3.374286956009605e-06,
      "loss": 0.55948687,
      "num_input_tokens_seen": 50151190,
      "step": 2338,
      "time_per_iteration": 2.5810139179229736
    },
    {
      "auxiliary_loss_clip": 0.01192565,
      "auxiliary_loss_mlp": 0.01030757,
      "balance_loss_clip": 1.06559896,
      "balance_loss_mlp": 1.02130938,
      "epoch": 0.2812481212048338,
      "flos": 12823629482880.0,
      "grad_norm": 2.024522059228888,
      "language_loss": 0.75487411,
      "learning_rate": 3.3737209119471405e-06,
      "loss": 0.77710736,
      "num_input_tokens_seen": 50167700,
      "step": 2339,
      "time_per_iteration": 2.5333142280578613
    },
    {
      "auxiliary_loss_clip": 0.01198482,
      "auxiliary_loss_mlp": 0.0103058,
      "balance_loss_clip": 1.06378555,
      "balance_loss_mlp": 1.02078724,
      "epoch": 0.28136836409547283,
      "flos": 15633064765440.0,
      "grad_norm": 3.3468805786252953,
      "language_loss": 0.63466692,
      "learning_rate": 3.373154659492306e-06,
      "loss": 0.65695763,
      "num_input_tokens_seen": 50185840,
      "step": 2340,
      "time_per_iteration": 2.509500741958618
    },
    {
      "auxiliary_loss_clip": 0.01178868,
      "auxiliary_loss_mlp": 0.01045446,
      "balance_loss_clip": 1.060274,
      "balance_loss_mlp": 1.03621376,
      "epoch": 0.28148860698611194,
      "flos": 19933726106880.0,
      "grad_norm": 1.8242314159819109,
      "language_loss": 0.85186571,
      "learning_rate": 3.3725881987310016e-06,
      "loss": 0.87410885,
      "num_input_tokens_seen": 50203375,
      "step": 2341,
      "time_per_iteration": 2.53665828704834
    },
    {
      "auxiliary_loss_clip": 0.01173563,
      "auxiliary_loss_mlp": 0.01033675,
      "balance_loss_clip": 1.05899489,
      "balance_loss_mlp": 1.02503204,
      "epoch": 0.28160884987675106,
      "flos": 17457219008640.0,
      "grad_norm": 1.958764281288809,
      "language_loss": 0.87675655,
      "learning_rate": 3.372021529749159e-06,
      "loss": 0.89882898,
      "num_input_tokens_seen": 50222435,
      "step": 2342,
      "time_per_iteration": 3.409428358078003
    },
    {
      "auxiliary_loss_clip": 0.01133496,
      "auxiliary_loss_mlp": 0.01033349,
      "balance_loss_clip": 1.05624151,
      "balance_loss_mlp": 1.02449799,
      "epoch": 0.2817290927673901,
      "flos": 16834743290880.0,
      "grad_norm": 1.856778383664567,
      "language_loss": 0.92201102,
      "learning_rate": 3.3714546526327405e-06,
      "loss": 0.94367945,
      "num_input_tokens_seen": 50240435,
      "step": 2343,
      "time_per_iteration": 3.4450154304504395
    },
    {
      "auxiliary_loss_clip": 0.01166647,
      "auxiliary_loss_mlp": 0.01031424,
      "balance_loss_clip": 1.05773675,
      "balance_loss_mlp": 1.02154708,
      "epoch": 0.2818493356580292,
      "flos": 15414081500160.0,
      "grad_norm": 1.9954686708102507,
      "language_loss": 0.87517333,
      "learning_rate": 3.3708875674677423e-06,
      "loss": 0.89715403,
      "num_input_tokens_seen": 50258410,
      "step": 2344,
      "time_per_iteration": 3.3381736278533936
    },
    {
      "auxiliary_loss_clip": 0.01186885,
      "auxiliary_loss_mlp": 0.0103178,
      "balance_loss_clip": 1.06462586,
      "balance_loss_mlp": 1.02197528,
      "epoch": 0.28196957854866833,
      "flos": 20412451595520.0,
      "grad_norm": 1.9092639907027233,
      "language_loss": 0.83694327,
      "learning_rate": 3.37032027434019e-06,
      "loss": 0.85912991,
      "num_input_tokens_seen": 50277930,
      "step": 2345,
      "time_per_iteration": 2.535391330718994
    },
    {
      "auxiliary_loss_clip": 0.01202432,
      "auxiliary_loss_mlp": 0.01031873,
      "balance_loss_clip": 1.06387603,
      "balance_loss_mlp": 1.02106702,
      "epoch": 0.2820898214393074,
      "flos": 19973120348160.0,
      "grad_norm": 5.171048350433416,
      "language_loss": 0.82612407,
      "learning_rate": 3.369752773336141e-06,
      "loss": 0.84846711,
      "num_input_tokens_seen": 50297410,
      "step": 2346,
      "time_per_iteration": 2.5055010318756104
    },
    {
      "auxiliary_loss_clip": 0.01173886,
      "auxiliary_loss_mlp": 0.01034297,
      "balance_loss_clip": 1.05696893,
      "balance_loss_mlp": 1.02402127,
      "epoch": 0.2822100643299465,
      "flos": 22528308188160.0,
      "grad_norm": 1.6579476078028912,
      "language_loss": 0.77877158,
      "learning_rate": 3.3691850645416864e-06,
      "loss": 0.80085343,
      "num_input_tokens_seen": 50317120,
      "step": 2347,
      "time_per_iteration": 2.5489501953125
    },
    {
      "auxiliary_loss_clip": 0.01197893,
      "auxiliary_loss_mlp": 0.01039699,
      "balance_loss_clip": 1.06306696,
      "balance_loss_mlp": 1.02986479,
      "epoch": 0.2823303072205856,
      "flos": 11546682007680.0,
      "grad_norm": 1.882846620029653,
      "language_loss": 0.8282944,
      "learning_rate": 3.368617148042945e-06,
      "loss": 0.85067034,
      "num_input_tokens_seen": 50334790,
      "step": 2348,
      "time_per_iteration": 2.505981683731079
    },
    {
      "auxiliary_loss_clip": 0.01170172,
      "auxiliary_loss_mlp": 0.0103684,
      "balance_loss_clip": 1.05508637,
      "balance_loss_mlp": 1.0265466,
      "epoch": 0.28245055011122466,
      "flos": 18259894281600.0,
      "grad_norm": 1.8691618871655928,
      "language_loss": 0.8425113,
      "learning_rate": 3.368049023926071e-06,
      "loss": 0.86458147,
      "num_input_tokens_seen": 50353785,
      "step": 2349,
      "time_per_iteration": 2.526679039001465
    },
    {
      "auxiliary_loss_clip": 0.011914,
      "auxiliary_loss_mlp": 0.0103685,
      "balance_loss_clip": 1.06355131,
      "balance_loss_mlp": 1.0282495,
      "epoch": 0.2825707930018638,
      "flos": 24608110504320.0,
      "grad_norm": 1.595431489534568,
      "language_loss": 0.83579248,
      "learning_rate": 3.3674806922772476e-06,
      "loss": 0.85807496,
      "num_input_tokens_seen": 50374670,
      "step": 2350,
      "time_per_iteration": 2.624443531036377
    },
    {
      "auxiliary_loss_clip": 0.01170053,
      "auxiliary_loss_mlp": 0.01040415,
      "balance_loss_clip": 1.05922985,
      "balance_loss_mlp": 1.03089666,
      "epoch": 0.28269103589250283,
      "flos": 25226994862080.0,
      "grad_norm": 2.136161510632417,
      "language_loss": 0.74909902,
      "learning_rate": 3.3669121531826904e-06,
      "loss": 0.77120376,
      "num_input_tokens_seen": 50395650,
      "step": 2351,
      "time_per_iteration": 2.659956216812134
    },
    {
      "auxiliary_loss_clip": 0.01160098,
      "auxiliary_loss_mlp": 0.01029089,
      "balance_loss_clip": 1.06141984,
      "balance_loss_mlp": 1.02030897,
      "epoch": 0.28281127878314194,
      "flos": 19281552819840.0,
      "grad_norm": 2.173373283419757,
      "language_loss": 0.83295619,
      "learning_rate": 3.366343406728647e-06,
      "loss": 0.85484803,
      "num_input_tokens_seen": 50415100,
      "step": 2352,
      "time_per_iteration": 2.5878121852874756
    },
    {
      "auxiliary_loss_clip": 0.01183629,
      "auxiliary_loss_mlp": 0.01030229,
      "balance_loss_clip": 1.05737889,
      "balance_loss_mlp": 1.02094913,
      "epoch": 0.28293152167378105,
      "flos": 23878405710720.0,
      "grad_norm": 1.707425677985109,
      "language_loss": 0.68638068,
      "learning_rate": 3.3657744530013946e-06,
      "loss": 0.70851934,
      "num_input_tokens_seen": 50434335,
      "step": 2353,
      "time_per_iteration": 2.5580530166625977
    },
    {
      "auxiliary_loss_clip": 0.01195874,
      "auxiliary_loss_mlp": 0.01030115,
      "balance_loss_clip": 1.06339359,
      "balance_loss_mlp": 1.02051306,
      "epoch": 0.2830517645644201,
      "flos": 43866965928960.0,
      "grad_norm": 2.0762422535772638,
      "language_loss": 0.71071017,
      "learning_rate": 3.3652052920872437e-06,
      "loss": 0.73297,
      "num_input_tokens_seen": 50457200,
      "step": 2354,
      "time_per_iteration": 2.7134580612182617
    },
    {
      "auxiliary_loss_clip": 0.01179273,
      "auxiliary_loss_mlp": 0.01037555,
      "balance_loss_clip": 1.05931056,
      "balance_loss_mlp": 1.02764273,
      "epoch": 0.2831720074550592,
      "flos": 26651750803200.0,
      "grad_norm": 1.8624680207898485,
      "language_loss": 0.85815382,
      "learning_rate": 3.3646359240725355e-06,
      "loss": 0.8803221,
      "num_input_tokens_seen": 50476390,
      "step": 2355,
      "time_per_iteration": 2.614729642868042
    },
    {
      "auxiliary_loss_clip": 0.01185177,
      "auxiliary_loss_mlp": 0.00764724,
      "balance_loss_clip": 1.06031001,
      "balance_loss_mlp": 1.00082004,
      "epoch": 0.2832922503456983,
      "flos": 31029979564800.0,
      "grad_norm": 2.694359359149424,
      "language_loss": 0.67484367,
      "learning_rate": 3.364066349043643e-06,
      "loss": 0.69434267,
      "num_input_tokens_seen": 50497595,
      "step": 2356,
      "time_per_iteration": 2.6167221069335938
    },
    {
      "auxiliary_loss_clip": 0.01175585,
      "auxiliary_loss_mlp": 0.01032289,
      "balance_loss_clip": 1.06018806,
      "balance_loss_mlp": 1.02372348,
      "epoch": 0.2834124932363374,
      "flos": 20405699838720.0,
      "grad_norm": 1.9004247575826367,
      "language_loss": 0.82089382,
      "learning_rate": 3.363496567086969e-06,
      "loss": 0.84297258,
      "num_input_tokens_seen": 50514690,
      "step": 2357,
      "time_per_iteration": 2.53696870803833
    },
    {
      "auxiliary_loss_clip": 0.01206444,
      "auxiliary_loss_mlp": 0.01030636,
      "balance_loss_clip": 1.06446064,
      "balance_loss_mlp": 1.02153432,
      "epoch": 0.2835327361269765,
      "flos": 39384848056320.0,
      "grad_norm": 1.9190401425507195,
      "language_loss": 0.75401032,
      "learning_rate": 3.3629265782889506e-06,
      "loss": 0.77638113,
      "num_input_tokens_seen": 50536515,
      "step": 2358,
      "time_per_iteration": 2.6603875160217285
    },
    {
      "auxiliary_loss_clip": 0.01157554,
      "auxiliary_loss_mlp": 0.01033714,
      "balance_loss_clip": 1.055071,
      "balance_loss_mlp": 1.02403402,
      "epoch": 0.2836529790176156,
      "flos": 30261598801920.0,
      "grad_norm": 2.042476570994474,
      "language_loss": 0.72166002,
      "learning_rate": 3.362356382736054e-06,
      "loss": 0.74357271,
      "num_input_tokens_seen": 50557120,
      "step": 2359,
      "time_per_iteration": 2.6420087814331055
    },
    {
      "auxiliary_loss_clip": 0.0116007,
      "auxiliary_loss_mlp": 0.01023801,
      "balance_loss_clip": 1.05396056,
      "balance_loss_mlp": 1.01566529,
      "epoch": 0.28377322190825466,
      "flos": 12677796264960.0,
      "grad_norm": 1.9568675531477338,
      "language_loss": 0.90907407,
      "learning_rate": 3.361785980514777e-06,
      "loss": 0.93091279,
      "num_input_tokens_seen": 50573320,
      "step": 2360,
      "time_per_iteration": 2.6443564891815186
    },
    {
      "auxiliary_loss_clip": 0.01128404,
      "auxiliary_loss_mlp": 0.01035024,
      "balance_loss_clip": 1.05643725,
      "balance_loss_mlp": 1.02615535,
      "epoch": 0.28389346479889377,
      "flos": 18296666830080.0,
      "grad_norm": 1.9874025011676186,
      "language_loss": 0.76633704,
      "learning_rate": 3.361215371711649e-06,
      "loss": 0.78797132,
      "num_input_tokens_seen": 50592415,
      "step": 2361,
      "time_per_iteration": 2.6237235069274902
    },
    {
      "auxiliary_loss_clip": 0.01155492,
      "auxiliary_loss_mlp": 0.01031293,
      "balance_loss_clip": 1.05587792,
      "balance_loss_mlp": 1.02279377,
      "epoch": 0.2840137076895329,
      "flos": 20406992728320.0,
      "grad_norm": 1.9902492788309643,
      "language_loss": 0.83466303,
      "learning_rate": 3.3606445564132326e-06,
      "loss": 0.8565309,
      "num_input_tokens_seen": 50609710,
      "step": 2362,
      "time_per_iteration": 2.636603832244873
    },
    {
      "auxiliary_loss_clip": 0.01208778,
      "auxiliary_loss_mlp": 0.00764205,
      "balance_loss_clip": 1.06684673,
      "balance_loss_mlp": 1.00074553,
      "epoch": 0.28413395058017193,
      "flos": 20048030161920.0,
      "grad_norm": 1.960051778617603,
      "language_loss": 0.82106245,
      "learning_rate": 3.360073534706118e-06,
      "loss": 0.8407923,
      "num_input_tokens_seen": 50626865,
      "step": 2363,
      "time_per_iteration": 2.5006539821624756
    },
    {
      "auxiliary_loss_clip": 0.0117748,
      "auxiliary_loss_mlp": 0.01027523,
      "balance_loss_clip": 1.05975437,
      "balance_loss_mlp": 1.01862371,
      "epoch": 0.28425419347081105,
      "flos": 37663613256960.0,
      "grad_norm": 2.5619835828188657,
      "language_loss": 0.75949162,
      "learning_rate": 3.35950230667693e-06,
      "loss": 0.78154165,
      "num_input_tokens_seen": 50648560,
      "step": 2364,
      "time_per_iteration": 3.575005292892456
    },
    {
      "auxiliary_loss_clip": 0.01193141,
      "auxiliary_loss_mlp": 0.01026669,
      "balance_loss_clip": 1.06227612,
      "balance_loss_mlp": 1.01805031,
      "epoch": 0.28437443636145016,
      "flos": 13845072539520.0,
      "grad_norm": 2.300519316214218,
      "language_loss": 0.85227096,
      "learning_rate": 3.358930872412323e-06,
      "loss": 0.87446904,
      "num_input_tokens_seen": 50665725,
      "step": 2365,
      "time_per_iteration": 2.4773616790771484
    },
    {
      "auxiliary_loss_clip": 0.0118766,
      "auxiliary_loss_mlp": 0.01033034,
      "balance_loss_clip": 1.06158233,
      "balance_loss_mlp": 1.02386665,
      "epoch": 0.2844946792520892,
      "flos": 22747794243840.0,
      "grad_norm": 1.7623387457713378,
      "language_loss": 0.80856657,
      "learning_rate": 3.3583592319989825e-06,
      "loss": 0.83077353,
      "num_input_tokens_seen": 50685095,
      "step": 2366,
      "time_per_iteration": 2.582291603088379
    },
    {
      "auxiliary_loss_clip": 0.01198795,
      "auxiliary_loss_mlp": 0.01037993,
      "balance_loss_clip": 1.06252277,
      "balance_loss_mlp": 1.02849197,
      "epoch": 0.2846149221427283,
      "flos": 32415987709440.0,
      "grad_norm": 2.1967308208184906,
      "language_loss": 0.68716174,
      "learning_rate": 3.357787385523627e-06,
      "loss": 0.70952964,
      "num_input_tokens_seen": 50706500,
      "step": 2367,
      "time_per_iteration": 2.585573196411133
    },
    {
      "auxiliary_loss_clip": 0.01141871,
      "auxiliary_loss_mlp": 0.01031395,
      "balance_loss_clip": 1.05591512,
      "balance_loss_mlp": 1.02290118,
      "epoch": 0.2847351650333674,
      "flos": 28475976873600.0,
      "grad_norm": 2.025690913299581,
      "language_loss": 0.82715809,
      "learning_rate": 3.3572153330730048e-06,
      "loss": 0.84889078,
      "num_input_tokens_seen": 50727595,
      "step": 2368,
      "time_per_iteration": 2.7367939949035645
    },
    {
      "auxiliary_loss_clip": 0.01094788,
      "auxiliary_loss_mlp": 0.01001877,
      "balance_loss_clip": 1.04459929,
      "balance_loss_mlp": 1.00039852,
      "epoch": 0.2848554079240065,
      "flos": 55753399704960.0,
      "grad_norm": 0.8242773002180156,
      "language_loss": 0.64665413,
      "learning_rate": 3.3566430747338956e-06,
      "loss": 0.66762078,
      "num_input_tokens_seen": 50782800,
      "step": 2369,
      "time_per_iteration": 3.959866762161255
    },
    {
      "auxiliary_loss_clip": 0.01192984,
      "auxiliary_loss_mlp": 0.01031914,
      "balance_loss_clip": 1.06040013,
      "balance_loss_mlp": 1.0227232,
      "epoch": 0.2849756508146456,
      "flos": 11836875985920.0,
      "grad_norm": 4.972296229661133,
      "language_loss": 0.86422235,
      "learning_rate": 3.35607061059311e-06,
      "loss": 0.88647127,
      "num_input_tokens_seen": 50797730,
      "step": 2370,
      "time_per_iteration": 3.390984296798706
    },
    {
      "auxiliary_loss_clip": 0.01201395,
      "auxiliary_loss_mlp": 0.01033095,
      "balance_loss_clip": 1.06159592,
      "balance_loss_mlp": 1.02454221,
      "epoch": 0.28509589370528465,
      "flos": 25155209531520.0,
      "grad_norm": 1.8411600341994314,
      "language_loss": 0.75162351,
      "learning_rate": 3.3554979407374917e-06,
      "loss": 0.77396846,
      "num_input_tokens_seen": 50819840,
      "step": 2371,
      "time_per_iteration": 2.599081039428711
    },
    {
      "auxiliary_loss_clip": 0.01192201,
      "auxiliary_loss_mlp": 0.01033268,
      "balance_loss_clip": 1.06195831,
      "balance_loss_mlp": 1.02445197,
      "epoch": 0.28521613659592376,
      "flos": 19974808287360.0,
      "grad_norm": 2.168944280064147,
      "language_loss": 0.73397607,
      "learning_rate": 3.3549250652539134e-06,
      "loss": 0.75623071,
      "num_input_tokens_seen": 50838935,
      "step": 2372,
      "time_per_iteration": 2.517530679702759
    },
    {
      "auxiliary_loss_clip": 0.01175333,
      "auxiliary_loss_mlp": 0.01032283,
      "balance_loss_clip": 1.05793881,
      "balance_loss_mlp": 1.02278793,
      "epoch": 0.2853363794865629,
      "flos": 23367971491200.0,
      "grad_norm": 1.8746988481138647,
      "language_loss": 0.81731111,
      "learning_rate": 3.3543519842292794e-06,
      "loss": 0.8393873,
      "num_input_tokens_seen": 50858590,
      "step": 2373,
      "time_per_iteration": 2.567857265472412
    },
    {
      "auxiliary_loss_clip": 0.01206605,
      "auxiliary_loss_mlp": 0.00763809,
      "balance_loss_clip": 1.06479335,
      "balance_loss_mlp": 1.0007211,
      "epoch": 0.28545662237720193,
      "flos": 19861940776320.0,
      "grad_norm": 1.744301002779992,
      "language_loss": 0.83457518,
      "learning_rate": 3.353778697750527e-06,
      "loss": 0.85427928,
      "num_input_tokens_seen": 50876995,
      "step": 2374,
      "time_per_iteration": 2.4623165130615234
    },
    {
      "auxiliary_loss_clip": 0.011677,
      "auxiliary_loss_mlp": 0.01027711,
      "balance_loss_clip": 1.05745029,
      "balance_loss_mlp": 1.01866889,
      "epoch": 0.28557686526784104,
      "flos": 23879016241920.0,
      "grad_norm": 1.9550237495053546,
      "language_loss": 0.89293051,
      "learning_rate": 3.353205205904622e-06,
      "loss": 0.91488457,
      "num_input_tokens_seen": 50896105,
      "step": 2375,
      "time_per_iteration": 2.658015727996826
    },
    {
      "auxiliary_loss_clip": 0.01176211,
      "auxiliary_loss_mlp": 0.0103181,
      "balance_loss_clip": 1.059587,
      "balance_loss_mlp": 1.02300668,
      "epoch": 0.28569710815848015,
      "flos": 44890384233600.0,
      "grad_norm": 2.1578884171303163,
      "language_loss": 0.72087818,
      "learning_rate": 3.3526315087785637e-06,
      "loss": 0.74295843,
      "num_input_tokens_seen": 50917220,
      "step": 2376,
      "time_per_iteration": 2.7428653240203857
    },
    {
      "auxiliary_loss_clip": 0.01125875,
      "auxiliary_loss_mlp": 0.01031991,
      "balance_loss_clip": 1.05305409,
      "balance_loss_mlp": 1.02302682,
      "epoch": 0.2858173510491192,
      "flos": 26829759628800.0,
      "grad_norm": 1.6070539904570438,
      "language_loss": 0.80692399,
      "learning_rate": 3.3520576064593805e-06,
      "loss": 0.82850266,
      "num_input_tokens_seen": 50937175,
      "step": 2377,
      "time_per_iteration": 2.633937358856201
    },
    {
      "auxiliary_loss_clip": 0.01194351,
      "auxiliary_loss_mlp": 0.01027429,
      "balance_loss_clip": 1.06182778,
      "balance_loss_mlp": 1.01830363,
      "epoch": 0.2859375939397583,
      "flos": 23148916398720.0,
      "grad_norm": 1.7961619675202265,
      "language_loss": 0.81772435,
      "learning_rate": 3.3514834990341337e-06,
      "loss": 0.8399421,
      "num_input_tokens_seen": 50957500,
      "step": 2378,
      "time_per_iteration": 2.606462240219116
    },
    {
      "auxiliary_loss_clip": 0.01184098,
      "auxiliary_loss_mlp": 0.0103082,
      "balance_loss_clip": 1.0625329,
      "balance_loss_mlp": 1.0225054,
      "epoch": 0.2860578368303974,
      "flos": 12129799397760.0,
      "grad_norm": 3.7731482053835363,
      "language_loss": 0.92979473,
      "learning_rate": 3.3509091865899144e-06,
      "loss": 0.95194387,
      "num_input_tokens_seen": 50972690,
      "step": 2379,
      "time_per_iteration": 2.6021652221679688
    },
    {
      "auxiliary_loss_clip": 0.01205216,
      "auxiliary_loss_mlp": 0.01032062,
      "balance_loss_clip": 1.06268048,
      "balance_loss_mlp": 1.02290678,
      "epoch": 0.2861780797210365,
      "flos": 19938035738880.0,
      "grad_norm": 2.329486631853123,
      "language_loss": 0.7042048,
      "learning_rate": 3.350334669213846e-06,
      "loss": 0.72657758,
      "num_input_tokens_seen": 50990095,
      "step": 2380,
      "time_per_iteration": 2.533942937850952
    },
    {
      "auxiliary_loss_clip": 0.01187849,
      "auxiliary_loss_mlp": 0.01033934,
      "balance_loss_clip": 1.06122661,
      "balance_loss_mlp": 1.02572083,
      "epoch": 0.2862983226116756,
      "flos": 27563127609600.0,
      "grad_norm": 1.9372811296714165,
      "language_loss": 0.75630999,
      "learning_rate": 3.3497599469930816e-06,
      "loss": 0.77852774,
      "num_input_tokens_seen": 51008305,
      "step": 2381,
      "time_per_iteration": 2.552480936050415
    },
    {
      "auxiliary_loss_clip": 0.0120487,
      "auxiliary_loss_mlp": 0.01031317,
      "balance_loss_clip": 1.0609355,
      "balance_loss_mlp": 1.02201271,
      "epoch": 0.28641856550231465,
      "flos": 22053964158720.0,
      "grad_norm": 2.169885444424467,
      "language_loss": 0.83164561,
      "learning_rate": 3.349185020014807e-06,
      "loss": 0.85400754,
      "num_input_tokens_seen": 51025570,
      "step": 2382,
      "time_per_iteration": 2.498779535293579
    },
    {
      "auxiliary_loss_clip": 0.01193425,
      "auxiliary_loss_mlp": 0.01029867,
      "balance_loss_clip": 1.06109154,
      "balance_loss_mlp": 1.02130163,
      "epoch": 0.28653880839295376,
      "flos": 22378775869440.0,
      "grad_norm": 2.104845382266116,
      "language_loss": 0.74216336,
      "learning_rate": 3.348609888366237e-06,
      "loss": 0.76439625,
      "num_input_tokens_seen": 51044585,
      "step": 2383,
      "time_per_iteration": 2.539055824279785
    },
    {
      "auxiliary_loss_clip": 0.01127379,
      "auxiliary_loss_mlp": 0.0102561,
      "balance_loss_clip": 1.05329919,
      "balance_loss_mlp": 1.01644325,
      "epoch": 0.28665905128359287,
      "flos": 23367971491200.0,
      "grad_norm": 2.32982882217772,
      "language_loss": 0.62526208,
      "learning_rate": 3.348034552134619e-06,
      "loss": 0.64679199,
      "num_input_tokens_seen": 51063990,
      "step": 2384,
      "time_per_iteration": 2.6518490314483643
    },
    {
      "auxiliary_loss_clip": 0.01140508,
      "auxiliary_loss_mlp": 0.01032827,
      "balance_loss_clip": 1.05743515,
      "balance_loss_mlp": 1.02445817,
      "epoch": 0.2867792941742319,
      "flos": 20881695893760.0,
      "grad_norm": 1.9825600712093907,
      "language_loss": 0.83897316,
      "learning_rate": 3.3474590114072316e-06,
      "loss": 0.86070657,
      "num_input_tokens_seen": 51081990,
      "step": 2385,
      "time_per_iteration": 2.6022582054138184
    },
    {
      "auxiliary_loss_clip": 0.0116075,
      "auxiliary_loss_mlp": 0.0103603,
      "balance_loss_clip": 1.06148148,
      "balance_loss_mlp": 1.02677917,
      "epoch": 0.28689953706487104,
      "flos": 20664005518080.0,
      "grad_norm": 2.4936825176014845,
      "language_loss": 0.82716608,
      "learning_rate": 3.3468832662713836e-06,
      "loss": 0.84913391,
      "num_input_tokens_seen": 51100235,
      "step": 2386,
      "time_per_iteration": 2.5869791507720947
    },
    {
      "auxiliary_loss_clip": 0.01155358,
      "auxiliary_loss_mlp": 0.01037734,
      "balance_loss_clip": 1.05651855,
      "balance_loss_mlp": 1.02873945,
      "epoch": 0.28701977995551015,
      "flos": 12675533708160.0,
      "grad_norm": 2.1766105322883322,
      "language_loss": 0.83761138,
      "learning_rate": 3.346307316814415e-06,
      "loss": 0.85954225,
      "num_input_tokens_seen": 51115405,
      "step": 2387,
      "time_per_iteration": 2.56998348236084
    },
    {
      "auxiliary_loss_clip": 0.0119017,
      "auxiliary_loss_mlp": 0.01030491,
      "balance_loss_clip": 1.06199944,
      "balance_loss_mlp": 1.02121699,
      "epoch": 0.2871400228461492,
      "flos": 21252366293760.0,
      "grad_norm": 2.1630717103264665,
      "language_loss": 0.75839806,
      "learning_rate": 3.3457311631236965e-06,
      "loss": 0.78060466,
      "num_input_tokens_seen": 51136390,
      "step": 2388,
      "time_per_iteration": 2.5333447456359863
    },
    {
      "auxiliary_loss_clip": 0.01163842,
      "auxiliary_loss_mlp": 0.01030202,
      "balance_loss_clip": 1.05806637,
      "balance_loss_mlp": 1.02139854,
      "epoch": 0.2872602657367883,
      "flos": 25119262995840.0,
      "grad_norm": 1.755746825594979,
      "language_loss": 0.84420305,
      "learning_rate": 3.345154805286631e-06,
      "loss": 0.86614358,
      "num_input_tokens_seen": 51156650,
      "step": 2389,
      "time_per_iteration": 2.6632096767425537
    },
    {
      "auxiliary_loss_clip": 0.01186223,
      "auxiliary_loss_mlp": 0.0103386,
      "balance_loss_clip": 1.06000936,
      "balance_loss_mlp": 1.02463937,
      "epoch": 0.2873805086274274,
      "flos": 16646606830080.0,
      "grad_norm": 2.4479971218267007,
      "language_loss": 0.76206875,
      "learning_rate": 3.344578243390651e-06,
      "loss": 0.78426957,
      "num_input_tokens_seen": 51172210,
      "step": 2390,
      "time_per_iteration": 2.51564884185791
    },
    {
      "auxiliary_loss_clip": 0.0117361,
      "auxiliary_loss_mlp": 0.01032658,
      "balance_loss_clip": 1.0606041,
      "balance_loss_mlp": 1.02313328,
      "epoch": 0.2875007515180665,
      "flos": 17420123237760.0,
      "grad_norm": 3.1989744779227776,
      "language_loss": 0.78559852,
      "learning_rate": 3.3440014775232206e-06,
      "loss": 0.80766124,
      "num_input_tokens_seen": 51190265,
      "step": 2391,
      "time_per_iteration": 3.3988842964172363
    },
    {
      "auxiliary_loss_clip": 0.01164769,
      "auxiliary_loss_mlp": 0.01030922,
      "balance_loss_clip": 1.05933332,
      "balance_loss_mlp": 1.02253604,
      "epoch": 0.2876209944087056,
      "flos": 23434190213760.0,
      "grad_norm": 2.9111860552634603,
      "language_loss": 0.70960832,
      "learning_rate": 3.343424507771834e-06,
      "loss": 0.73156524,
      "num_input_tokens_seen": 51208475,
      "step": 2392,
      "time_per_iteration": 2.6180832386016846
    },
    {
      "auxiliary_loss_clip": 0.01160356,
      "auxiliary_loss_mlp": 0.01028745,
      "balance_loss_clip": 1.05859554,
      "balance_loss_mlp": 1.020329,
      "epoch": 0.2877412372993447,
      "flos": 13735509079680.0,
      "grad_norm": 1.7787883190164946,
      "language_loss": 0.8678295,
      "learning_rate": 3.342847334224018e-06,
      "loss": 0.8897205,
      "num_input_tokens_seen": 51225875,
      "step": 2393,
      "time_per_iteration": 2.6140360832214355
    },
    {
      "auxiliary_loss_clip": 0.01110196,
      "auxiliary_loss_mlp": 0.01006328,
      "balance_loss_clip": 1.04217052,
      "balance_loss_mlp": 1.0049628,
      "epoch": 0.28786148018998375,
      "flos": 58079695104000.0,
      "grad_norm": 0.9416801896126653,
      "language_loss": 0.62427354,
      "learning_rate": 3.342269956967329e-06,
      "loss": 0.64543879,
      "num_input_tokens_seen": 51287780,
      "step": 2394,
      "time_per_iteration": 3.1465885639190674
    },
    {
      "auxiliary_loss_clip": 0.01196602,
      "auxiliary_loss_mlp": 0.01036542,
      "balance_loss_clip": 1.06365514,
      "balance_loss_mlp": 1.02614737,
      "epoch": 0.28798172308062286,
      "flos": 23435052140160.0,
      "grad_norm": 2.5055395850598643,
      "language_loss": 0.71856058,
      "learning_rate": 3.341692376089355e-06,
      "loss": 0.74089205,
      "num_input_tokens_seen": 51303335,
      "step": 2395,
      "time_per_iteration": 3.4253551959991455
    },
    {
      "auxiliary_loss_clip": 0.01189538,
      "auxiliary_loss_mlp": 0.01033825,
      "balance_loss_clip": 1.06328869,
      "balance_loss_mlp": 1.02490187,
      "epoch": 0.288101965971262,
      "flos": 25110033200640.0,
      "grad_norm": 4.208070991132828,
      "language_loss": 0.84362888,
      "learning_rate": 3.3411145916777146e-06,
      "loss": 0.86586249,
      "num_input_tokens_seen": 51317495,
      "step": 2396,
      "time_per_iteration": 3.314282178878784
    },
    {
      "auxiliary_loss_clip": 0.01168151,
      "auxiliary_loss_mlp": 0.01031198,
      "balance_loss_clip": 1.05717874,
      "balance_loss_mlp": 1.02188182,
      "epoch": 0.28822220886190103,
      "flos": 16252559654400.0,
      "grad_norm": 7.363153001333871,
      "language_loss": 0.90462404,
      "learning_rate": 3.3405366038200566e-06,
      "loss": 0.92661756,
      "num_input_tokens_seen": 51336430,
      "step": 2397,
      "time_per_iteration": 3.2972218990325928
    },
    {
      "auxiliary_loss_clip": 0.01180515,
      "auxiliary_loss_mlp": 0.01039798,
      "balance_loss_clip": 1.06512213,
      "balance_loss_mlp": 1.03036833,
      "epoch": 0.28834245175254014,
      "flos": 24535642815360.0,
      "grad_norm": 2.316113287464621,
      "language_loss": 0.8474431,
      "learning_rate": 3.3399584126040617e-06,
      "loss": 0.86964619,
      "num_input_tokens_seen": 51355930,
      "step": 2398,
      "time_per_iteration": 2.5883123874664307
    },
    {
      "auxiliary_loss_clip": 0.01205295,
      "auxiliary_loss_mlp": 0.00763223,
      "balance_loss_clip": 1.06315017,
      "balance_loss_mlp": 1.00055063,
      "epoch": 0.2884626946431792,
      "flos": 24571445696640.0,
      "grad_norm": 1.895950222794461,
      "language_loss": 0.90676975,
      "learning_rate": 3.339380018117441e-06,
      "loss": 0.92645496,
      "num_input_tokens_seen": 51376765,
      "step": 2399,
      "time_per_iteration": 2.533411979675293
    },
    {
      "auxiliary_loss_clip": 0.01188791,
      "auxiliary_loss_mlp": 0.01029815,
      "balance_loss_clip": 1.0625298,
      "balance_loss_mlp": 1.02110028,
      "epoch": 0.2885829375338183,
      "flos": 16544657053440.0,
      "grad_norm": 2.5877069647403452,
      "language_loss": 0.78251559,
      "learning_rate": 3.3388014204479366e-06,
      "loss": 0.80470169,
      "num_input_tokens_seen": 51394570,
      "step": 2400,
      "time_per_iteration": 2.5092945098876953
    },
    {
      "auxiliary_loss_clip": 0.01210963,
      "auxiliary_loss_mlp": 0.01029922,
      "balance_loss_clip": 1.06550336,
      "balance_loss_mlp": 1.02107632,
      "epoch": 0.2887031804244574,
      "flos": 24061226958720.0,
      "grad_norm": 2.6541472970759554,
      "language_loss": 0.91612792,
      "learning_rate": 3.338222619683321e-06,
      "loss": 0.9385367,
      "num_input_tokens_seen": 51414535,
      "step": 2401,
      "time_per_iteration": 2.5581748485565186
    },
    {
      "auxiliary_loss_clip": 0.011797,
      "auxiliary_loss_mlp": 0.01028987,
      "balance_loss_clip": 1.06140471,
      "balance_loss_mlp": 1.01964736,
      "epoch": 0.2888234233150965,
      "flos": 23330696152320.0,
      "grad_norm": 2.4906168921043914,
      "language_loss": 0.73581588,
      "learning_rate": 3.337643615911398e-06,
      "loss": 0.75790274,
      "num_input_tokens_seen": 51434160,
      "step": 2402,
      "time_per_iteration": 2.5718328952789307
    },
    {
      "auxiliary_loss_clip": 0.01192058,
      "auxiliary_loss_mlp": 0.01028843,
      "balance_loss_clip": 1.06080925,
      "balance_loss_mlp": 1.01890087,
      "epoch": 0.2889436662057356,
      "flos": 22272767856000.0,
      "grad_norm": 2.117520059593653,
      "language_loss": 0.7878207,
      "learning_rate": 3.3370644092200026e-06,
      "loss": 0.81002975,
      "num_input_tokens_seen": 51451435,
      "step": 2403,
      "time_per_iteration": 2.558042526245117
    },
    {
      "auxiliary_loss_clip": 0.01145025,
      "auxiliary_loss_mlp": 0.01032651,
      "balance_loss_clip": 1.05169439,
      "balance_loss_mlp": 1.02333522,
      "epoch": 0.2890639090963747,
      "flos": 21616931381760.0,
      "grad_norm": 1.7973179924388445,
      "language_loss": 0.78104806,
      "learning_rate": 3.3364849996969985e-06,
      "loss": 0.80282485,
      "num_input_tokens_seen": 51471455,
      "step": 2404,
      "time_per_iteration": 2.6071979999542236
    },
    {
      "auxiliary_loss_clip": 0.01188986,
      "auxiliary_loss_mlp": 0.01030196,
      "balance_loss_clip": 1.06104934,
      "balance_loss_mlp": 1.02102888,
      "epoch": 0.28918415198701375,
      "flos": 28585540333440.0,
      "grad_norm": 1.806707488056981,
      "language_loss": 0.8532275,
      "learning_rate": 3.335905387430283e-06,
      "loss": 0.87541926,
      "num_input_tokens_seen": 51492890,
      "step": 2405,
      "time_per_iteration": 2.5849053859710693
    },
    {
      "auxiliary_loss_clip": 0.01178508,
      "auxiliary_loss_mlp": 0.01033211,
      "balance_loss_clip": 1.05829358,
      "balance_loss_mlp": 1.02399671,
      "epoch": 0.28930439487765286,
      "flos": 21944688007680.0,
      "grad_norm": 1.7859010656047312,
      "language_loss": 0.83107346,
      "learning_rate": 3.335325572507782e-06,
      "loss": 0.85319066,
      "num_input_tokens_seen": 51513390,
      "step": 2406,
      "time_per_iteration": 2.5744731426239014
    },
    {
      "auxiliary_loss_clip": 0.01210634,
      "auxiliary_loss_mlp": 0.00764099,
      "balance_loss_clip": 1.06816113,
      "balance_loss_mlp": 1.00064349,
      "epoch": 0.28942463776829197,
      "flos": 19281911955840.0,
      "grad_norm": 1.6883569413838524,
      "language_loss": 0.73805565,
      "learning_rate": 3.3347455550174537e-06,
      "loss": 0.75780296,
      "num_input_tokens_seen": 51532730,
      "step": 2407,
      "time_per_iteration": 2.475522518157959
    },
    {
      "auxiliary_loss_clip": 0.01157693,
      "auxiliary_loss_mlp": 0.01034389,
      "balance_loss_clip": 1.05476999,
      "balance_loss_mlp": 1.024405,
      "epoch": 0.289544880658931,
      "flos": 14645700737280.0,
      "grad_norm": 2.0141154729951216,
      "language_loss": 0.68109119,
      "learning_rate": 3.3341653350472864e-06,
      "loss": 0.70301205,
      "num_input_tokens_seen": 51549560,
      "step": 2408,
      "time_per_iteration": 2.5330286026000977
    },
    {
      "auxiliary_loss_clip": 0.01214259,
      "auxiliary_loss_mlp": 0.01033442,
      "balance_loss_clip": 1.06336641,
      "balance_loss_mlp": 1.02283239,
      "epoch": 0.28966512354957014,
      "flos": 28621881918720.0,
      "grad_norm": 2.3235009440326757,
      "language_loss": 0.69177634,
      "learning_rate": 3.333584912685298e-06,
      "loss": 0.71425337,
      "num_input_tokens_seen": 51568180,
      "step": 2409,
      "time_per_iteration": 2.5282223224639893
    },
    {
      "auxiliary_loss_clip": 0.01079926,
      "auxiliary_loss_mlp": 0.01009002,
      "balance_loss_clip": 1.03628111,
      "balance_loss_mlp": 1.00772655,
      "epoch": 0.28978536644020925,
      "flos": 64711784511360.0,
      "grad_norm": 0.9681170968111037,
      "language_loss": 0.55545115,
      "learning_rate": 3.3330042880195385e-06,
      "loss": 0.57634044,
      "num_input_tokens_seen": 51622530,
      "step": 2410,
      "time_per_iteration": 3.09198260307312
    },
    {
      "auxiliary_loss_clip": 0.01176145,
      "auxiliary_loss_mlp": 0.01028157,
      "balance_loss_clip": 1.05848563,
      "balance_loss_mlp": 1.01870334,
      "epoch": 0.2899056093308483,
      "flos": 18624638937600.0,
      "grad_norm": 1.93755332741465,
      "language_loss": 0.78646594,
      "learning_rate": 3.3324234611380888e-06,
      "loss": 0.80850899,
      "num_input_tokens_seen": 51641260,
      "step": 2411,
      "time_per_iteration": 2.5475380420684814
    },
    {
      "auxiliary_loss_clip": 0.01151313,
      "auxiliary_loss_mlp": 0.01032779,
      "balance_loss_clip": 1.05500996,
      "balance_loss_mlp": 1.02390361,
      "epoch": 0.2900258522214874,
      "flos": 22893735202560.0,
      "grad_norm": 2.657662894469066,
      "language_loss": 0.81773156,
      "learning_rate": 3.3318424321290596e-06,
      "loss": 0.83957249,
      "num_input_tokens_seen": 51660975,
      "step": 2412,
      "time_per_iteration": 2.63137149810791
    },
    {
      "auxiliary_loss_clip": 0.01076642,
      "auxiliary_loss_mlp": 0.0100363,
      "balance_loss_clip": 1.03442025,
      "balance_loss_mlp": 1.00226521,
      "epoch": 0.2901460951121265,
      "flos": 71106036013440.0,
      "grad_norm": 0.8528914580462031,
      "language_loss": 0.59982955,
      "learning_rate": 3.3312612010805917e-06,
      "loss": 0.62063229,
      "num_input_tokens_seen": 51720550,
      "step": 2413,
      "time_per_iteration": 3.1731765270233154
    },
    {
      "auxiliary_loss_clip": 0.01162893,
      "auxiliary_loss_mlp": 0.01038415,
      "balance_loss_clip": 1.05737257,
      "balance_loss_mlp": 1.02894998,
      "epoch": 0.2902663380027656,
      "flos": 32160986081280.0,
      "grad_norm": 1.713133232171617,
      "language_loss": 0.70072764,
      "learning_rate": 3.330679768080858e-06,
      "loss": 0.72274065,
      "num_input_tokens_seen": 51744435,
      "step": 2414,
      "time_per_iteration": 2.6292827129364014
    },
    {
      "auxiliary_loss_clip": 0.01190576,
      "auxiliary_loss_mlp": 0.01034585,
      "balance_loss_clip": 1.06367862,
      "balance_loss_mlp": 1.02460098,
      "epoch": 0.2903865808934047,
      "flos": 29351658539520.0,
      "grad_norm": 2.024128948640041,
      "language_loss": 0.83503389,
      "learning_rate": 3.3300981332180627e-06,
      "loss": 0.8572855,
      "num_input_tokens_seen": 51763640,
      "step": 2415,
      "time_per_iteration": 2.5794622898101807
    },
    {
      "auxiliary_loss_clip": 0.01167465,
      "auxiliary_loss_mlp": 0.01028051,
      "balance_loss_clip": 1.05756402,
      "balance_loss_mlp": 1.01902103,
      "epoch": 0.29050682378404374,
      "flos": 17089026647040.0,
      "grad_norm": 2.137085589400161,
      "language_loss": 0.80126452,
      "learning_rate": 3.3295162965804373e-06,
      "loss": 0.82321966,
      "num_input_tokens_seen": 51782135,
      "step": 2416,
      "time_per_iteration": 2.5965566635131836
    },
    {
      "auxiliary_loss_clip": 0.01156201,
      "auxiliary_loss_mlp": 0.01027398,
      "balance_loss_clip": 1.05722833,
      "balance_loss_mlp": 1.01815963,
      "epoch": 0.29062706667468285,
      "flos": 17858233422720.0,
      "grad_norm": 2.0182079626067235,
      "language_loss": 0.78811514,
      "learning_rate": 3.328934258256247e-06,
      "loss": 0.80995113,
      "num_input_tokens_seen": 51800200,
      "step": 2417,
      "time_per_iteration": 3.513564348220825
    },
    {
      "auxiliary_loss_clip": 0.0119104,
      "auxiliary_loss_mlp": 0.01030553,
      "balance_loss_clip": 1.05934966,
      "balance_loss_mlp": 1.02039635,
      "epoch": 0.29074730956532197,
      "flos": 24279815174400.0,
      "grad_norm": 2.024376027443427,
      "language_loss": 0.67472589,
      "learning_rate": 3.3283520183337856e-06,
      "loss": 0.69694179,
      "num_input_tokens_seen": 51819905,
      "step": 2418,
      "time_per_iteration": 2.5593769550323486
    },
    {
      "auxiliary_loss_clip": 0.01172277,
      "auxiliary_loss_mlp": 0.01031434,
      "balance_loss_clip": 1.05761409,
      "balance_loss_mlp": 1.02201653,
      "epoch": 0.290867552455961,
      "flos": 22340961826560.0,
      "grad_norm": 1.5750965800784684,
      "language_loss": 0.68862599,
      "learning_rate": 3.3277695769013797e-06,
      "loss": 0.71066308,
      "num_input_tokens_seen": 51839350,
      "step": 2419,
      "time_per_iteration": 2.5420775413513184
    },
    {
      "auxiliary_loss_clip": 0.01194084,
      "auxiliary_loss_mlp": 0.01031481,
      "balance_loss_clip": 1.06259513,
      "balance_loss_mlp": 1.02207494,
      "epoch": 0.29098779534660013,
      "flos": 23186155824000.0,
      "grad_norm": 2.1426285722053944,
      "language_loss": 0.77466422,
      "learning_rate": 3.327186934047385e-06,
      "loss": 0.79691982,
      "num_input_tokens_seen": 51858045,
      "step": 2420,
      "time_per_iteration": 2.5408473014831543
    },
    {
      "auxiliary_loss_clip": 0.01163673,
      "auxiliary_loss_mlp": 0.01033997,
      "balance_loss_clip": 1.05411804,
      "balance_loss_mlp": 1.02470434,
      "epoch": 0.29110803823723924,
      "flos": 15304194817920.0,
      "grad_norm": 1.7916491406087527,
      "language_loss": 0.65689802,
      "learning_rate": 3.3266040898601877e-06,
      "loss": 0.67887473,
      "num_input_tokens_seen": 51875880,
      "step": 2421,
      "time_per_iteration": 2.517857313156128
    },
    {
      "auxiliary_loss_clip": 0.01140338,
      "auxiliary_loss_mlp": 0.01030631,
      "balance_loss_clip": 1.05222094,
      "balance_loss_mlp": 1.0214932,
      "epoch": 0.2912282811278783,
      "flos": 22595352923520.0,
      "grad_norm": 2.053182994260173,
      "language_loss": 0.77941227,
      "learning_rate": 3.3260210444282045e-06,
      "loss": 0.80112201,
      "num_input_tokens_seen": 51893835,
      "step": 2422,
      "time_per_iteration": 3.5465195178985596
    },
    {
      "auxiliary_loss_clip": 0.01186018,
      "auxiliary_loss_mlp": 0.01035057,
      "balance_loss_clip": 1.06199372,
      "balance_loss_mlp": 1.02566361,
      "epoch": 0.2913485240185174,
      "flos": 24497900599680.0,
      "grad_norm": 2.136328364839051,
      "language_loss": 0.73118502,
      "learning_rate": 3.325437797839883e-06,
      "loss": 0.75339574,
      "num_input_tokens_seen": 51912205,
      "step": 2423,
      "time_per_iteration": 3.264315128326416
    },
    {
      "auxiliary_loss_clip": 0.01204405,
      "auxiliary_loss_mlp": 0.01034084,
      "balance_loss_clip": 1.06104994,
      "balance_loss_mlp": 1.0247376,
      "epoch": 0.2914687669091565,
      "flos": 17931024334080.0,
      "grad_norm": 2.105803453504304,
      "language_loss": 0.74570858,
      "learning_rate": 3.3248543501837015e-06,
      "loss": 0.76809347,
      "num_input_tokens_seen": 51929410,
      "step": 2424,
      "time_per_iteration": 2.444683790206909
    },
    {
      "auxiliary_loss_clip": 0.0115476,
      "auxiliary_loss_mlp": 0.01033075,
      "balance_loss_clip": 1.05951738,
      "balance_loss_mlp": 1.0239495,
      "epoch": 0.2915890097997956,
      "flos": 22529313768960.0,
      "grad_norm": 1.7549213412504703,
      "language_loss": 0.77552181,
      "learning_rate": 3.3242707015481684e-06,
      "loss": 0.79740012,
      "num_input_tokens_seen": 51949345,
      "step": 2425,
      "time_per_iteration": 2.625549077987671
    },
    {
      "auxiliary_loss_clip": 0.01171152,
      "auxiliary_loss_mlp": 0.01030623,
      "balance_loss_clip": 1.05472851,
      "balance_loss_mlp": 1.02218342,
      "epoch": 0.2917092526904347,
      "flos": 13845216193920.0,
      "grad_norm": 1.637159052522053,
      "language_loss": 0.80268264,
      "learning_rate": 3.323686852021823e-06,
      "loss": 0.82470036,
      "num_input_tokens_seen": 51966855,
      "step": 2426,
      "time_per_iteration": 2.5675816535949707
    },
    {
      "auxiliary_loss_clip": 0.0116416,
      "auxiliary_loss_mlp": 0.01031326,
      "balance_loss_clip": 1.05450225,
      "balance_loss_mlp": 1.02187288,
      "epoch": 0.2918294955810738,
      "flos": 22674859678080.0,
      "grad_norm": 2.3121532436044405,
      "language_loss": 0.79587591,
      "learning_rate": 3.323102801693235e-06,
      "loss": 0.8178308,
      "num_input_tokens_seen": 51985620,
      "step": 2427,
      "time_per_iteration": 2.65246319770813
    },
    {
      "auxiliary_loss_clip": 0.01182783,
      "auxiliary_loss_mlp": 0.01030354,
      "balance_loss_clip": 1.05762768,
      "balance_loss_mlp": 1.02089453,
      "epoch": 0.29194973847171285,
      "flos": 23438284364160.0,
      "grad_norm": 2.4546048710733985,
      "language_loss": 0.80223823,
      "learning_rate": 3.322518550651003e-06,
      "loss": 0.82436955,
      "num_input_tokens_seen": 52004930,
      "step": 2428,
      "time_per_iteration": 2.5681416988372803
    },
    {
      "auxiliary_loss_clip": 0.01180822,
      "auxiliary_loss_mlp": 0.01035829,
      "balance_loss_clip": 1.05698776,
      "balance_loss_mlp": 1.02663839,
      "epoch": 0.29206998136235196,
      "flos": 21909064694400.0,
      "grad_norm": 2.137436896155721,
      "language_loss": 0.81402034,
      "learning_rate": 3.3219340989837586e-06,
      "loss": 0.83618689,
      "num_input_tokens_seen": 52024920,
      "step": 2429,
      "time_per_iteration": 2.661130666732788
    },
    {
      "auxiliary_loss_clip": 0.01174282,
      "auxiliary_loss_mlp": 0.0103343,
      "balance_loss_clip": 1.05909109,
      "balance_loss_mlp": 1.02479947,
      "epoch": 0.292190224252991,
      "flos": 23215925220480.0,
      "grad_norm": 1.9097660838340935,
      "language_loss": 0.80582821,
      "learning_rate": 3.3213494467801625e-06,
      "loss": 0.82790542,
      "num_input_tokens_seen": 52044095,
      "step": 2430,
      "time_per_iteration": 2.6785995960235596
    },
    {
      "auxiliary_loss_clip": 0.01113519,
      "auxiliary_loss_mlp": 0.01029596,
      "balance_loss_clip": 1.04592931,
      "balance_loss_mlp": 1.02052474,
      "epoch": 0.2923104671436301,
      "flos": 20740818752640.0,
      "grad_norm": 1.8227797343507777,
      "language_loss": 0.71225917,
      "learning_rate": 3.3207645941289063e-06,
      "loss": 0.73369032,
      "num_input_tokens_seen": 52062440,
      "step": 2431,
      "time_per_iteration": 2.6438469886779785
    },
    {
      "auxiliary_loss_clip": 0.01191784,
      "auxiliary_loss_mlp": 0.00763923,
      "balance_loss_clip": 1.06377077,
      "balance_loss_mlp": 1.0002867,
      "epoch": 0.29243071003426924,
      "flos": 35809114999680.0,
      "grad_norm": 1.7465996955748733,
      "language_loss": 0.80065852,
      "learning_rate": 3.320179541118711e-06,
      "loss": 0.82021558,
      "num_input_tokens_seen": 52084940,
      "step": 2432,
      "time_per_iteration": 2.6420559883117676
    },
    {
      "auxiliary_loss_clip": 0.01100604,
      "auxiliary_loss_mlp": 0.01003887,
      "balance_loss_clip": 1.03223062,
      "balance_loss_mlp": 1.00246227,
      "epoch": 0.2925509529249083,
      "flos": 58081598524800.0,
      "grad_norm": 1.0249214197562737,
      "language_loss": 0.60312903,
      "learning_rate": 3.3195942878383293e-06,
      "loss": 0.62417388,
      "num_input_tokens_seen": 52141040,
      "step": 2433,
      "time_per_iteration": 3.103043556213379
    },
    {
      "auxiliary_loss_clip": 0.01193019,
      "auxiliary_loss_mlp": 0.01031312,
      "balance_loss_clip": 1.06189704,
      "balance_loss_mlp": 1.02141118,
      "epoch": 0.2926711958155474,
      "flos": 21397122103680.0,
      "grad_norm": 1.9670036998358922,
      "language_loss": 0.77871996,
      "learning_rate": 3.319008834376543e-06,
      "loss": 0.80096328,
      "num_input_tokens_seen": 52160730,
      "step": 2434,
      "time_per_iteration": 2.486548662185669
    },
    {
      "auxiliary_loss_clip": 0.01164531,
      "auxiliary_loss_mlp": 0.01025148,
      "balance_loss_clip": 1.05157483,
      "balance_loss_mlp": 1.01607037,
      "epoch": 0.2927914387061865,
      "flos": 23185796688000.0,
      "grad_norm": 2.0954825512035,
      "language_loss": 0.88984632,
      "learning_rate": 3.3184231808221654e-06,
      "loss": 0.91174316,
      "num_input_tokens_seen": 52175055,
      "step": 2435,
      "time_per_iteration": 2.5504398345947266
    },
    {
      "auxiliary_loss_clip": 0.01164275,
      "auxiliary_loss_mlp": 0.01037997,
      "balance_loss_clip": 1.06017447,
      "balance_loss_mlp": 1.0284369,
      "epoch": 0.29291168159682557,
      "flos": 22455553190400.0,
      "grad_norm": 1.8407655570802917,
      "language_loss": 0.62755907,
      "learning_rate": 3.3178373272640394e-06,
      "loss": 0.64958179,
      "num_input_tokens_seen": 52194150,
      "step": 2436,
      "time_per_iteration": 2.5636203289031982
    },
    {
      "auxiliary_loss_clip": 0.01204314,
      "auxiliary_loss_mlp": 0.01036166,
      "balance_loss_clip": 1.06290472,
      "balance_loss_mlp": 1.02752304,
      "epoch": 0.2930319244874647,
      "flos": 21170632896000.0,
      "grad_norm": 2.177517658942623,
      "language_loss": 0.85170913,
      "learning_rate": 3.3172512737910387e-06,
      "loss": 0.87411386,
      "num_input_tokens_seen": 52211660,
      "step": 2437,
      "time_per_iteration": 2.451910972595215
    },
    {
      "auxiliary_loss_clip": 0.01190476,
      "auxiliary_loss_mlp": 0.01033878,
      "balance_loss_clip": 1.05854487,
      "balance_loss_mlp": 1.0245676,
      "epoch": 0.2931521673781038,
      "flos": 31357843931520.0,
      "grad_norm": 2.8979296588964254,
      "language_loss": 0.87840325,
      "learning_rate": 3.3166650204920674e-06,
      "loss": 0.90064681,
      "num_input_tokens_seen": 52232830,
      "step": 2438,
      "time_per_iteration": 2.5606577396392822
    },
    {
      "auxiliary_loss_clip": 0.0119167,
      "auxiliary_loss_mlp": 0.01033876,
      "balance_loss_clip": 1.06327248,
      "balance_loss_mlp": 1.02433634,
      "epoch": 0.29327241026874284,
      "flos": 24200990778240.0,
      "grad_norm": 1.5372924706964617,
      "language_loss": 0.81654668,
      "learning_rate": 3.316078567456059e-06,
      "loss": 0.83880216,
      "num_input_tokens_seen": 52250670,
      "step": 2439,
      "time_per_iteration": 2.5211644172668457
    },
    {
      "auxiliary_loss_clip": 0.01136472,
      "auxiliary_loss_mlp": 0.01028248,
      "balance_loss_clip": 1.05421662,
      "balance_loss_mlp": 1.02004027,
      "epoch": 0.29339265315938196,
      "flos": 24242611662720.0,
      "grad_norm": 2.1372627892072105,
      "language_loss": 0.75933146,
      "learning_rate": 3.3154919147719786e-06,
      "loss": 0.78097868,
      "num_input_tokens_seen": 52271685,
      "step": 2440,
      "time_per_iteration": 2.6627840995788574
    },
    {
      "auxiliary_loss_clip": 0.01190477,
      "auxiliary_loss_mlp": 0.01030369,
      "balance_loss_clip": 1.06052673,
      "balance_loss_mlp": 1.02110076,
      "epoch": 0.29351289605002107,
      "flos": 16946641134720.0,
      "grad_norm": 2.1196209954178333,
      "language_loss": 0.86603832,
      "learning_rate": 3.31490506252882e-06,
      "loss": 0.88824677,
      "num_input_tokens_seen": 52291065,
      "step": 2441,
      "time_per_iteration": 2.5005228519439697
    },
    {
      "auxiliary_loss_clip": 0.01147855,
      "auxiliary_loss_mlp": 0.01032235,
      "balance_loss_clip": 1.05032372,
      "balance_loss_mlp": 1.02397394,
      "epoch": 0.2936331389406601,
      "flos": 19829082810240.0,
      "grad_norm": 1.8255698780371186,
      "language_loss": 0.84314984,
      "learning_rate": 3.31431801081561e-06,
      "loss": 0.86495072,
      "num_input_tokens_seen": 52310000,
      "step": 2442,
      "time_per_iteration": 2.536606788635254
    },
    {
      "auxiliary_loss_clip": 0.01087172,
      "auxiliary_loss_mlp": 0.01011709,
      "balance_loss_clip": 1.03608942,
      "balance_loss_mlp": 1.0104872,
      "epoch": 0.29375338183129923,
      "flos": 71416844398080.0,
      "grad_norm": 0.8978043828204144,
      "language_loss": 0.67893684,
      "learning_rate": 3.313730759721402e-06,
      "loss": 0.69992566,
      "num_input_tokens_seen": 52372930,
      "step": 2443,
      "time_per_iteration": 3.2036445140838623
    },
    {
      "auxiliary_loss_clip": 0.01172636,
      "auxiliary_loss_mlp": 0.01035066,
      "balance_loss_clip": 1.05867529,
      "balance_loss_mlp": 1.02524877,
      "epoch": 0.29387362472193834,
      "flos": 22054502862720.0,
      "grad_norm": 1.9646828932350573,
      "language_loss": 0.86361015,
      "learning_rate": 3.313143309335282e-06,
      "loss": 0.88568711,
      "num_input_tokens_seen": 52391420,
      "step": 2444,
      "time_per_iteration": 3.3804564476013184
    },
    {
      "auxiliary_loss_clip": 0.01158347,
      "auxiliary_loss_mlp": 0.01034168,
      "balance_loss_clip": 1.05759907,
      "balance_loss_mlp": 1.02468514,
      "epoch": 0.2939938676125774,
      "flos": 22966418373120.0,
      "grad_norm": 1.9760051808641652,
      "language_loss": 0.84948897,
      "learning_rate": 3.3125556597463665e-06,
      "loss": 0.87141418,
      "num_input_tokens_seen": 52410725,
      "step": 2445,
      "time_per_iteration": 2.570573568344116
    },
    {
      "auxiliary_loss_clip": 0.01187592,
      "auxiliary_loss_mlp": 0.01033759,
      "balance_loss_clip": 1.06216717,
      "balance_loss_mlp": 1.02493179,
      "epoch": 0.2941141105032165,
      "flos": 31358705857920.0,
      "grad_norm": 1.567619141861922,
      "language_loss": 0.66339487,
      "learning_rate": 3.311967811043801e-06,
      "loss": 0.68560839,
      "num_input_tokens_seen": 52432645,
      "step": 2446,
      "time_per_iteration": 2.623185873031616
    },
    {
      "auxiliary_loss_clip": 0.01190521,
      "auxiliary_loss_mlp": 0.01031455,
      "balance_loss_clip": 1.06228948,
      "balance_loss_mlp": 1.0221504,
      "epoch": 0.29423435339385556,
      "flos": 23222138273280.0,
      "grad_norm": 2.5557226140536344,
      "language_loss": 0.82255226,
      "learning_rate": 3.3113797633167617e-06,
      "loss": 0.84477198,
      "num_input_tokens_seen": 52450940,
      "step": 2447,
      "time_per_iteration": 2.616151809692383
    },
    {
      "auxiliary_loss_clip": 0.01203828,
      "auxiliary_loss_mlp": 0.01029877,
      "balance_loss_clip": 1.06127059,
      "balance_loss_mlp": 1.02025044,
      "epoch": 0.2943545962844947,
      "flos": 26864054138880.0,
      "grad_norm": 2.116558527049001,
      "language_loss": 0.68866307,
      "learning_rate": 3.310791516654455e-06,
      "loss": 0.71100014,
      "num_input_tokens_seen": 52468000,
      "step": 2448,
      "time_per_iteration": 3.415570020675659
    },
    {
      "auxiliary_loss_clip": 0.01169073,
      "auxiliary_loss_mlp": 0.01042707,
      "balance_loss_clip": 1.05729854,
      "balance_loss_mlp": 1.03240764,
      "epoch": 0.2944748391751338,
      "flos": 20231677422720.0,
      "grad_norm": 1.825830585048486,
      "language_loss": 0.79675436,
      "learning_rate": 3.3102030711461177e-06,
      "loss": 0.81887221,
      "num_input_tokens_seen": 52487575,
      "step": 2449,
      "time_per_iteration": 4.269005537033081
    },
    {
      "auxiliary_loss_clip": 0.01164726,
      "auxiliary_loss_mlp": 0.01024555,
      "balance_loss_clip": 1.05778742,
      "balance_loss_mlp": 1.01538229,
      "epoch": 0.29459508206577284,
      "flos": 15960965045760.0,
      "grad_norm": 1.8538415440303369,
      "language_loss": 0.67971587,
      "learning_rate": 3.3096144268810156e-06,
      "loss": 0.70160866,
      "num_input_tokens_seen": 52506335,
      "step": 2450,
      "time_per_iteration": 2.6147491931915283
    },
    {
      "auxiliary_loss_clip": 0.01179106,
      "auxiliary_loss_mlp": 0.01031909,
      "balance_loss_clip": 1.05796123,
      "balance_loss_mlp": 1.02250957,
      "epoch": 0.29471532495641195,
      "flos": 20412882558720.0,
      "grad_norm": 1.9659645250214002,
      "language_loss": 0.7295301,
      "learning_rate": 3.3090255839484462e-06,
      "loss": 0.75164026,
      "num_input_tokens_seen": 52524330,
      "step": 2451,
      "time_per_iteration": 2.6420652866363525
    },
    {
      "auxiliary_loss_clip": 0.01176882,
      "auxiliary_loss_mlp": 0.01030402,
      "balance_loss_clip": 1.05749226,
      "balance_loss_mlp": 1.02075231,
      "epoch": 0.29483556784705106,
      "flos": 20376576887040.0,
      "grad_norm": 1.925991110262362,
      "language_loss": 0.85528541,
      "learning_rate": 3.3084365424377366e-06,
      "loss": 0.8773582,
      "num_input_tokens_seen": 52543095,
      "step": 2452,
      "time_per_iteration": 2.5142624378204346
    },
    {
      "auxiliary_loss_clip": 0.01073416,
      "auxiliary_loss_mlp": 0.01008848,
      "balance_loss_clip": 1.04353976,
      "balance_loss_mlp": 1.00715494,
      "epoch": 0.2949558107376901,
      "flos": 68555660595840.0,
      "grad_norm": 0.72802823671306,
      "language_loss": 0.5594064,
      "learning_rate": 3.307847302438245e-06,
      "loss": 0.58022904,
      "num_input_tokens_seen": 52597075,
      "step": 2453,
      "time_per_iteration": 3.0465087890625
    },
    {
      "auxiliary_loss_clip": 0.01128962,
      "auxiliary_loss_mlp": 0.01029565,
      "balance_loss_clip": 1.04890275,
      "balance_loss_mlp": 1.01953936,
      "epoch": 0.2950760536283292,
      "flos": 16107085572480.0,
      "grad_norm": 2.246497937726557,
      "language_loss": 0.78037846,
      "learning_rate": 3.3072578640393562e-06,
      "loss": 0.80196381,
      "num_input_tokens_seen": 52614410,
      "step": 2454,
      "time_per_iteration": 2.612377405166626
    },
    {
      "auxiliary_loss_clip": 0.01174708,
      "auxiliary_loss_mlp": 0.01030006,
      "balance_loss_clip": 1.05923843,
      "balance_loss_mlp": 1.02053535,
      "epoch": 0.29519629651896834,
      "flos": 20483626394880.0,
      "grad_norm": 1.8481795362133178,
      "language_loss": 0.79444516,
      "learning_rate": 3.3066682273304886e-06,
      "loss": 0.81649226,
      "num_input_tokens_seen": 52632055,
      "step": 2455,
      "time_per_iteration": 2.5631654262542725
    },
    {
      "auxiliary_loss_clip": 0.01195319,
      "auxiliary_loss_mlp": 0.0076407,
      "balance_loss_clip": 1.06213617,
      "balance_loss_mlp": 1.00032496,
      "epoch": 0.2953165394096074,
      "flos": 18916484941440.0,
      "grad_norm": 1.9910736544904315,
      "language_loss": 0.78806609,
      "learning_rate": 3.3060783924010904e-06,
      "loss": 0.80765998,
      "num_input_tokens_seen": 52649980,
      "step": 2456,
      "time_per_iteration": 2.5193769931793213
    },
    {
      "auxiliary_loss_clip": 0.01163171,
      "auxiliary_loss_mlp": 0.01034789,
      "balance_loss_clip": 1.05919313,
      "balance_loss_mlp": 1.02501416,
      "epoch": 0.2954367823002465,
      "flos": 20624467622400.0,
      "grad_norm": 2.471307453115101,
      "language_loss": 0.85131407,
      "learning_rate": 3.3054883593406387e-06,
      "loss": 0.8732937,
      "num_input_tokens_seen": 52664730,
      "step": 2457,
      "time_per_iteration": 2.5412800312042236
    },
    {
      "auxiliary_loss_clip": 0.01176856,
      "auxiliary_loss_mlp": 0.01034044,
      "balance_loss_clip": 1.05884945,
      "balance_loss_mlp": 1.02504992,
      "epoch": 0.2955570251908856,
      "flos": 31175525473920.0,
      "grad_norm": 2.1421960615479567,
      "language_loss": 0.65481204,
      "learning_rate": 3.3048981282386404e-06,
      "loss": 0.67692101,
      "num_input_tokens_seen": 52686040,
      "step": 2458,
      "time_per_iteration": 2.6395134925842285
    },
    {
      "auxiliary_loss_clip": 0.0114987,
      "auxiliary_loss_mlp": 0.01029788,
      "balance_loss_clip": 1.05724669,
      "balance_loss_mlp": 1.02119303,
      "epoch": 0.29567726808152467,
      "flos": 21650328051840.0,
      "grad_norm": 1.934980542574379,
      "language_loss": 0.82540143,
      "learning_rate": 3.304307699184634e-06,
      "loss": 0.84719801,
      "num_input_tokens_seen": 52704630,
      "step": 2459,
      "time_per_iteration": 2.562892436981201
    },
    {
      "auxiliary_loss_clip": 0.01175545,
      "auxiliary_loss_mlp": 0.01034541,
      "balance_loss_clip": 1.06029212,
      "balance_loss_mlp": 1.02565408,
      "epoch": 0.2957975109721638,
      "flos": 24243868638720.0,
      "grad_norm": 1.7621834198712325,
      "language_loss": 0.78859651,
      "learning_rate": 3.3037170722681866e-06,
      "loss": 0.81069744,
      "num_input_tokens_seen": 52725465,
      "step": 2460,
      "time_per_iteration": 2.54333758354187
    },
    {
      "auxiliary_loss_clip": 0.01156518,
      "auxiliary_loss_mlp": 0.01032624,
      "balance_loss_clip": 1.0592941,
      "balance_loss_mlp": 1.02322483,
      "epoch": 0.29591775386280283,
      "flos": 13479717352320.0,
      "grad_norm": 1.761281639389518,
      "language_loss": 0.68101501,
      "learning_rate": 3.3031262475788956e-06,
      "loss": 0.70290643,
      "num_input_tokens_seen": 52742405,
      "step": 2461,
      "time_per_iteration": 2.526418447494507
    },
    {
      "auxiliary_loss_clip": 0.01173169,
      "auxiliary_loss_mlp": 0.01032337,
      "balance_loss_clip": 1.05831647,
      "balance_loss_mlp": 1.02332461,
      "epoch": 0.29603799675344195,
      "flos": 17749783284480.0,
      "grad_norm": 1.7226588489405246,
      "language_loss": 0.73160857,
      "learning_rate": 3.3025352252063897e-06,
      "loss": 0.7536636,
      "num_input_tokens_seen": 52761100,
      "step": 2462,
      "time_per_iteration": 2.5071654319763184
    },
    {
      "auxiliary_loss_clip": 0.0118954,
      "auxiliary_loss_mlp": 0.01038403,
      "balance_loss_clip": 1.06385076,
      "balance_loss_mlp": 1.02912831,
      "epoch": 0.29615823964408106,
      "flos": 22783920347520.0,
      "grad_norm": 1.9316008937784697,
      "language_loss": 0.7524932,
      "learning_rate": 3.3019440052403252e-06,
      "loss": 0.77477264,
      "num_input_tokens_seen": 52780965,
      "step": 2463,
      "time_per_iteration": 2.5210492610931396
    },
    {
      "auxiliary_loss_clip": 0.01176218,
      "auxiliary_loss_mlp": 0.01030969,
      "balance_loss_clip": 1.05787027,
      "balance_loss_mlp": 1.021909,
      "epoch": 0.2962784825347201,
      "flos": 23514199758720.0,
      "grad_norm": 1.9874792727442891,
      "language_loss": 0.70831293,
      "learning_rate": 3.30135258777039e-06,
      "loss": 0.73038483,
      "num_input_tokens_seen": 52800335,
      "step": 2464,
      "time_per_iteration": 2.548659086227417
    },
    {
      "auxiliary_loss_clip": 0.01194284,
      "auxiliary_loss_mlp": 0.00764029,
      "balance_loss_clip": 1.06067729,
      "balance_loss_mlp": 1.00035286,
      "epoch": 0.2963987254253592,
      "flos": 16362769559040.0,
      "grad_norm": 2.0135302193605282,
      "language_loss": 0.70604622,
      "learning_rate": 3.3007609728863024e-06,
      "loss": 0.72562933,
      "num_input_tokens_seen": 52818425,
      "step": 2465,
      "time_per_iteration": 2.4830453395843506
    },
    {
      "auxiliary_loss_clip": 0.01124545,
      "auxiliary_loss_mlp": 0.01029915,
      "balance_loss_clip": 1.05726361,
      "balance_loss_mlp": 1.02039647,
      "epoch": 0.29651896831599833,
      "flos": 33472263980160.0,
      "grad_norm": 2.0939213696398236,
      "language_loss": 0.7299872,
      "learning_rate": 3.300169160677809e-06,
      "loss": 0.75153184,
      "num_input_tokens_seen": 52842340,
      "step": 2466,
      "time_per_iteration": 2.7332284450531006
    },
    {
      "auxiliary_loss_clip": 0.01168318,
      "auxiliary_loss_mlp": 0.01027985,
      "balance_loss_clip": 1.05975151,
      "balance_loss_mlp": 1.018466,
      "epoch": 0.2966392112066374,
      "flos": 23805363404160.0,
      "grad_norm": 2.252350045765582,
      "language_loss": 0.7770915,
      "learning_rate": 3.2995771512346878e-06,
      "loss": 0.7990545,
      "num_input_tokens_seen": 52860690,
      "step": 2467,
      "time_per_iteration": 2.5818231105804443
    },
    {
      "auxiliary_loss_clip": 0.01208669,
      "auxiliary_loss_mlp": 0.00764232,
      "balance_loss_clip": 1.06461656,
      "balance_loss_mlp": 1.00033641,
      "epoch": 0.2967594540972765,
      "flos": 19938466702080.0,
      "grad_norm": 9.878362557482154,
      "language_loss": 0.73227632,
      "learning_rate": 3.298984944646746e-06,
      "loss": 0.7520054,
      "num_input_tokens_seen": 52879370,
      "step": 2468,
      "time_per_iteration": 2.4837563037872314
    },
    {
      "auxiliary_loss_clip": 0.01193126,
      "auxiliary_loss_mlp": 0.0076337,
      "balance_loss_clip": 1.0625639,
      "balance_loss_mlp": 1.00027835,
      "epoch": 0.2968796969879156,
      "flos": 23732823888000.0,
      "grad_norm": 1.8758742520349645,
      "language_loss": 0.81659961,
      "learning_rate": 3.298392541003822e-06,
      "loss": 0.83616459,
      "num_input_tokens_seen": 52898775,
      "step": 2469,
      "time_per_iteration": 2.5382080078125
    },
    {
      "auxiliary_loss_clip": 0.01174975,
      "auxiliary_loss_mlp": 0.0103081,
      "balance_loss_clip": 1.06108212,
      "balance_loss_mlp": 1.02200603,
      "epoch": 0.29699993987855466,
      "flos": 22893699288960.0,
      "grad_norm": 1.6177336234424458,
      "language_loss": 0.89689589,
      "learning_rate": 3.2977999403957806e-06,
      "loss": 0.91895366,
      "num_input_tokens_seen": 52917535,
      "step": 2470,
      "time_per_iteration": 3.4227635860443115
    },
    {
      "auxiliary_loss_clip": 0.01206573,
      "auxiliary_loss_mlp": 0.01037172,
      "balance_loss_clip": 1.06557202,
      "balance_loss_mlp": 1.02745628,
      "epoch": 0.2971201827691938,
      "flos": 33832555349760.0,
      "grad_norm": 2.9443293596005033,
      "language_loss": 0.66731179,
      "learning_rate": 3.2972071429125207e-06,
      "loss": 0.68974924,
      "num_input_tokens_seen": 52938755,
      "step": 2471,
      "time_per_iteration": 2.5778756141662598
    },
    {
      "auxiliary_loss_clip": 0.01155426,
      "auxiliary_loss_mlp": 0.01031614,
      "balance_loss_clip": 1.05770791,
      "balance_loss_mlp": 1.02232766,
      "epoch": 0.2972404256598329,
      "flos": 22054359208320.0,
      "grad_norm": 1.9541774829106588,
      "language_loss": 0.8816393,
      "learning_rate": 3.2966141486439682e-06,
      "loss": 0.90350974,
      "num_input_tokens_seen": 52957945,
      "step": 2472,
      "time_per_iteration": 2.54937744140625
    },
    {
      "auxiliary_loss_clip": 0.01134126,
      "auxiliary_loss_mlp": 0.0103045,
      "balance_loss_clip": 1.05253124,
      "balance_loss_mlp": 1.02101469,
      "epoch": 0.29736066855047194,
      "flos": 31978595796480.0,
      "grad_norm": 2.4185275375393225,
      "language_loss": 0.64247286,
      "learning_rate": 3.29602095768008e-06,
      "loss": 0.66411865,
      "num_input_tokens_seen": 52978460,
      "step": 2473,
      "time_per_iteration": 2.736271381378174
    },
    {
      "auxiliary_loss_clip": 0.01169971,
      "auxiliary_loss_mlp": 0.0103006,
      "balance_loss_clip": 1.0612036,
      "balance_loss_mlp": 1.02129173,
      "epoch": 0.29748091144111105,
      "flos": 33510401245440.0,
      "grad_norm": 1.8397306232531603,
      "language_loss": 0.63817883,
      "learning_rate": 3.2954275701108437e-06,
      "loss": 0.66017914,
      "num_input_tokens_seen": 52999640,
      "step": 2474,
      "time_per_iteration": 2.6391453742980957
    },
    {
      "auxiliary_loss_clip": 0.01142061,
      "auxiliary_loss_mlp": 0.01026019,
      "balance_loss_clip": 1.05406857,
      "balance_loss_mlp": 1.01678634,
      "epoch": 0.29760115433175016,
      "flos": 41283373409280.0,
      "grad_norm": 2.338121174326109,
      "language_loss": 0.68623805,
      "learning_rate": 3.294833986026275e-06,
      "loss": 0.70791882,
      "num_input_tokens_seen": 53022880,
      "step": 2475,
      "time_per_iteration": 3.6481127738952637
    },
    {
      "auxiliary_loss_clip": 0.01153067,
      "auxiliary_loss_mlp": 0.01025789,
      "balance_loss_clip": 1.05640996,
      "balance_loss_mlp": 1.01688433,
      "epoch": 0.2977213972223892,
      "flos": 24493339572480.0,
      "grad_norm": 2.0947412880151806,
      "language_loss": 0.85193896,
      "learning_rate": 3.29424020551642e-06,
      "loss": 0.87372756,
      "num_input_tokens_seen": 53041515,
      "step": 2476,
      "time_per_iteration": 4.150704622268677
    },
    {
      "auxiliary_loss_clip": 0.01210018,
      "auxiliary_loss_mlp": 0.01035184,
      "balance_loss_clip": 1.06459832,
      "balance_loss_mlp": 1.02477074,
      "epoch": 0.2978416401130283,
      "flos": 21285116519040.0,
      "grad_norm": 1.978533878096831,
      "language_loss": 0.72296906,
      "learning_rate": 3.2936462286713546e-06,
      "loss": 0.74542105,
      "num_input_tokens_seen": 53059865,
      "step": 2477,
      "time_per_iteration": 2.480623245239258
    },
    {
      "auxiliary_loss_clip": 0.01191504,
      "auxiliary_loss_mlp": 0.00764295,
      "balance_loss_clip": 1.06174481,
      "balance_loss_mlp": 1.00030088,
      "epoch": 0.2979618830036674,
      "flos": 25772154554880.0,
      "grad_norm": 1.996844815724552,
      "language_loss": 0.77014327,
      "learning_rate": 3.2930520555811846e-06,
      "loss": 0.78970122,
      "num_input_tokens_seen": 53079490,
      "step": 2478,
      "time_per_iteration": 2.5560431480407715
    },
    {
      "auxiliary_loss_clip": 0.01095932,
      "auxiliary_loss_mlp": 0.01033001,
      "balance_loss_clip": 1.04948032,
      "balance_loss_mlp": 1.02344012,
      "epoch": 0.2980821258943065,
      "flos": 23476996247040.0,
      "grad_norm": 1.6958406021904748,
      "language_loss": 0.80130553,
      "learning_rate": 3.292457686336046e-06,
      "loss": 0.82259488,
      "num_input_tokens_seen": 53098810,
      "step": 2479,
      "time_per_iteration": 2.7030694484710693
    },
    {
      "auxiliary_loss_clip": 0.01091854,
      "auxiliary_loss_mlp": 0.01009119,
      "balance_loss_clip": 1.03518629,
      "balance_loss_mlp": 1.00764132,
      "epoch": 0.2982023687849456,
      "flos": 69752314195200.0,
      "grad_norm": 0.874127644883929,
      "language_loss": 0.61287028,
      "learning_rate": 3.291863121026105e-06,
      "loss": 0.63388002,
      "num_input_tokens_seen": 53162590,
      "step": 2480,
      "time_per_iteration": 3.1950089931488037
    },
    {
      "auxiliary_loss_clip": 0.01191739,
      "auxiliary_loss_mlp": 0.01033224,
      "balance_loss_clip": 1.06191647,
      "balance_loss_mlp": 1.02424145,
      "epoch": 0.29832261167558466,
      "flos": 29825930741760.0,
      "grad_norm": 1.8341941873830745,
      "language_loss": 0.76604187,
      "learning_rate": 3.2912683597415547e-06,
      "loss": 0.78829145,
      "num_input_tokens_seen": 53186675,
      "step": 2481,
      "time_per_iteration": 2.6439967155456543
    },
    {
      "auxiliary_loss_clip": 0.011644,
      "auxiliary_loss_mlp": 0.01035994,
      "balance_loss_clip": 1.05876565,
      "balance_loss_mlp": 1.02667737,
      "epoch": 0.29844285456622377,
      "flos": 33910158683520.0,
      "grad_norm": 2.051309937754646,
      "language_loss": 0.78282851,
      "learning_rate": 3.2906734025726213e-06,
      "loss": 0.80483246,
      "num_input_tokens_seen": 53205940,
      "step": 2482,
      "time_per_iteration": 2.6849558353424072
    },
    {
      "auxiliary_loss_clip": 0.01198728,
      "auxiliary_loss_mlp": 0.01039036,
      "balance_loss_clip": 1.06457233,
      "balance_loss_mlp": 1.02949929,
      "epoch": 0.2985630974568629,
      "flos": 23876933253120.0,
      "grad_norm": 1.7798961196721699,
      "language_loss": 0.88462925,
      "learning_rate": 3.290078249609559e-06,
      "loss": 0.90700692,
      "num_input_tokens_seen": 53225360,
      "step": 2483,
      "time_per_iteration": 2.5498769283294678
    },
    {
      "auxiliary_loss_clip": 0.01188429,
      "auxiliary_loss_mlp": 0.01034313,
      "balance_loss_clip": 1.06426525,
      "balance_loss_mlp": 1.02515769,
      "epoch": 0.29868334034750194,
      "flos": 21799106184960.0,
      "grad_norm": 1.8362350643667995,
      "language_loss": 0.87950534,
      "learning_rate": 3.2894829009426514e-06,
      "loss": 0.9017328,
      "num_input_tokens_seen": 53243195,
      "step": 2484,
      "time_per_iteration": 2.5167906284332275
    },
    {
      "auxiliary_loss_clip": 0.01188402,
      "auxiliary_loss_mlp": 0.01033614,
      "balance_loss_clip": 1.06123793,
      "balance_loss_mlp": 1.02445829,
      "epoch": 0.29880358323814105,
      "flos": 25666649331840.0,
      "grad_norm": 1.8898556178261638,
      "language_loss": 0.77748764,
      "learning_rate": 3.288887356662213e-06,
      "loss": 0.79970771,
      "num_input_tokens_seen": 53264530,
      "step": 2485,
      "time_per_iteration": 2.5755698680877686
    },
    {
      "auxiliary_loss_clip": 0.01094887,
      "auxiliary_loss_mlp": 0.01006101,
      "balance_loss_clip": 1.03364277,
      "balance_loss_mlp": 1.00483787,
      "epoch": 0.29892382612878016,
      "flos": 71005846003200.0,
      "grad_norm": 0.7694142112376068,
      "language_loss": 0.59711593,
      "learning_rate": 3.288291616858588e-06,
      "loss": 0.6181258,
      "num_input_tokens_seen": 53319920,
      "step": 2486,
      "time_per_iteration": 2.9750685691833496
    },
    {
      "auxiliary_loss_clip": 0.01141358,
      "auxiliary_loss_mlp": 0.01027051,
      "balance_loss_clip": 1.05825543,
      "balance_loss_mlp": 1.01824117,
      "epoch": 0.2990440690194192,
      "flos": 25481134563840.0,
      "grad_norm": 1.6290586183661782,
      "language_loss": 0.7689122,
      "learning_rate": 3.287695681622149e-06,
      "loss": 0.79059625,
      "num_input_tokens_seen": 53339270,
      "step": 2487,
      "time_per_iteration": 2.6511435508728027
    },
    {
      "auxiliary_loss_clip": 0.01179766,
      "auxiliary_loss_mlp": 0.01027272,
      "balance_loss_clip": 1.05895138,
      "balance_loss_mlp": 1.01882613,
      "epoch": 0.2991643119100583,
      "flos": 23732357011200.0,
      "grad_norm": 1.9882470204192673,
      "language_loss": 0.81252182,
      "learning_rate": 3.2870995510432982e-06,
      "loss": 0.83459222,
      "num_input_tokens_seen": 53357750,
      "step": 2488,
      "time_per_iteration": 2.5806496143341064
    },
    {
      "auxiliary_loss_clip": 0.01182326,
      "auxiliary_loss_mlp": 0.01035152,
      "balance_loss_clip": 1.05971909,
      "balance_loss_mlp": 1.02672434,
      "epoch": 0.29928455480069743,
      "flos": 27417545786880.0,
      "grad_norm": 1.7434606052828534,
      "language_loss": 0.76765078,
      "learning_rate": 3.2865032252124697e-06,
      "loss": 0.78982556,
      "num_input_tokens_seen": 53378265,
      "step": 2489,
      "time_per_iteration": 2.5856635570526123
    },
    {
      "auxiliary_loss_clip": 0.01174397,
      "auxiliary_loss_mlp": 0.01035883,
      "balance_loss_clip": 1.0585072,
      "balance_loss_mlp": 1.02685928,
      "epoch": 0.2994047976913365,
      "flos": 33692935184640.0,
      "grad_norm": 1.3682145206630378,
      "language_loss": 0.77613735,
      "learning_rate": 3.2859067042201243e-06,
      "loss": 0.79824018,
      "num_input_tokens_seen": 53400305,
      "step": 2490,
      "time_per_iteration": 2.6538589000701904
    },
    {
      "auxiliary_loss_clip": 0.01110008,
      "auxiliary_loss_mlp": 0.01033365,
      "balance_loss_clip": 1.05195951,
      "balance_loss_mlp": 1.02441239,
      "epoch": 0.2995250405819756,
      "flos": 16763963541120.0,
      "grad_norm": 1.9833230117530085,
      "language_loss": 0.77764446,
      "learning_rate": 3.2853099881567544e-06,
      "loss": 0.79907817,
      "num_input_tokens_seen": 53418705,
      "step": 2491,
      "time_per_iteration": 2.609543561935425
    },
    {
      "auxiliary_loss_clip": 0.01202512,
      "auxiliary_loss_mlp": 0.01034757,
      "balance_loss_clip": 1.06415343,
      "balance_loss_mlp": 1.0265789,
      "epoch": 0.29964528347261465,
      "flos": 22963976248320.0,
      "grad_norm": 2.097809545643159,
      "language_loss": 0.79014528,
      "learning_rate": 3.284713077112881e-06,
      "loss": 0.81251794,
      "num_input_tokens_seen": 53438135,
      "step": 2492,
      "time_per_iteration": 2.5228989124298096
    },
    {
      "auxiliary_loss_clip": 0.01169831,
      "auxiliary_loss_mlp": 0.01032711,
      "balance_loss_clip": 1.06279361,
      "balance_loss_mlp": 1.02304316,
      "epoch": 0.29976552636325376,
      "flos": 16938021870720.0,
      "grad_norm": 3.0825661009400624,
      "language_loss": 0.86485302,
      "learning_rate": 3.284115971179056e-06,
      "loss": 0.88687849,
      "num_input_tokens_seen": 53452165,
      "step": 2493,
      "time_per_iteration": 2.545217514038086
    },
    {
      "auxiliary_loss_clip": 0.01136519,
      "auxiliary_loss_mlp": 0.01033522,
      "balance_loss_clip": 1.05784714,
      "balance_loss_mlp": 1.02459359,
      "epoch": 0.2998857692538929,
      "flos": 17056455989760.0,
      "grad_norm": 1.8039814319555105,
      "language_loss": 0.78785151,
      "learning_rate": 3.283518670445859e-06,
      "loss": 0.80955195,
      "num_input_tokens_seen": 53470075,
      "step": 2494,
      "time_per_iteration": 2.604356050491333
    },
    {
      "auxiliary_loss_clip": 0.01074994,
      "auxiliary_loss_mlp": 0.00753826,
      "balance_loss_clip": 1.02911782,
      "balance_loss_mlp": 1.00020719,
      "epoch": 0.30000601214453193,
      "flos": 68831528025600.0,
      "grad_norm": 0.6880120674253565,
      "language_loss": 0.54308426,
      "learning_rate": 3.2829211750038995e-06,
      "loss": 0.56137252,
      "num_input_tokens_seen": 53538705,
      "step": 2495,
      "time_per_iteration": 3.1627235412597656
    },
    {
      "auxiliary_loss_clip": 0.01156512,
      "auxiliary_loss_mlp": 0.01033936,
      "balance_loss_clip": 1.05592608,
      "balance_loss_mlp": 1.02512026,
      "epoch": 0.30012625503517104,
      "flos": 17603267708160.0,
      "grad_norm": 1.9796274089981198,
      "language_loss": 0.89079237,
      "learning_rate": 3.2823234849438183e-06,
      "loss": 0.91269684,
      "num_input_tokens_seen": 53556740,
      "step": 2496,
      "time_per_iteration": 2.5424180030822754
    },
    {
      "auxiliary_loss_clip": 0.01177575,
      "auxiliary_loss_mlp": 0.01032121,
      "balance_loss_clip": 1.06031883,
      "balance_loss_mlp": 1.02330506,
      "epoch": 0.30024649792581015,
      "flos": 21252581775360.0,
      "grad_norm": 1.9246867939105723,
      "language_loss": 0.75855315,
      "learning_rate": 3.2817256003562836e-06,
      "loss": 0.78065014,
      "num_input_tokens_seen": 53577115,
      "step": 2497,
      "time_per_iteration": 3.4003679752349854
    },
    {
      "auxiliary_loss_clip": 0.01134962,
      "auxiliary_loss_mlp": 0.01040467,
      "balance_loss_clip": 1.05717862,
      "balance_loss_mlp": 1.03079939,
      "epoch": 0.3003667408164492,
      "flos": 23003262748800.0,
      "grad_norm": 1.7296803073515479,
      "language_loss": 0.66278738,
      "learning_rate": 3.281127521331995e-06,
      "loss": 0.6845417,
      "num_input_tokens_seen": 53598295,
      "step": 2498,
      "time_per_iteration": 2.653770923614502
    },
    {
      "auxiliary_loss_clip": 0.01103264,
      "auxiliary_loss_mlp": 0.01002994,
      "balance_loss_clip": 1.02895892,
      "balance_loss_mlp": 1.00180769,
      "epoch": 0.3004869837070883,
      "flos": 64232340750720.0,
      "grad_norm": 0.8835941656506333,
      "language_loss": 0.60625434,
      "learning_rate": 3.2805292479616798e-06,
      "loss": 0.62731689,
      "num_input_tokens_seen": 53657160,
      "step": 2499,
      "time_per_iteration": 2.97001314163208
    },
    {
      "auxiliary_loss_clip": 0.01177163,
      "auxiliary_loss_mlp": 0.01030096,
      "balance_loss_clip": 1.05830836,
      "balance_loss_mlp": 1.0209167,
      "epoch": 0.30060722659772743,
      "flos": 26248653400320.0,
      "grad_norm": 2.2195220560177997,
      "language_loss": 0.91660655,
      "learning_rate": 3.2799307803360955e-06,
      "loss": 0.9386791,
      "num_input_tokens_seen": 53673090,
      "step": 2500,
      "time_per_iteration": 2.6368207931518555
    },
    {
      "auxiliary_loss_clip": 0.01199409,
      "auxiliary_loss_mlp": 0.01033264,
      "balance_loss_clip": 1.06164384,
      "balance_loss_mlp": 1.02465725,
      "epoch": 0.3007274694883665,
      "flos": 24970879912320.0,
      "grad_norm": 1.3896059002488632,
      "language_loss": 0.8169753,
      "learning_rate": 3.27933211854603e-06,
      "loss": 0.83930206,
      "num_input_tokens_seen": 53692145,
      "step": 2501,
      "time_per_iteration": 3.4357757568359375
    },
    {
      "auxiliary_loss_clip": 0.01176098,
      "auxiliary_loss_mlp": 0.01030171,
      "balance_loss_clip": 1.0614953,
      "balance_loss_mlp": 1.02126038,
      "epoch": 0.3008477123790056,
      "flos": 17055845458560.0,
      "grad_norm": 1.5334444339985713,
      "language_loss": 0.8710283,
      "learning_rate": 3.278733262682299e-06,
      "loss": 0.89309108,
      "num_input_tokens_seen": 53710000,
      "step": 2502,
      "time_per_iteration": 2.5217535495758057
    },
    {
      "auxiliary_loss_clip": 0.01202946,
      "auxiliary_loss_mlp": 0.01029761,
      "balance_loss_clip": 1.06204009,
      "balance_loss_mlp": 1.02104104,
      "epoch": 0.3009679552696447,
      "flos": 21506398254720.0,
      "grad_norm": 2.1236831730283776,
      "language_loss": 0.82534826,
      "learning_rate": 3.2781342128357484e-06,
      "loss": 0.84767532,
      "num_input_tokens_seen": 53729355,
      "step": 2503,
      "time_per_iteration": 4.005939960479736
    },
    {
      "auxiliary_loss_clip": 0.01159268,
      "auxiliary_loss_mlp": 0.01032485,
      "balance_loss_clip": 1.05540967,
      "balance_loss_mlp": 1.02297807,
      "epoch": 0.30108819816028376,
      "flos": 21134004001920.0,
      "grad_norm": 2.192504855647531,
      "language_loss": 0.80452847,
      "learning_rate": 3.2775349690972547e-06,
      "loss": 0.826446,
      "num_input_tokens_seen": 53743505,
      "step": 2504,
      "time_per_iteration": 2.5541675090789795
    },
    {
      "auxiliary_loss_clip": 0.01086458,
      "auxiliary_loss_mlp": 0.01005833,
      "balance_loss_clip": 1.02872694,
      "balance_loss_mlp": 1.0046525,
      "epoch": 0.30120844105092287,
      "flos": 71126434938240.0,
      "grad_norm": 0.7594273797325565,
      "language_loss": 0.5185554,
      "learning_rate": 3.276935531557722e-06,
      "loss": 0.5394783,
      "num_input_tokens_seen": 53808725,
      "step": 2505,
      "time_per_iteration": 3.1689682006835938
    },
    {
      "auxiliary_loss_clip": 0.0114669,
      "auxiliary_loss_mlp": 0.01035879,
      "balance_loss_clip": 1.05268741,
      "balance_loss_mlp": 1.02696228,
      "epoch": 0.301328683941562,
      "flos": 20264571302400.0,
      "grad_norm": 2.0878670427861725,
      "language_loss": 0.79518831,
      "learning_rate": 3.2763359003080837e-06,
      "loss": 0.81701398,
      "num_input_tokens_seen": 53825680,
      "step": 2506,
      "time_per_iteration": 2.5947065353393555
    },
    {
      "auxiliary_loss_clip": 0.01077126,
      "auxiliary_loss_mlp": 0.01004033,
      "balance_loss_clip": 1.02478361,
      "balance_loss_mlp": 1.00269222,
      "epoch": 0.30144892683220104,
      "flos": 70648212240000.0,
      "grad_norm": 0.7999185397598263,
      "language_loss": 0.624529,
      "learning_rate": 3.2757360754393047e-06,
      "loss": 0.64534056,
      "num_input_tokens_seen": 53889750,
      "step": 2507,
      "time_per_iteration": 3.1939477920532227
    },
    {
      "auxiliary_loss_clip": 0.01185691,
      "auxiliary_loss_mlp": 0.01028733,
      "balance_loss_clip": 1.05850244,
      "balance_loss_mlp": 1.01961398,
      "epoch": 0.30156916972284015,
      "flos": 22820549241600.0,
      "grad_norm": 2.5404471931931263,
      "language_loss": 0.63833511,
      "learning_rate": 3.2751360570423767e-06,
      "loss": 0.66047931,
      "num_input_tokens_seen": 53908135,
      "step": 2508,
      "time_per_iteration": 2.5242459774017334
    },
    {
      "auxiliary_loss_clip": 0.01173782,
      "auxiliary_loss_mlp": 0.01034597,
      "balance_loss_clip": 1.05881381,
      "balance_loss_mlp": 1.02562666,
      "epoch": 0.3016894126134792,
      "flos": 29899188529920.0,
      "grad_norm": 2.0208810286106322,
      "language_loss": 0.7583909,
      "learning_rate": 3.2745358452083236e-06,
      "loss": 0.78047472,
      "num_input_tokens_seen": 53931035,
      "step": 2509,
      "time_per_iteration": 2.669614315032959
    },
    {
      "auxiliary_loss_clip": 0.01187899,
      "auxiliary_loss_mlp": 0.01028985,
      "balance_loss_clip": 1.06250525,
      "balance_loss_mlp": 1.02098036,
      "epoch": 0.3018096555041183,
      "flos": 21546331200000.0,
      "grad_norm": 1.3699594928564742,
      "language_loss": 0.82284606,
      "learning_rate": 3.2739354400281955e-06,
      "loss": 0.84501493,
      "num_input_tokens_seen": 53952255,
      "step": 2510,
      "time_per_iteration": 2.6003730297088623
    },
    {
      "auxiliary_loss_clip": 0.01065413,
      "auxiliary_loss_mlp": 0.00753369,
      "balance_loss_clip": 1.02279091,
      "balance_loss_mlp": 1.00027013,
      "epoch": 0.3019298983947574,
      "flos": 59136294597120.0,
      "grad_norm": 0.8677804210448475,
      "language_loss": 0.63737381,
      "learning_rate": 3.2733348415930744e-06,
      "loss": 0.65556163,
      "num_input_tokens_seen": 54014125,
      "step": 2511,
      "time_per_iteration": 3.18686842918396
    },
    {
      "auxiliary_loss_clip": 0.01153976,
      "auxiliary_loss_mlp": 0.01029852,
      "balance_loss_clip": 1.05634844,
      "balance_loss_mlp": 1.02103055,
      "epoch": 0.3020501412853965,
      "flos": 34423070941440.0,
      "grad_norm": 1.753753272570682,
      "language_loss": 0.80697787,
      "learning_rate": 3.27273404999407e-06,
      "loss": 0.82881618,
      "num_input_tokens_seen": 54036345,
      "step": 2512,
      "time_per_iteration": 2.7594552040100098
    },
    {
      "auxiliary_loss_clip": 0.01076693,
      "auxiliary_loss_mlp": 0.01001928,
      "balance_loss_clip": 1.02240872,
      "balance_loss_mlp": 1.00051546,
      "epoch": 0.3021703841760356,
      "flos": 71008288128000.0,
      "grad_norm": 0.7945268294047647,
      "language_loss": 0.60523999,
      "learning_rate": 3.272133065322322e-06,
      "loss": 0.62602621,
      "num_input_tokens_seen": 54094615,
      "step": 2513,
      "time_per_iteration": 3.1161978244781494
    },
    {
      "auxiliary_loss_clip": 0.01198852,
      "auxiliary_loss_mlp": 0.01033194,
      "balance_loss_clip": 1.05981231,
      "balance_loss_mlp": 1.02458715,
      "epoch": 0.3022906270666747,
      "flos": 21510528318720.0,
      "grad_norm": 2.294592755166712,
      "language_loss": 0.79291415,
      "learning_rate": 3.271531887669e-06,
      "loss": 0.81523466,
      "num_input_tokens_seen": 54114675,
      "step": 2514,
      "time_per_iteration": 2.560870885848999
    },
    {
      "auxiliary_loss_clip": 0.01146909,
      "auxiliary_loss_mlp": 0.01031409,
      "balance_loss_clip": 1.0526731,
      "balance_loss_mlp": 1.0221231,
      "epoch": 0.30241086995731375,
      "flos": 31132001168640.0,
      "grad_norm": 2.3291015231033088,
      "language_loss": 0.63726056,
      "learning_rate": 3.2709305171253015e-06,
      "loss": 0.65904373,
      "num_input_tokens_seen": 54134795,
      "step": 2515,
      "time_per_iteration": 2.691617250442505
    },
    {
      "auxiliary_loss_clip": 0.0118586,
      "auxiliary_loss_mlp": 0.01029907,
      "balance_loss_clip": 1.06004691,
      "balance_loss_mlp": 1.02132368,
      "epoch": 0.30253111284795287,
      "flos": 23511542152320.0,
      "grad_norm": 1.767714350659697,
      "language_loss": 0.7750448,
      "learning_rate": 3.2703289537824536e-06,
      "loss": 0.79720253,
      "num_input_tokens_seen": 54154595,
      "step": 2516,
      "time_per_iteration": 2.5222325325012207
    },
    {
      "auxiliary_loss_clip": 0.01145503,
      "auxiliary_loss_mlp": 0.01037356,
      "balance_loss_clip": 1.05513978,
      "balance_loss_mlp": 1.0281949,
      "epoch": 0.302651355738592,
      "flos": 18725367651840.0,
      "grad_norm": 2.5750889189759074,
      "language_loss": 0.79021883,
      "learning_rate": 3.269727197731714e-06,
      "loss": 0.81204736,
      "num_input_tokens_seen": 54167360,
      "step": 2517,
      "time_per_iteration": 2.5560317039489746
    },
    {
      "auxiliary_loss_clip": 0.01138605,
      "auxiliary_loss_mlp": 0.01035049,
      "balance_loss_clip": 1.05475283,
      "balance_loss_mlp": 1.02629876,
      "epoch": 0.30277159862923103,
      "flos": 22418888382720.0,
      "grad_norm": 1.6179424218917458,
      "language_loss": 0.78088784,
      "learning_rate": 3.269125249064367e-06,
      "loss": 0.8026244,
      "num_input_tokens_seen": 54187055,
      "step": 2518,
      "time_per_iteration": 2.5996103286743164
    },
    {
      "auxiliary_loss_clip": 0.01201619,
      "auxiliary_loss_mlp": 0.01029556,
      "balance_loss_clip": 1.06070828,
      "balance_loss_mlp": 1.02100289,
      "epoch": 0.30289184151987014,
      "flos": 22273126992000.0,
      "grad_norm": 1.6123060852828135,
      "language_loss": 0.83222878,
      "learning_rate": 3.2685231078717297e-06,
      "loss": 0.85454059,
      "num_input_tokens_seen": 54207245,
      "step": 2519,
      "time_per_iteration": 2.5161924362182617
    },
    {
      "auxiliary_loss_clip": 0.01148396,
      "auxiliary_loss_mlp": 0.00763591,
      "balance_loss_clip": 1.05572081,
      "balance_loss_mlp": 1.00041389,
      "epoch": 0.30301208441050925,
      "flos": 25225594231680.0,
      "grad_norm": 2.634598646533618,
      "language_loss": 0.75251359,
      "learning_rate": 3.267920774245145e-06,
      "loss": 0.77163351,
      "num_input_tokens_seen": 54226650,
      "step": 2520,
      "time_per_iteration": 2.594172239303589
    },
    {
      "auxiliary_loss_clip": 0.01191567,
      "auxiliary_loss_mlp": 0.01038977,
      "balance_loss_clip": 1.06405401,
      "balance_loss_mlp": 1.02931499,
      "epoch": 0.3031323273011483,
      "flos": 23039245198080.0,
      "grad_norm": 2.1235876154280096,
      "language_loss": 0.84863114,
      "learning_rate": 3.2673182482759876e-06,
      "loss": 0.87093663,
      "num_input_tokens_seen": 54245765,
      "step": 2521,
      "time_per_iteration": 2.506431818008423
    },
    {
      "auxiliary_loss_clip": 0.01186747,
      "auxiliary_loss_mlp": 0.01029515,
      "balance_loss_clip": 1.06061029,
      "balance_loss_mlp": 1.02059186,
      "epoch": 0.3032525701917874,
      "flos": 18876695650560.0,
      "grad_norm": 2.65023783129727,
      "language_loss": 0.66362751,
      "learning_rate": 3.266715530055659e-06,
      "loss": 0.68579012,
      "num_input_tokens_seen": 54263915,
      "step": 2522,
      "time_per_iteration": 2.4898760318756104
    },
    {
      "auxiliary_loss_clip": 0.01177743,
      "auxiliary_loss_mlp": 0.01029182,
      "balance_loss_clip": 1.0578686,
      "balance_loss_mlp": 1.02010381,
      "epoch": 0.30337281308242653,
      "flos": 17782641250560.0,
      "grad_norm": 1.5567698222303759,
      "language_loss": 0.80538595,
      "learning_rate": 3.2661126196755927e-06,
      "loss": 0.82745516,
      "num_input_tokens_seen": 54283025,
      "step": 2523,
      "time_per_iteration": 3.3456547260284424
    },
    {
      "auxiliary_loss_clip": 0.01093292,
      "auxiliary_loss_mlp": 0.0100313,
      "balance_loss_clip": 1.02000046,
      "balance_loss_mlp": 1.00193238,
      "epoch": 0.3034930559730656,
      "flos": 57824298426240.0,
      "grad_norm": 0.7842039617537705,
      "language_loss": 0.55972791,
      "learning_rate": 3.265509517227248e-06,
      "loss": 0.58069217,
      "num_input_tokens_seen": 54339840,
      "step": 2524,
      "time_per_iteration": 3.0423965454101562
    },
    {
      "auxiliary_loss_clip": 0.01173376,
      "auxiliary_loss_mlp": 0.01029172,
      "balance_loss_clip": 1.05697024,
      "balance_loss_mlp": 1.02056503,
      "epoch": 0.3036132988637047,
      "flos": 14755587419520.0,
      "grad_norm": 1.6294588722256331,
      "language_loss": 0.81412554,
      "learning_rate": 3.264906222802115e-06,
      "loss": 0.836151,
      "num_input_tokens_seen": 54357690,
      "step": 2525,
      "time_per_iteration": 2.520167589187622
    },
    {
      "auxiliary_loss_clip": 0.01203987,
      "auxiliary_loss_mlp": 0.01032519,
      "balance_loss_clip": 1.06117499,
      "balance_loss_mlp": 1.02296424,
      "epoch": 0.30373354175434375,
      "flos": 21033203460480.0,
      "grad_norm": 1.963109081989034,
      "language_loss": 0.77811176,
      "learning_rate": 3.264302736491715e-06,
      "loss": 0.80047679,
      "num_input_tokens_seen": 54377810,
      "step": 2526,
      "time_per_iteration": 2.5642902851104736
    },
    {
      "auxiliary_loss_clip": 0.0118515,
      "auxiliary_loss_mlp": 0.01028345,
      "balance_loss_clip": 1.06349432,
      "balance_loss_mlp": 1.01973176,
      "epoch": 0.30385378464498286,
      "flos": 21143233797120.0,
      "grad_norm": 1.7335452160130531,
      "language_loss": 0.87648231,
      "learning_rate": 3.263699058387594e-06,
      "loss": 0.89861721,
      "num_input_tokens_seen": 54395245,
      "step": 2527,
      "time_per_iteration": 2.5394296646118164
    },
    {
      "auxiliary_loss_clip": 0.01152684,
      "auxiliary_loss_mlp": 0.01035535,
      "balance_loss_clip": 1.0527072,
      "balance_loss_mlp": 1.02629042,
      "epoch": 0.30397402753562197,
      "flos": 20629244131200.0,
      "grad_norm": 2.1590391549208627,
      "language_loss": 0.90146124,
      "learning_rate": 3.2630951885813315e-06,
      "loss": 0.92334342,
      "num_input_tokens_seen": 54412640,
      "step": 2528,
      "time_per_iteration": 3.4062328338623047
    },
    {
      "auxiliary_loss_clip": 0.01170894,
      "auxiliary_loss_mlp": 0.01034786,
      "balance_loss_clip": 1.05523014,
      "balance_loss_mlp": 1.02592254,
      "epoch": 0.304094270426261,
      "flos": 15085678429440.0,
      "grad_norm": 2.254779655991013,
      "language_loss": 0.78191692,
      "learning_rate": 3.262491127164533e-06,
      "loss": 0.80397367,
      "num_input_tokens_seen": 54431455,
      "step": 2529,
      "time_per_iteration": 3.35711669921875
    },
    {
      "auxiliary_loss_clip": 0.01178079,
      "auxiliary_loss_mlp": 0.00763844,
      "balance_loss_clip": 1.05872631,
      "balance_loss_mlp": 1.00048196,
      "epoch": 0.30421451331690014,
      "flos": 13845216193920.0,
      "grad_norm": 2.3109062809461705,
      "language_loss": 0.80576789,
      "learning_rate": 3.2618868742288337e-06,
      "loss": 0.82518709,
      "num_input_tokens_seen": 54448380,
      "step": 2530,
      "time_per_iteration": 2.5056116580963135
    },
    {
      "auxiliary_loss_clip": 0.01184193,
      "auxiliary_loss_mlp": 0.01034752,
      "balance_loss_clip": 1.05911517,
      "balance_loss_mlp": 1.02624011,
      "epoch": 0.30433475620753925,
      "flos": 17384212615680.0,
      "grad_norm": 1.9498125816392136,
      "language_loss": 0.721264,
      "learning_rate": 3.261282429865899e-06,
      "loss": 0.7434535,
      "num_input_tokens_seen": 54466385,
      "step": 2531,
      "time_per_iteration": 2.5009498596191406
    },
    {
      "auxiliary_loss_clip": 0.0117875,
      "auxiliary_loss_mlp": 0.00762875,
      "balance_loss_clip": 1.06110644,
      "balance_loss_mlp": 1.00043011,
      "epoch": 0.3044549990981783,
      "flos": 18916951818240.0,
      "grad_norm": 1.7131432541660319,
      "language_loss": 0.72472709,
      "learning_rate": 3.2606777941674225e-06,
      "loss": 0.74414337,
      "num_input_tokens_seen": 54485040,
      "step": 2532,
      "time_per_iteration": 2.523322820663452
    },
    {
      "auxiliary_loss_clip": 0.01136456,
      "auxiliary_loss_mlp": 0.01034818,
      "balance_loss_clip": 1.05527604,
      "balance_loss_mlp": 1.02533531,
      "epoch": 0.3045752419888174,
      "flos": 21068431724160.0,
      "grad_norm": 2.048658010233344,
      "language_loss": 0.84644771,
      "learning_rate": 3.2600729672251276e-06,
      "loss": 0.86816043,
      "num_input_tokens_seen": 54502755,
      "step": 2533,
      "time_per_iteration": 2.6052637100219727
    },
    {
      "auxiliary_loss_clip": 0.01201848,
      "auxiliary_loss_mlp": 0.00764031,
      "balance_loss_clip": 1.06364822,
      "balance_loss_mlp": 1.00044715,
      "epoch": 0.3046954848794565,
      "flos": 29096405516160.0,
      "grad_norm": 1.957431760527712,
      "language_loss": 0.65837675,
      "learning_rate": 3.259467949130765e-06,
      "loss": 0.67803556,
      "num_input_tokens_seen": 54524165,
      "step": 2534,
      "time_per_iteration": 2.528587579727173
    },
    {
      "auxiliary_loss_clip": 0.01175538,
      "auxiliary_loss_mlp": 0.01032652,
      "balance_loss_clip": 1.06147075,
      "balance_loss_mlp": 1.0237174,
      "epoch": 0.3048157277700956,
      "flos": 20295346279680.0,
      "grad_norm": 2.717079864066357,
      "language_loss": 0.82668656,
      "learning_rate": 3.2588627399761164e-06,
      "loss": 0.84876847,
      "num_input_tokens_seen": 54540160,
      "step": 2535,
      "time_per_iteration": 2.52421498298645
    },
    {
      "auxiliary_loss_clip": 0.01169873,
      "auxiliary_loss_mlp": 0.01030812,
      "balance_loss_clip": 1.05769825,
      "balance_loss_mlp": 1.02286112,
      "epoch": 0.3049359706607347,
      "flos": 22739929165440.0,
      "grad_norm": 1.8297408170458171,
      "language_loss": 0.70868313,
      "learning_rate": 3.2582573398529903e-06,
      "loss": 0.73068994,
      "num_input_tokens_seen": 54557515,
      "step": 2536,
      "time_per_iteration": 2.525496244430542
    },
    {
      "auxiliary_loss_clip": 0.01157652,
      "auxiliary_loss_mlp": 0.01030913,
      "balance_loss_clip": 1.05547166,
      "balance_loss_mlp": 1.02116156,
      "epoch": 0.3050562135513738,
      "flos": 18434634969600.0,
      "grad_norm": 2.2877642762140815,
      "language_loss": 0.73673189,
      "learning_rate": 3.2576517488532265e-06,
      "loss": 0.75861752,
      "num_input_tokens_seen": 54573865,
      "step": 2537,
      "time_per_iteration": 2.5248055458068848
    },
    {
      "auxiliary_loss_clip": 0.01183149,
      "auxiliary_loss_mlp": 0.01031619,
      "balance_loss_clip": 1.05673671,
      "balance_loss_mlp": 1.02342308,
      "epoch": 0.30517645644201286,
      "flos": 20370327920640.0,
      "grad_norm": 1.996552654536769,
      "language_loss": 0.87468207,
      "learning_rate": 3.257045967068692e-06,
      "loss": 0.89682972,
      "num_input_tokens_seen": 54593120,
      "step": 2538,
      "time_per_iteration": 2.4772703647613525
    },
    {
      "auxiliary_loss_clip": 0.0120401,
      "auxiliary_loss_mlp": 0.01036768,
      "balance_loss_clip": 1.06302202,
      "balance_loss_mlp": 1.02708817,
      "epoch": 0.30529669933265197,
      "flos": 21945118970880.0,
      "grad_norm": 1.5150788562623656,
      "language_loss": 0.82151115,
      "learning_rate": 3.2564399945912848e-06,
      "loss": 0.84391892,
      "num_input_tokens_seen": 54612910,
      "step": 2539,
      "time_per_iteration": 2.465324878692627
    },
    {
      "auxiliary_loss_clip": 0.01147856,
      "auxiliary_loss_mlp": 0.01031328,
      "balance_loss_clip": 1.0562706,
      "balance_loss_mlp": 1.02314401,
      "epoch": 0.305416942223291,
      "flos": 21835411856640.0,
      "grad_norm": 2.112055452973483,
      "language_loss": 0.81893671,
      "learning_rate": 3.2558338315129287e-06,
      "loss": 0.84072852,
      "num_input_tokens_seen": 54631055,
      "step": 2540,
      "time_per_iteration": 2.579378128051758
    },
    {
      "auxiliary_loss_clip": 0.0117988,
      "auxiliary_loss_mlp": 0.01031593,
      "balance_loss_clip": 1.05811334,
      "balance_loss_mlp": 1.02225316,
      "epoch": 0.30553718511393013,
      "flos": 33911810709120.0,
      "grad_norm": 2.3360721068148647,
      "language_loss": 0.75758135,
      "learning_rate": 3.2552274779255785e-06,
      "loss": 0.77969611,
      "num_input_tokens_seen": 54651985,
      "step": 2541,
      "time_per_iteration": 2.568037509918213
    },
    {
      "auxiliary_loss_clip": 0.01188058,
      "auxiliary_loss_mlp": 0.01035742,
      "balance_loss_clip": 1.06177032,
      "balance_loss_mlp": 1.0269084,
      "epoch": 0.30565742800456924,
      "flos": 22268530051200.0,
      "grad_norm": 2.816432528797108,
      "language_loss": 0.7695508,
      "learning_rate": 3.2546209339212184e-06,
      "loss": 0.7917887,
      "num_input_tokens_seen": 54671005,
      "step": 2542,
      "time_per_iteration": 2.503436326980591
    },
    {
      "auxiliary_loss_clip": 0.01174199,
      "auxiliary_loss_mlp": 0.0103293,
      "balance_loss_clip": 1.05772519,
      "balance_loss_mlp": 1.02360773,
      "epoch": 0.3057776708952083,
      "flos": 22565044823040.0,
      "grad_norm": 4.817841391028305,
      "language_loss": 0.77093232,
      "learning_rate": 3.25401419959186e-06,
      "loss": 0.79300356,
      "num_input_tokens_seen": 54691615,
      "step": 2543,
      "time_per_iteration": 2.577075242996216
    },
    {
      "auxiliary_loss_clip": 0.01184035,
      "auxiliary_loss_mlp": 0.010404,
      "balance_loss_clip": 1.06369662,
      "balance_loss_mlp": 1.03090477,
      "epoch": 0.3058979137858474,
      "flos": 21799213925760.0,
      "grad_norm": 1.985366899540621,
      "language_loss": 0.76251411,
      "learning_rate": 3.253407275029545e-06,
      "loss": 0.78475845,
      "num_input_tokens_seen": 54710520,
      "step": 2544,
      "time_per_iteration": 2.552225112915039
    },
    {
      "auxiliary_loss_clip": 0.01164375,
      "auxiliary_loss_mlp": 0.01031448,
      "balance_loss_clip": 1.06153047,
      "balance_loss_mlp": 1.02136254,
      "epoch": 0.3060181566764865,
      "flos": 26979435601920.0,
      "grad_norm": 3.547934076312213,
      "language_loss": 0.79899305,
      "learning_rate": 3.2528001603263425e-06,
      "loss": 0.82095122,
      "num_input_tokens_seen": 54732590,
      "step": 2545,
      "time_per_iteration": 2.676417112350464
    },
    {
      "auxiliary_loss_clip": 0.01189625,
      "auxiliary_loss_mlp": 0.01032753,
      "balance_loss_clip": 1.06468487,
      "balance_loss_mlp": 1.02368116,
      "epoch": 0.3061383995671256,
      "flos": 19865101173120.0,
      "grad_norm": 1.7584696023057218,
      "language_loss": 0.81316751,
      "learning_rate": 3.2521928555743514e-06,
      "loss": 0.83539128,
      "num_input_tokens_seen": 54749935,
      "step": 2546,
      "time_per_iteration": 2.502725124359131
    },
    {
      "auxiliary_loss_clip": 0.01165588,
      "auxiliary_loss_mlp": 0.00764089,
      "balance_loss_clip": 1.05534077,
      "balance_loss_mlp": 1.0003953,
      "epoch": 0.3062586424577647,
      "flos": 22127509255680.0,
      "grad_norm": 1.7343426097250114,
      "language_loss": 0.67941946,
      "learning_rate": 3.2515853608657e-06,
      "loss": 0.69871622,
      "num_input_tokens_seen": 54767935,
      "step": 2547,
      "time_per_iteration": 2.6365807056427
    },
    {
      "auxiliary_loss_clip": 0.01183558,
      "auxiliary_loss_mlp": 0.01034689,
      "balance_loss_clip": 1.06005025,
      "balance_loss_mlp": 1.02557516,
      "epoch": 0.3063788853484038,
      "flos": 20845497962880.0,
      "grad_norm": 1.9789148819918887,
      "language_loss": 0.75077724,
      "learning_rate": 3.250977676292545e-06,
      "loss": 0.77295965,
      "num_input_tokens_seen": 54786175,
      "step": 2548,
      "time_per_iteration": 2.62123966217041
    },
    {
      "auxiliary_loss_clip": 0.01176399,
      "auxiliary_loss_mlp": 0.01028232,
      "balance_loss_clip": 1.05896759,
      "balance_loss_mlp": 1.01921952,
      "epoch": 0.30649912823904285,
      "flos": 16209717707520.0,
      "grad_norm": 2.1408589079634472,
      "language_loss": 0.79484624,
      "learning_rate": 3.2503698019470712e-06,
      "loss": 0.8168925,
      "num_input_tokens_seen": 54801945,
      "step": 2549,
      "time_per_iteration": 3.4255385398864746
    },
    {
      "auxiliary_loss_clip": 0.01185958,
      "auxiliary_loss_mlp": 0.01035496,
      "balance_loss_clip": 1.05819559,
      "balance_loss_mlp": 1.02554846,
      "epoch": 0.30661937112968196,
      "flos": 18617815353600.0,
      "grad_norm": 2.9209145172391393,
      "language_loss": 0.77924049,
      "learning_rate": 3.249761737921492e-06,
      "loss": 0.80145496,
      "num_input_tokens_seen": 54818475,
      "step": 2550,
      "time_per_iteration": 2.5074427127838135
    },
    {
      "auxiliary_loss_clip": 0.01172711,
      "auxiliary_loss_mlp": 0.01039531,
      "balance_loss_clip": 1.06199098,
      "balance_loss_mlp": 1.0304476,
      "epoch": 0.30673961402032107,
      "flos": 31390809638400.0,
      "grad_norm": 1.930094948696054,
      "language_loss": 0.74417871,
      "learning_rate": 3.249153484308051e-06,
      "loss": 0.76630116,
      "num_input_tokens_seen": 54837090,
      "step": 2551,
      "time_per_iteration": 2.6570584774017334
    },
    {
      "auxiliary_loss_clip": 0.0113492,
      "auxiliary_loss_mlp": 0.01031278,
      "balance_loss_clip": 1.05287671,
      "balance_loss_mlp": 1.02201533,
      "epoch": 0.3068598569109601,
      "flos": 20229809915520.0,
      "grad_norm": 1.8974327612896555,
      "language_loss": 0.78112525,
      "learning_rate": 3.2485450411990194e-06,
      "loss": 0.8027873,
      "num_input_tokens_seen": 54856445,
      "step": 2552,
      "time_per_iteration": 2.609426736831665
    },
    {
      "auxiliary_loss_clip": 0.01203668,
      "auxiliary_loss_mlp": 0.01034169,
      "balance_loss_clip": 1.06156111,
      "balance_loss_mlp": 1.02482295,
      "epoch": 0.30698009980159924,
      "flos": 29601991399680.0,
      "grad_norm": 1.67458772363467,
      "language_loss": 0.82118469,
      "learning_rate": 3.2479364086866983e-06,
      "loss": 0.84356302,
      "num_input_tokens_seen": 54876700,
      "step": 2553,
      "time_per_iteration": 2.568513870239258
    },
    {
      "auxiliary_loss_clip": 0.01175701,
      "auxiliary_loss_mlp": 0.00764556,
      "balance_loss_clip": 1.06309676,
      "balance_loss_mlp": 1.00036275,
      "epoch": 0.30710034269223835,
      "flos": 23842423261440.0,
      "grad_norm": 2.0202817862949427,
      "language_loss": 0.81202972,
      "learning_rate": 3.247327586863416e-06,
      "loss": 0.83143228,
      "num_input_tokens_seen": 54897580,
      "step": 2554,
      "time_per_iteration": 2.635573387145996
    },
    {
      "auxiliary_loss_clip": 0.01164287,
      "auxiliary_loss_mlp": 0.01031833,
      "balance_loss_clip": 1.05838156,
      "balance_loss_mlp": 1.02221262,
      "epoch": 0.3072205855828774,
      "flos": 25884986152320.0,
      "grad_norm": 2.5539249194740923,
      "language_loss": 0.76973808,
      "learning_rate": 3.2467185758215304e-06,
      "loss": 0.79169923,
      "num_input_tokens_seen": 54917320,
      "step": 2555,
      "time_per_iteration": 4.326486110687256
    },
    {
      "auxiliary_loss_clip": 0.01164073,
      "auxiliary_loss_mlp": 0.00764421,
      "balance_loss_clip": 1.06065357,
      "balance_loss_mlp": 1.0003407,
      "epoch": 0.3073408284735165,
      "flos": 22236390357120.0,
      "grad_norm": 2.3399816496710737,
      "language_loss": 0.85378468,
      "learning_rate": 3.246109375653428e-06,
      "loss": 0.87306958,
      "num_input_tokens_seen": 54934085,
      "step": 2556,
      "time_per_iteration": 2.566512107849121
    },
    {
      "auxiliary_loss_clip": 0.01202081,
      "auxiliary_loss_mlp": 0.01032906,
      "balance_loss_clip": 1.06287003,
      "balance_loss_mlp": 1.02345324,
      "epoch": 0.30746107136415557,
      "flos": 19500284689920.0,
      "grad_norm": 1.8613564569188183,
      "language_loss": 0.78450274,
      "learning_rate": 3.2454999864515243e-06,
      "loss": 0.80685258,
      "num_input_tokens_seen": 54953460,
      "step": 2557,
      "time_per_iteration": 2.485262632369995
    },
    {
      "auxiliary_loss_clip": 0.01169424,
      "auxiliary_loss_mlp": 0.00764566,
      "balance_loss_clip": 1.05856824,
      "balance_loss_mlp": 1.00033796,
      "epoch": 0.3075813142547947,
      "flos": 21724806902400.0,
      "grad_norm": 1.9734679764863288,
      "language_loss": 0.69481289,
      "learning_rate": 3.244890408308263e-06,
      "loss": 0.71415275,
      "num_input_tokens_seen": 54974165,
      "step": 2558,
      "time_per_iteration": 2.5684025287628174
    },
    {
      "auxiliary_loss_clip": 0.01143103,
      "auxiliary_loss_mlp": 0.01027458,
      "balance_loss_clip": 1.0528121,
      "balance_loss_mlp": 1.01857102,
      "epoch": 0.3077015571454338,
      "flos": 24097963593600.0,
      "grad_norm": 2.0964095580441033,
      "language_loss": 0.61549819,
      "learning_rate": 3.2442806413161165e-06,
      "loss": 0.63720381,
      "num_input_tokens_seen": 54993810,
      "step": 2559,
      "time_per_iteration": 2.6360771656036377
    },
    {
      "auxiliary_loss_clip": 0.01145844,
      "auxiliary_loss_mlp": 0.01036367,
      "balance_loss_clip": 1.05490255,
      "balance_loss_mlp": 1.02663326,
      "epoch": 0.30782180003607285,
      "flos": 18405476104320.0,
      "grad_norm": 2.262498332346023,
      "language_loss": 0.75669622,
      "learning_rate": 3.243670685567586e-06,
      "loss": 0.77851838,
      "num_input_tokens_seen": 55011210,
      "step": 2560,
      "time_per_iteration": 2.574000835418701
    },
    {
      "auxiliary_loss_clip": 0.01167786,
      "auxiliary_loss_mlp": 0.00763195,
      "balance_loss_clip": 1.05660057,
      "balance_loss_mlp": 1.000337,
      "epoch": 0.30794204292671196,
      "flos": 23878549365120.0,
      "grad_norm": 2.2030241756908877,
      "language_loss": 0.80321157,
      "learning_rate": 3.2430605411552012e-06,
      "loss": 0.82252139,
      "num_input_tokens_seen": 55031325,
      "step": 2561,
      "time_per_iteration": 2.574800968170166
    },
    {
      "auxiliary_loss_clip": 0.0106763,
      "auxiliary_loss_mlp": 0.01002978,
      "balance_loss_clip": 1.02392912,
      "balance_loss_mlp": 1.00163066,
      "epoch": 0.30806228581735107,
      "flos": 67927800816000.0,
      "grad_norm": 0.895991085028544,
      "language_loss": 0.70571971,
      "learning_rate": 3.2424502081715205e-06,
      "loss": 0.72642577,
      "num_input_tokens_seen": 55094440,
      "step": 2562,
      "time_per_iteration": 3.1551928520202637
    },
    {
      "auxiliary_loss_clip": 0.01173007,
      "auxiliary_loss_mlp": 0.01035635,
      "balance_loss_clip": 1.05844688,
      "balance_loss_mlp": 1.02624738,
      "epoch": 0.3081825287079901,
      "flos": 23843213360640.0,
      "grad_norm": 1.7024109787593373,
      "language_loss": 0.78258002,
      "learning_rate": 3.241839686709132e-06,
      "loss": 0.80466652,
      "num_input_tokens_seen": 55115375,
      "step": 2563,
      "time_per_iteration": 2.5709052085876465
    },
    {
      "auxiliary_loss_clip": 0.01182179,
      "auxiliary_loss_mlp": 0.01033847,
      "balance_loss_clip": 1.05560935,
      "balance_loss_mlp": 1.02456093,
      "epoch": 0.30830277159862923,
      "flos": 16209969102720.0,
      "grad_norm": 2.2087492961752875,
      "language_loss": 0.81828892,
      "learning_rate": 3.2412289768606495e-06,
      "loss": 0.84044921,
      "num_input_tokens_seen": 55131945,
      "step": 2564,
      "time_per_iteration": 2.4789037704467773
    },
    {
      "auxiliary_loss_clip": 0.01188191,
      "auxiliary_loss_mlp": 0.01036984,
      "balance_loss_clip": 1.06065464,
      "balance_loss_mlp": 1.02835298,
      "epoch": 0.30842301448926834,
      "flos": 29349503723520.0,
      "grad_norm": 1.7893080932667713,
      "language_loss": 0.82809997,
      "learning_rate": 3.240618078718718e-06,
      "loss": 0.85035169,
      "num_input_tokens_seen": 55153405,
      "step": 2565,
      "time_per_iteration": 2.6351876258850098
    },
    {
      "auxiliary_loss_clip": 0.01153988,
      "auxiliary_loss_mlp": 0.01034538,
      "balance_loss_clip": 1.05391157,
      "balance_loss_mlp": 1.02522802,
      "epoch": 0.3085432573799074,
      "flos": 21945190798080.0,
      "grad_norm": 2.7851221701190694,
      "language_loss": 0.74114138,
      "learning_rate": 3.240006992376011e-06,
      "loss": 0.7630266,
      "num_input_tokens_seen": 55173030,
      "step": 2566,
      "time_per_iteration": 2.579993724822998
    },
    {
      "auxiliary_loss_clip": 0.01175019,
      "auxiliary_loss_mlp": 0.0104024,
      "balance_loss_clip": 1.05949759,
      "balance_loss_mlp": 1.03106689,
      "epoch": 0.3086635002705465,
      "flos": 22054718344320.0,
      "grad_norm": 2.198744559276731,
      "language_loss": 0.75980198,
      "learning_rate": 3.2393957179252284e-06,
      "loss": 0.78195465,
      "num_input_tokens_seen": 55189565,
      "step": 2567,
      "time_per_iteration": 2.548823833465576
    },
    {
      "auxiliary_loss_clip": 0.0119968,
      "auxiliary_loss_mlp": 0.01032221,
      "balance_loss_clip": 1.06107259,
      "balance_loss_mlp": 1.02361417,
      "epoch": 0.3087837431611856,
      "flos": 32665925520000.0,
      "grad_norm": 2.21318109491651,
      "language_loss": 0.8092196,
      "learning_rate": 3.2387842554591016e-06,
      "loss": 0.83153862,
      "num_input_tokens_seen": 55210380,
      "step": 2568,
      "time_per_iteration": 2.575373411178589
    },
    {
      "auxiliary_loss_clip": 0.0119888,
      "auxiliary_loss_mlp": 0.01038499,
      "balance_loss_clip": 1.0610764,
      "balance_loss_mlp": 1.02949309,
      "epoch": 0.3089039860518247,
      "flos": 17599245384960.0,
      "grad_norm": 2.0483665100440356,
      "language_loss": 0.87769192,
      "learning_rate": 3.238172605070388e-06,
      "loss": 0.90006578,
      "num_input_tokens_seen": 55225795,
      "step": 2569,
      "time_per_iteration": 2.4413695335388184
    },
    {
      "auxiliary_loss_clip": 0.01182767,
      "auxiliary_loss_mlp": 0.00764516,
      "balance_loss_clip": 1.05796194,
      "balance_loss_mlp": 1.00036359,
      "epoch": 0.3090242289424638,
      "flos": 14383839611520.0,
      "grad_norm": 2.2588303445951907,
      "language_loss": 0.78495175,
      "learning_rate": 3.2375607668518745e-06,
      "loss": 0.80442452,
      "num_input_tokens_seen": 55238830,
      "step": 2570,
      "time_per_iteration": 2.4739160537719727
    },
    {
      "auxiliary_loss_clip": 0.01161531,
      "auxiliary_loss_mlp": 0.01033084,
      "balance_loss_clip": 1.05574048,
      "balance_loss_mlp": 1.02392864,
      "epoch": 0.30914447183310284,
      "flos": 16068625084800.0,
      "grad_norm": 2.4154320674368708,
      "language_loss": 0.90108687,
      "learning_rate": 3.236948740896377e-06,
      "loss": 0.923033,
      "num_input_tokens_seen": 55253630,
      "step": 2571,
      "time_per_iteration": 2.5089807510375977
    },
    {
      "auxiliary_loss_clip": 0.01184009,
      "auxiliary_loss_mlp": 0.01033267,
      "balance_loss_clip": 1.05975842,
      "balance_loss_mlp": 1.02452934,
      "epoch": 0.30926471472374195,
      "flos": 32230221546240.0,
      "grad_norm": 1.3806592437244731,
      "language_loss": 0.84018606,
      "learning_rate": 3.2363365272967384e-06,
      "loss": 0.86235881,
      "num_input_tokens_seen": 55276200,
      "step": 2572,
      "time_per_iteration": 2.575801134109497
    },
    {
      "auxiliary_loss_clip": 0.01182971,
      "auxiliary_loss_mlp": 0.01036158,
      "balance_loss_clip": 1.06215227,
      "balance_loss_mlp": 1.02581048,
      "epoch": 0.30938495761438106,
      "flos": 20370722970240.0,
      "grad_norm": 1.8761373595849862,
      "language_loss": 0.81487721,
      "learning_rate": 3.235724126145832e-06,
      "loss": 0.8370685,
      "num_input_tokens_seen": 55292235,
      "step": 2573,
      "time_per_iteration": 2.4904286861419678
    },
    {
      "auxiliary_loss_clip": 0.01174138,
      "auxiliary_loss_mlp": 0.01036856,
      "balance_loss_clip": 1.05594969,
      "balance_loss_mlp": 1.02745676,
      "epoch": 0.3095052005050201,
      "flos": 24061155131520.0,
      "grad_norm": 1.493099705913763,
      "language_loss": 0.77647698,
      "learning_rate": 3.235111537536558e-06,
      "loss": 0.79858696,
      "num_input_tokens_seen": 55313050,
      "step": 2574,
      "time_per_iteration": 2.524824619293213
    },
    {
      "auxiliary_loss_clip": 0.01186069,
      "auxiliary_loss_mlp": 0.01026742,
      "balance_loss_clip": 1.06078708,
      "balance_loss_mlp": 1.01811147,
      "epoch": 0.30962544339565923,
      "flos": 23401547729280.0,
      "grad_norm": 1.8130970592096554,
      "language_loss": 0.8289758,
      "learning_rate": 3.2344987615618456e-06,
      "loss": 0.8511039,
      "num_input_tokens_seen": 55332885,
      "step": 2575,
      "time_per_iteration": 2.562351942062378
    },
    {
      "auxiliary_loss_clip": 0.01153689,
      "auxiliary_loss_mlp": 0.0103978,
      "balance_loss_clip": 1.05712855,
      "balance_loss_mlp": 1.03079736,
      "epoch": 0.30974568628629834,
      "flos": 33799984692480.0,
      "grad_norm": 1.5430371627268409,
      "language_loss": 0.78230202,
      "learning_rate": 3.2338857983146533e-06,
      "loss": 0.80423671,
      "num_input_tokens_seen": 55354385,
      "step": 2576,
      "time_per_iteration": 2.6436009407043457
    },
    {
      "auxiliary_loss_clip": 0.01159953,
      "auxiliary_loss_mlp": 0.01027954,
      "balance_loss_clip": 1.05686581,
      "balance_loss_mlp": 1.01823258,
      "epoch": 0.3098659291769374,
      "flos": 20229594433920.0,
      "grad_norm": 1.7415915432371871,
      "language_loss": 0.76381147,
      "learning_rate": 3.233272647887966e-06,
      "loss": 0.78569055,
      "num_input_tokens_seen": 55373275,
      "step": 2577,
      "time_per_iteration": 3.376648187637329
    },
    {
      "auxiliary_loss_clip": 0.01201486,
      "auxiliary_loss_mlp": 0.0103839,
      "balance_loss_clip": 1.06256604,
      "balance_loss_mlp": 1.02931213,
      "epoch": 0.3099861720675765,
      "flos": 24748556682240.0,
      "grad_norm": 1.5511575236520265,
      "language_loss": 0.8996799,
      "learning_rate": 3.2326593103747985e-06,
      "loss": 0.92207867,
      "num_input_tokens_seen": 55392290,
      "step": 2578,
      "time_per_iteration": 2.5045881271362305
    },
    {
      "auxiliary_loss_clip": 0.01182392,
      "auxiliary_loss_mlp": 0.01033662,
      "balance_loss_clip": 1.06043911,
      "balance_loss_mlp": 1.02469134,
      "epoch": 0.3101064149582156,
      "flos": 11765485704960.0,
      "grad_norm": 3.4427079003615595,
      "language_loss": 0.85038143,
      "learning_rate": 3.2320457858681936e-06,
      "loss": 0.87254196,
      "num_input_tokens_seen": 55410680,
      "step": 2579,
      "time_per_iteration": 2.4898135662078857
    },
    {
      "auxiliary_loss_clip": 0.01169894,
      "auxiliary_loss_mlp": 0.01027,
      "balance_loss_clip": 1.05594587,
      "balance_loss_mlp": 1.0181253,
      "epoch": 0.31022665784885467,
      "flos": 23033247626880.0,
      "grad_norm": 2.0790316675104306,
      "language_loss": 0.85322726,
      "learning_rate": 3.2314320744612228e-06,
      "loss": 0.87519616,
      "num_input_tokens_seen": 55425980,
      "step": 2580,
      "time_per_iteration": 2.5371475219726562
    },
    {
      "auxiliary_loss_clip": 0.0118271,
      "auxiliary_loss_mlp": 0.01029878,
      "balance_loss_clip": 1.06065238,
      "balance_loss_mlp": 1.02134824,
      "epoch": 0.3103469007394938,
      "flos": 16289188548480.0,
      "grad_norm": 1.6085496797856134,
      "language_loss": 0.76590157,
      "learning_rate": 3.2308181762469854e-06,
      "loss": 0.78802741,
      "num_input_tokens_seen": 55443925,
      "step": 2581,
      "time_per_iteration": 2.503220558166504
    },
    {
      "auxiliary_loss_clip": 0.01200649,
      "auxiliary_loss_mlp": 0.01035645,
      "balance_loss_clip": 1.0603869,
      "balance_loss_mlp": 1.02676368,
      "epoch": 0.3104671436301329,
      "flos": 30515271626880.0,
      "grad_norm": 2.1078312850665246,
      "language_loss": 0.78558457,
      "learning_rate": 3.230204091318609e-06,
      "loss": 0.80794752,
      "num_input_tokens_seen": 55464465,
      "step": 2582,
      "time_per_iteration": 4.275298833847046
    },
    {
      "auxiliary_loss_clip": 0.01196594,
      "auxiliary_loss_mlp": 0.00762884,
      "balance_loss_clip": 1.0596447,
      "balance_loss_mlp": 1.00037193,
      "epoch": 0.31058738652077195,
      "flos": 20047240062720.0,
      "grad_norm": 1.7650193911695984,
      "language_loss": 0.84690374,
      "learning_rate": 3.2295898197692503e-06,
      "loss": 0.86649853,
      "num_input_tokens_seen": 55483425,
      "step": 2583,
      "time_per_iteration": 2.4883065223693848
    },
    {
      "auxiliary_loss_clip": 0.01197684,
      "auxiliary_loss_mlp": 0.0103378,
      "balance_loss_clip": 1.05907989,
      "balance_loss_mlp": 1.02538133,
      "epoch": 0.31070762941141106,
      "flos": 28074639237120.0,
      "grad_norm": 1.6860276225057864,
      "language_loss": 0.78884518,
      "learning_rate": 3.228975361692094e-06,
      "loss": 0.81115985,
      "num_input_tokens_seen": 55504445,
      "step": 2584,
      "time_per_iteration": 2.5420374870300293
    },
    {
      "auxiliary_loss_clip": 0.0119043,
      "auxiliary_loss_mlp": 0.00764102,
      "balance_loss_clip": 1.05986571,
      "balance_loss_mlp": 1.00039804,
      "epoch": 0.31082787230205017,
      "flos": 20521907314560.0,
      "grad_norm": 2.476101687890319,
      "language_loss": 0.80007064,
      "learning_rate": 3.228360717180352e-06,
      "loss": 0.81961596,
      "num_input_tokens_seen": 55521970,
      "step": 2585,
      "time_per_iteration": 2.5032527446746826
    },
    {
      "auxiliary_loss_clip": 0.01096014,
      "auxiliary_loss_mlp": 0.00753815,
      "balance_loss_clip": 1.02352905,
      "balance_loss_mlp": 1.00082803,
      "epoch": 0.3109481151926892,
      "flos": 62445928723200.0,
      "grad_norm": 0.8243884665032329,
      "language_loss": 0.59454364,
      "learning_rate": 3.227745886327266e-06,
      "loss": 0.61304194,
      "num_input_tokens_seen": 55580665,
      "step": 2586,
      "time_per_iteration": 3.005657434463501
    },
    {
      "auxiliary_loss_clip": 0.01095846,
      "auxiliary_loss_mlp": 0.01005528,
      "balance_loss_clip": 1.02343237,
      "balance_loss_mlp": 1.00436008,
      "epoch": 0.31106835808332833,
      "flos": 44746744723200.0,
      "grad_norm": 0.80831643848649,
      "language_loss": 0.55916452,
      "learning_rate": 3.227130869226105e-06,
      "loss": 0.58017826,
      "num_input_tokens_seen": 55637825,
      "step": 2587,
      "time_per_iteration": 3.029630661010742
    },
    {
      "auxiliary_loss_clip": 0.01181659,
      "auxiliary_loss_mlp": 0.01023639,
      "balance_loss_clip": 1.05678034,
      "balance_loss_mlp": 1.01563406,
      "epoch": 0.3111886009739674,
      "flos": 23403056100480.0,
      "grad_norm": 2.407839259251215,
      "language_loss": 0.83063811,
      "learning_rate": 3.226515665970167e-06,
      "loss": 0.85269111,
      "num_input_tokens_seen": 55655365,
      "step": 2588,
      "time_per_iteration": 2.5212247371673584
    },
    {
      "auxiliary_loss_clip": 0.01182968,
      "auxiliary_loss_mlp": 0.01028551,
      "balance_loss_clip": 1.0575577,
      "balance_loss_mlp": 1.0193181,
      "epoch": 0.3113088438646065,
      "flos": 17530728192000.0,
      "grad_norm": 2.3574393718216236,
      "language_loss": 0.85905403,
      "learning_rate": 3.225900276652777e-06,
      "loss": 0.8811692,
      "num_input_tokens_seen": 55672140,
      "step": 2589,
      "time_per_iteration": 2.507843017578125
    },
    {
      "auxiliary_loss_clip": 0.01174194,
      "auxiliary_loss_mlp": 0.01035051,
      "balance_loss_clip": 1.05657244,
      "balance_loss_mlp": 1.02644444,
      "epoch": 0.3114290867552456,
      "flos": 28365802882560.0,
      "grad_norm": 1.711456013896892,
      "language_loss": 0.75411683,
      "learning_rate": 3.2252847013672906e-06,
      "loss": 0.77620929,
      "num_input_tokens_seen": 55694800,
      "step": 2590,
      "time_per_iteration": 2.6080386638641357
    },
    {
      "auxiliary_loss_clip": 0.01145343,
      "auxiliary_loss_mlp": 0.01027895,
      "balance_loss_clip": 1.05118597,
      "balance_loss_mlp": 1.01910901,
      "epoch": 0.31154932964588467,
      "flos": 27379157126400.0,
      "grad_norm": 1.9189915256378762,
      "language_loss": 0.76221371,
      "learning_rate": 3.224668940207089e-06,
      "loss": 0.7839461,
      "num_input_tokens_seen": 55713785,
      "step": 2591,
      "time_per_iteration": 2.7283730506896973
    },
    {
      "auxiliary_loss_clip": 0.01129153,
      "auxiliary_loss_mlp": 0.01038358,
      "balance_loss_clip": 1.04771507,
      "balance_loss_mlp": 1.02913737,
      "epoch": 0.3116695725365238,
      "flos": 26541864120960.0,
      "grad_norm": 1.7434602583579224,
      "language_loss": 0.86475694,
      "learning_rate": 3.2240529932655828e-06,
      "loss": 0.88643205,
      "num_input_tokens_seen": 55733050,
      "step": 2592,
      "time_per_iteration": 2.645447015762329
    },
    {
      "auxiliary_loss_clip": 0.01165464,
      "auxiliary_loss_mlp": 0.01037197,
      "balance_loss_clip": 1.05661762,
      "balance_loss_mlp": 1.02818441,
      "epoch": 0.3117898154271629,
      "flos": 21177600134400.0,
      "grad_norm": 2.420129463294857,
      "language_loss": 0.88537049,
      "learning_rate": 3.223436860636211e-06,
      "loss": 0.90739703,
      "num_input_tokens_seen": 55748685,
      "step": 2593,
      "time_per_iteration": 2.565955400466919
    },
    {
      "auxiliary_loss_clip": 0.01198644,
      "auxiliary_loss_mlp": 0.01037605,
      "balance_loss_clip": 1.06061149,
      "balance_loss_mlp": 1.02911162,
      "epoch": 0.31191005831780194,
      "flos": 27272430840960.0,
      "grad_norm": 1.6335165093259305,
      "language_loss": 0.7407468,
      "learning_rate": 3.2228205424124403e-06,
      "loss": 0.76310927,
      "num_input_tokens_seen": 55771840,
      "step": 2594,
      "time_per_iteration": 2.57686710357666
    },
    {
      "auxiliary_loss_clip": 0.01155782,
      "auxiliary_loss_mlp": 0.01025274,
      "balance_loss_clip": 1.05371058,
      "balance_loss_mlp": 1.01619005,
      "epoch": 0.31203030120844105,
      "flos": 12963501043200.0,
      "grad_norm": 2.4080262192992126,
      "language_loss": 0.74761307,
      "learning_rate": 3.222204038687765e-06,
      "loss": 0.7694236,
      "num_input_tokens_seen": 55784975,
      "step": 2595,
      "time_per_iteration": 2.5144760608673096
    },
    {
      "auxiliary_loss_clip": 0.01178774,
      "auxiliary_loss_mlp": 0.01028422,
      "balance_loss_clip": 1.05625701,
      "balance_loss_mlp": 1.02026248,
      "epoch": 0.31215054409908016,
      "flos": 27562014288000.0,
      "grad_norm": 1.7144754742947774,
      "language_loss": 0.87691414,
      "learning_rate": 3.221587349555709e-06,
      "loss": 0.8989861,
      "num_input_tokens_seen": 55805235,
      "step": 2596,
      "time_per_iteration": 2.611163854598999
    },
    {
      "auxiliary_loss_clip": 0.01173395,
      "auxiliary_loss_mlp": 0.01026693,
      "balance_loss_clip": 1.05747008,
      "balance_loss_mlp": 1.01781225,
      "epoch": 0.3122707869897192,
      "flos": 21506326427520.0,
      "grad_norm": 1.6947963425357557,
      "language_loss": 0.69178277,
      "learning_rate": 3.2209704751098236e-06,
      "loss": 0.71378362,
      "num_input_tokens_seen": 55824265,
      "step": 2597,
      "time_per_iteration": 2.5642731189727783
    },
    {
      "auxiliary_loss_clip": 0.01172083,
      "auxiliary_loss_mlp": 0.01029978,
      "balance_loss_clip": 1.05868542,
      "balance_loss_mlp": 1.02103162,
      "epoch": 0.31239102988035833,
      "flos": 15187017674880.0,
      "grad_norm": 2.015586882940876,
      "language_loss": 0.82808524,
      "learning_rate": 3.2203534154436875e-06,
      "loss": 0.85010588,
      "num_input_tokens_seen": 55838620,
      "step": 2598,
      "time_per_iteration": 2.5268592834472656
    },
    {
      "auxiliary_loss_clip": 0.0112323,
      "auxiliary_loss_mlp": 0.01040877,
      "balance_loss_clip": 1.0518713,
      "balance_loss_mlp": 1.03202569,
      "epoch": 0.31251127277099744,
      "flos": 22053712763520.0,
      "grad_norm": 1.9494711890694392,
      "language_loss": 0.75375074,
      "learning_rate": 3.2197361706509084e-06,
      "loss": 0.77539182,
      "num_input_tokens_seen": 55859375,
      "step": 2599,
      "time_per_iteration": 2.629678964614868
    },
    {
      "auxiliary_loss_clip": 0.01201196,
      "auxiliary_loss_mlp": 0.01033082,
      "balance_loss_clip": 1.05931282,
      "balance_loss_mlp": 1.02324128,
      "epoch": 0.3126315156616365,
      "flos": 15193984913280.0,
      "grad_norm": 2.536284799310931,
      "language_loss": 0.8332063,
      "learning_rate": 3.2191187408251228e-06,
      "loss": 0.8555491,
      "num_input_tokens_seen": 55876535,
      "step": 2600,
      "time_per_iteration": 2.469064950942993
    },
    {
      "auxiliary_loss_clip": 0.01190933,
      "auxiliary_loss_mlp": 0.0103469,
      "balance_loss_clip": 1.05906367,
      "balance_loss_mlp": 1.02470016,
      "epoch": 0.3127517585522756,
      "flos": 18145338831360.0,
      "grad_norm": 2.0636303833563083,
      "language_loss": 0.7878629,
      "learning_rate": 3.218501126059993e-06,
      "loss": 0.81011915,
      "num_input_tokens_seen": 55891930,
      "step": 2601,
      "time_per_iteration": 2.456859588623047
    },
    {
      "auxiliary_loss_clip": 0.01186845,
      "auxiliary_loss_mlp": 0.01028349,
      "balance_loss_clip": 1.05746675,
      "balance_loss_mlp": 1.01906276,
      "epoch": 0.31287200144291466,
      "flos": 21908633731200.0,
      "grad_norm": 1.9752061429597036,
      "language_loss": 0.81559646,
      "learning_rate": 3.2178833264492116e-06,
      "loss": 0.83774841,
      "num_input_tokens_seen": 55910635,
      "step": 2602,
      "time_per_iteration": 2.511857271194458
    },
    {
      "auxiliary_loss_clip": 0.01193463,
      "auxiliary_loss_mlp": 0.01027102,
      "balance_loss_clip": 1.06080794,
      "balance_loss_mlp": 1.01791739,
      "epoch": 0.31299224433355377,
      "flos": 29896997800320.0,
      "grad_norm": 2.0025992454860844,
      "language_loss": 0.76295877,
      "learning_rate": 3.217265342086498e-06,
      "loss": 0.78516448,
      "num_input_tokens_seen": 55931125,
      "step": 2603,
      "time_per_iteration": 3.4304370880126953
    },
    {
      "auxiliary_loss_clip": 0.01161095,
      "auxiliary_loss_mlp": 0.00765278,
      "balance_loss_clip": 1.05863607,
      "balance_loss_mlp": 1.00053036,
      "epoch": 0.3131124872241929,
      "flos": 11655886331520.0,
      "grad_norm": 4.3125303152270975,
      "language_loss": 0.72939312,
      "learning_rate": 3.216647173065599e-06,
      "loss": 0.74865687,
      "num_input_tokens_seen": 55946590,
      "step": 2604,
      "time_per_iteration": 2.5584330558776855
    },
    {
      "auxiliary_loss_clip": 0.01169084,
      "auxiliary_loss_mlp": 0.01034275,
      "balance_loss_clip": 1.05905271,
      "balance_loss_mlp": 1.02479219,
      "epoch": 0.31323273011483194,
      "flos": 49848785470080.0,
      "grad_norm": 1.8110611455828947,
      "language_loss": 0.73628491,
      "learning_rate": 3.216028819480292e-06,
      "loss": 0.75831854,
      "num_input_tokens_seen": 55967930,
      "step": 2605,
      "time_per_iteration": 2.773589849472046
    },
    {
      "auxiliary_loss_clip": 0.01157006,
      "auxiliary_loss_mlp": 0.01033852,
      "balance_loss_clip": 1.0549798,
      "balance_loss_mlp": 1.02457714,
      "epoch": 0.31335297300547105,
      "flos": 22601278667520.0,
      "grad_norm": 3.366422634366582,
      "language_loss": 0.75332034,
      "learning_rate": 3.2154102814243793e-06,
      "loss": 0.77522898,
      "num_input_tokens_seen": 55987070,
      "step": 2606,
      "time_per_iteration": 2.541259765625
    },
    {
      "auxiliary_loss_clip": 0.01160178,
      "auxiliary_loss_mlp": 0.01034541,
      "balance_loss_clip": 1.05788648,
      "balance_loss_mlp": 1.02585053,
      "epoch": 0.31347321589611016,
      "flos": 34710858708480.0,
      "grad_norm": 2.3671053727266798,
      "language_loss": 0.66825628,
      "learning_rate": 3.2147915589916937e-06,
      "loss": 0.69020343,
      "num_input_tokens_seen": 56008630,
      "step": 2607,
      "time_per_iteration": 2.675459861755371
    },
    {
      "auxiliary_loss_clip": 0.0116368,
      "auxiliary_loss_mlp": 0.01031091,
      "balance_loss_clip": 1.05455661,
      "balance_loss_mlp": 1.02198386,
      "epoch": 0.3135934587867492,
      "flos": 19755789108480.0,
      "grad_norm": 2.367441191309818,
      "language_loss": 0.8307637,
      "learning_rate": 3.2141726522760938e-06,
      "loss": 0.85271144,
      "num_input_tokens_seen": 56026690,
      "step": 2608,
      "time_per_iteration": 4.257196664810181
    },
    {
      "auxiliary_loss_clip": 0.01080903,
      "auxiliary_loss_mlp": 0.01003236,
      "balance_loss_clip": 1.02172875,
      "balance_loss_mlp": 1.00213897,
      "epoch": 0.3137137016773883,
      "flos": 65815535583360.0,
      "grad_norm": 0.7093106504270107,
      "language_loss": 0.52659452,
      "learning_rate": 3.213553561371469e-06,
      "loss": 0.54743588,
      "num_input_tokens_seen": 56090425,
      "step": 2609,
      "time_per_iteration": 3.9358878135681152
    },
    {
      "auxiliary_loss_clip": 0.01138251,
      "auxiliary_loss_mlp": 0.01033492,
      "balance_loss_clip": 1.05418134,
      "balance_loss_mlp": 1.02505767,
      "epoch": 0.31383394456802743,
      "flos": 16252739222400.0,
      "grad_norm": 2.1073021052085177,
      "language_loss": 0.95681918,
      "learning_rate": 3.212934286371733e-06,
      "loss": 0.97853661,
      "num_input_tokens_seen": 56107135,
      "step": 2610,
      "time_per_iteration": 2.6059062480926514
    },
    {
      "auxiliary_loss_clip": 0.01187071,
      "auxiliary_loss_mlp": 0.01029251,
      "balance_loss_clip": 1.06295073,
      "balance_loss_mlp": 1.01986957,
      "epoch": 0.3139541874586665,
      "flos": 38795517613440.0,
      "grad_norm": 2.136827029398247,
      "language_loss": 0.830742,
      "learning_rate": 3.2123148273708304e-06,
      "loss": 0.85290515,
      "num_input_tokens_seen": 56127325,
      "step": 2611,
      "time_per_iteration": 2.7035458087921143
    },
    {
      "auxiliary_loss_clip": 0.01197998,
      "auxiliary_loss_mlp": 0.01028919,
      "balance_loss_clip": 1.06103134,
      "balance_loss_mlp": 1.01931071,
      "epoch": 0.3140744303493056,
      "flos": 25046328430080.0,
      "grad_norm": 3.679265695046637,
      "language_loss": 0.7698766,
      "learning_rate": 3.211695184462733e-06,
      "loss": 0.79214579,
      "num_input_tokens_seen": 56148500,
      "step": 2612,
      "time_per_iteration": 2.5048346519470215
    },
    {
      "auxiliary_loss_clip": 0.01063136,
      "auxiliary_loss_mlp": 0.01003309,
      "balance_loss_clip": 1.02280474,
      "balance_loss_mlp": 1.002159,
      "epoch": 0.3141946732399447,
      "flos": 72504254782080.0,
      "grad_norm": 0.8938898730384528,
      "language_loss": 0.60477912,
      "learning_rate": 3.2110753577414383e-06,
      "loss": 0.62544358,
      "num_input_tokens_seen": 56210080,
      "step": 2613,
      "time_per_iteration": 3.12212872505188
    },
    {
      "auxiliary_loss_clip": 0.01171845,
      "auxiliary_loss_mlp": 0.0102732,
      "balance_loss_clip": 1.05562329,
      "balance_loss_mlp": 1.01834309,
      "epoch": 0.31431491613058377,
      "flos": 19239788280960.0,
      "grad_norm": 1.8428816284518195,
      "language_loss": 0.79341722,
      "learning_rate": 3.2104553473009757e-06,
      "loss": 0.81540883,
      "num_input_tokens_seen": 56228200,
      "step": 2614,
      "time_per_iteration": 2.6002743244171143
    },
    {
      "auxiliary_loss_clip": 0.01134423,
      "auxiliary_loss_mlp": 0.01029078,
      "balance_loss_clip": 1.05161834,
      "balance_loss_mlp": 1.02038729,
      "epoch": 0.3144351590212229,
      "flos": 36210596290560.0,
      "grad_norm": 2.0145890970336846,
      "language_loss": 0.68034756,
      "learning_rate": 3.209835153235399e-06,
      "loss": 0.70198256,
      "num_input_tokens_seen": 56249755,
      "step": 2615,
      "time_per_iteration": 2.747421979904175
    },
    {
      "auxiliary_loss_clip": 0.01144817,
      "auxiliary_loss_mlp": 0.01029793,
      "balance_loss_clip": 1.05205631,
      "balance_loss_mlp": 1.02116847,
      "epoch": 0.314555401911862,
      "flos": 18551740285440.0,
      "grad_norm": 1.982516614324758,
      "language_loss": 0.67601323,
      "learning_rate": 3.2092147756387916e-06,
      "loss": 0.69775939,
      "num_input_tokens_seen": 56270080,
      "step": 2616,
      "time_per_iteration": 2.594224691390991
    },
    {
      "auxiliary_loss_clip": 0.0116238,
      "auxiliary_loss_mlp": 0.01031769,
      "balance_loss_clip": 1.05348861,
      "balance_loss_mlp": 1.02202916,
      "epoch": 0.31467564480250104,
      "flos": 16362877299840.0,
      "grad_norm": 2.235411078431745,
      "language_loss": 0.83681297,
      "learning_rate": 3.208594214605264e-06,
      "loss": 0.85875446,
      "num_input_tokens_seen": 56288625,
      "step": 2617,
      "time_per_iteration": 2.5438573360443115
    },
    {
      "auxiliary_loss_clip": 0.01158166,
      "auxiliary_loss_mlp": 0.01031737,
      "balance_loss_clip": 1.05377257,
      "balance_loss_mlp": 1.02332664,
      "epoch": 0.31479588769314015,
      "flos": 21652375127040.0,
      "grad_norm": 1.9088795915386598,
      "language_loss": 0.77119839,
      "learning_rate": 3.2079734702289553e-06,
      "loss": 0.79309744,
      "num_input_tokens_seen": 56307520,
      "step": 2618,
      "time_per_iteration": 2.5629146099090576
    },
    {
      "auxiliary_loss_clip": 0.01080781,
      "auxiliary_loss_mlp": 0.00753882,
      "balance_loss_clip": 1.02235162,
      "balance_loss_mlp": 1.00093412,
      "epoch": 0.3149161305837792,
      "flos": 66051072040320.0,
      "grad_norm": 0.8130474375700619,
      "language_loss": 0.60432166,
      "learning_rate": 3.207352542604031e-06,
      "loss": 0.62266827,
      "num_input_tokens_seen": 56369855,
      "step": 2619,
      "time_per_iteration": 3.16501522064209
    },
    {
      "auxiliary_loss_clip": 0.0114237,
      "auxiliary_loss_mlp": 0.01035521,
      "balance_loss_clip": 1.05244112,
      "balance_loss_mlp": 1.02701545,
      "epoch": 0.3150363734744183,
      "flos": 28987201192320.0,
      "grad_norm": 1.472725751268888,
      "language_loss": 0.78521335,
      "learning_rate": 3.2067314318246864e-06,
      "loss": 0.80699223,
      "num_input_tokens_seen": 56390570,
      "step": 2620,
      "time_per_iteration": 2.6150474548339844
    },
    {
      "auxiliary_loss_clip": 0.0116014,
      "auxiliary_loss_mlp": 0.01027545,
      "balance_loss_clip": 1.05875874,
      "balance_loss_mlp": 1.01849079,
      "epoch": 0.31515661636505743,
      "flos": 27636600879360.0,
      "grad_norm": 1.834349620895664,
      "language_loss": 0.77787185,
      "learning_rate": 3.206110137985143e-06,
      "loss": 0.79974866,
      "num_input_tokens_seen": 56410775,
      "step": 2621,
      "time_per_iteration": 2.601893663406372
    },
    {
      "auxiliary_loss_clip": 0.01142134,
      "auxiliary_loss_mlp": 0.01031852,
      "balance_loss_clip": 1.05157673,
      "balance_loss_mlp": 1.02236319,
      "epoch": 0.3152768592556965,
      "flos": 24605632465920.0,
      "grad_norm": 2.046596875939888,
      "language_loss": 0.92146122,
      "learning_rate": 3.2054886611796505e-06,
      "loss": 0.94320112,
      "num_input_tokens_seen": 56429770,
      "step": 2622,
      "time_per_iteration": 2.5968804359436035
    },
    {
      "auxiliary_loss_clip": 0.01092255,
      "auxiliary_loss_mlp": 0.01005424,
      "balance_loss_clip": 1.02100754,
      "balance_loss_mlp": 1.00423741,
      "epoch": 0.3153971021463356,
      "flos": 68476908026880.0,
      "grad_norm": 0.9058345926050926,
      "language_loss": 0.63517946,
      "learning_rate": 3.204867001502487e-06,
      "loss": 0.65615624,
      "num_input_tokens_seen": 56488425,
      "step": 2623,
      "time_per_iteration": 3.0292139053344727
    },
    {
      "auxiliary_loss_clip": 0.01200761,
      "auxiliary_loss_mlp": 0.01037772,
      "balance_loss_clip": 1.06225538,
      "balance_loss_mlp": 1.02772212,
      "epoch": 0.3155173450369747,
      "flos": 25593714766080.0,
      "grad_norm": 2.7139432294518127,
      "language_loss": 0.80709815,
      "learning_rate": 3.2042451590479567e-06,
      "loss": 0.82948345,
      "num_input_tokens_seen": 56508940,
      "step": 2624,
      "time_per_iteration": 2.5205681324005127
    },
    {
      "auxiliary_loss_clip": 0.01195533,
      "auxiliary_loss_mlp": 0.01030614,
      "balance_loss_clip": 1.06097662,
      "balance_loss_mlp": 1.02177453,
      "epoch": 0.31563758792761376,
      "flos": 24309333175680.0,
      "grad_norm": 1.6041019879332563,
      "language_loss": 0.86744505,
      "learning_rate": 3.203623133910394e-06,
      "loss": 0.88970655,
      "num_input_tokens_seen": 56527245,
      "step": 2625,
      "time_per_iteration": 2.5068979263305664
    },
    {
      "auxiliary_loss_clip": 0.0112798,
      "auxiliary_loss_mlp": 0.01031815,
      "balance_loss_clip": 1.05176115,
      "balance_loss_mlp": 1.02295136,
      "epoch": 0.31575783081825287,
      "flos": 31903865550720.0,
      "grad_norm": 2.8678377148333003,
      "language_loss": 0.77635121,
      "learning_rate": 3.203000926184158e-06,
      "loss": 0.79794919,
      "num_input_tokens_seen": 56546170,
      "step": 2626,
      "time_per_iteration": 2.7037241458892822
    },
    {
      "auxiliary_loss_clip": 0.01196746,
      "auxiliary_loss_mlp": 0.0103294,
      "balance_loss_clip": 1.06001151,
      "balance_loss_mlp": 1.02461886,
      "epoch": 0.315878073708892,
      "flos": 30810960385920.0,
      "grad_norm": 1.60115179440242,
      "language_loss": 0.77441078,
      "learning_rate": 3.202378535963639e-06,
      "loss": 0.79670763,
      "num_input_tokens_seen": 56567085,
      "step": 2627,
      "time_per_iteration": 2.5422019958496094
    },
    {
      "auxiliary_loss_clip": 0.01160373,
      "auxiliary_loss_mlp": 0.00764797,
      "balance_loss_clip": 1.0549165,
      "balance_loss_mlp": 1.00059295,
      "epoch": 0.31599831659953104,
      "flos": 22200264253440.0,
      "grad_norm": 1.7362295986004062,
      "language_loss": 0.83955228,
      "learning_rate": 3.2017559633432516e-06,
      "loss": 0.85880399,
      "num_input_tokens_seen": 56586715,
      "step": 2628,
      "time_per_iteration": 2.552711009979248
    },
    {
      "auxiliary_loss_clip": 0.01175928,
      "auxiliary_loss_mlp": 0.01035718,
      "balance_loss_clip": 1.05587053,
      "balance_loss_mlp": 1.02692628,
      "epoch": 0.31611855949017015,
      "flos": 25593463370880.0,
      "grad_norm": 3.1274857547054555,
      "language_loss": 0.66390634,
      "learning_rate": 3.2011332084174398e-06,
      "loss": 0.68602282,
      "num_input_tokens_seen": 56607585,
      "step": 2629,
      "time_per_iteration": 2.5938720703125
    },
    {
      "auxiliary_loss_clip": 0.0118026,
      "auxiliary_loss_mlp": 0.01030081,
      "balance_loss_clip": 1.05879664,
      "balance_loss_mlp": 1.02122974,
      "epoch": 0.31623880238080926,
      "flos": 20594087694720.0,
      "grad_norm": 1.806039296557657,
      "language_loss": 0.89102179,
      "learning_rate": 3.2005102712806756e-06,
      "loss": 0.91312522,
      "num_input_tokens_seen": 56626415,
      "step": 2630,
      "time_per_iteration": 3.357517719268799
    },
    {
      "auxiliary_loss_clip": 0.01185915,
      "auxiliary_loss_mlp": 0.01033641,
      "balance_loss_clip": 1.05891585,
      "balance_loss_mlp": 1.02447951,
      "epoch": 0.3163590452714483,
      "flos": 12784917600000.0,
      "grad_norm": 2.2428645288132847,
      "language_loss": 0.72755176,
      "learning_rate": 3.1998871520274575e-06,
      "loss": 0.74974728,
      "num_input_tokens_seen": 56641750,
      "step": 2631,
      "time_per_iteration": 2.4795680046081543
    },
    {
      "auxiliary_loss_clip": 0.01169185,
      "auxiliary_loss_mlp": 0.01035875,
      "balance_loss_clip": 1.05409968,
      "balance_loss_mlp": 1.02701151,
      "epoch": 0.3164792881620874,
      "flos": 23041292273280.0,
      "grad_norm": 1.9105310617123425,
      "language_loss": 0.84661406,
      "learning_rate": 3.199263850752312e-06,
      "loss": 0.86866462,
      "num_input_tokens_seen": 56662585,
      "step": 2632,
      "time_per_iteration": 2.5387017726898193
    },
    {
      "auxiliary_loss_clip": 0.01183183,
      "auxiliary_loss_mlp": 0.01035867,
      "balance_loss_clip": 1.05679798,
      "balance_loss_mlp": 1.02616882,
      "epoch": 0.31659953105272653,
      "flos": 18296271780480.0,
      "grad_norm": 1.9980979232432317,
      "language_loss": 0.85386515,
      "learning_rate": 3.198640367549795e-06,
      "loss": 0.8760556,
      "num_input_tokens_seen": 56681480,
      "step": 2633,
      "time_per_iteration": 2.5256948471069336
    },
    {
      "auxiliary_loss_clip": 0.01182296,
      "auxiliary_loss_mlp": 0.00763639,
      "balance_loss_clip": 1.05698788,
      "balance_loss_mlp": 1.00052941,
      "epoch": 0.3167197739433656,
      "flos": 25703421880320.0,
      "grad_norm": 1.6505522849425667,
      "language_loss": 0.85933113,
      "learning_rate": 3.198016702514487e-06,
      "loss": 0.8787905,
      "num_input_tokens_seen": 56701760,
      "step": 2634,
      "time_per_iteration": 2.5501585006713867
    },
    {
      "auxiliary_loss_clip": 0.01195496,
      "auxiliary_loss_mlp": 0.01030855,
      "balance_loss_clip": 1.05887139,
      "balance_loss_mlp": 1.02205682,
      "epoch": 0.3168400168340047,
      "flos": 23546016230400.0,
      "grad_norm": 1.583378822339557,
      "language_loss": 0.84266776,
      "learning_rate": 3.1973928557409972e-06,
      "loss": 0.86493123,
      "num_input_tokens_seen": 56719800,
      "step": 2635,
      "time_per_iteration": 5.027114629745483
    },
    {
      "auxiliary_loss_clip": 0.01193628,
      "auxiliary_loss_mlp": 0.01031587,
      "balance_loss_clip": 1.05843687,
      "balance_loss_mlp": 1.02272344,
      "epoch": 0.31696025972464376,
      "flos": 28366449327360.0,
      "grad_norm": 1.9198474815738502,
      "language_loss": 0.71574003,
      "learning_rate": 3.1967688273239636e-06,
      "loss": 0.73799217,
      "num_input_tokens_seen": 56739605,
      "step": 2636,
      "time_per_iteration": 2.531128406524658
    },
    {
      "auxiliary_loss_clip": 0.01154394,
      "auxiliary_loss_mlp": 0.01032035,
      "balance_loss_clip": 1.05565,
      "balance_loss_mlp": 1.0230881,
      "epoch": 0.31708050261528287,
      "flos": 16399111144320.0,
      "grad_norm": 1.6785021220979,
      "language_loss": 0.82015347,
      "learning_rate": 3.1961446173580503e-06,
      "loss": 0.84201783,
      "num_input_tokens_seen": 56756545,
      "step": 2637,
      "time_per_iteration": 2.5340161323547363
    },
    {
      "auxiliary_loss_clip": 0.01165216,
      "auxiliary_loss_mlp": 0.01032469,
      "balance_loss_clip": 1.05676579,
      "balance_loss_mlp": 1.02384412,
      "epoch": 0.317200745505922,
      "flos": 26212347728640.0,
      "grad_norm": 1.648827139568316,
      "language_loss": 0.77255428,
      "learning_rate": 3.1955202259379502e-06,
      "loss": 0.79453117,
      "num_input_tokens_seen": 56778275,
      "step": 2638,
      "time_per_iteration": 2.572671890258789
    },
    {
      "auxiliary_loss_clip": 0.01179761,
      "auxiliary_loss_mlp": 0.01033718,
      "balance_loss_clip": 1.05564928,
      "balance_loss_mlp": 1.02509892,
      "epoch": 0.31732098839656103,
      "flos": 31350876693120.0,
      "grad_norm": 2.1162610733862754,
      "language_loss": 0.82747221,
      "learning_rate": 3.194895653158381e-06,
      "loss": 0.84960705,
      "num_input_tokens_seen": 56797215,
      "step": 2639,
      "time_per_iteration": 2.5851449966430664
    },
    {
      "auxiliary_loss_clip": 0.01089319,
      "auxiliary_loss_mlp": 0.01001988,
      "balance_loss_clip": 1.01750064,
      "balance_loss_mlp": 1.00092149,
      "epoch": 0.31744123128720014,
      "flos": 58989024835200.0,
      "grad_norm": 0.9047474169079357,
      "language_loss": 0.55607307,
      "learning_rate": 3.194270899114093e-06,
      "loss": 0.57698613,
      "num_input_tokens_seen": 56863010,
      "step": 2640,
      "time_per_iteration": 3.142221689224243
    },
    {
      "auxiliary_loss_clip": 0.01187806,
      "auxiliary_loss_mlp": 0.01034459,
      "balance_loss_clip": 1.05908442,
      "balance_loss_mlp": 1.02434397,
      "epoch": 0.31756147417783925,
      "flos": 17417573372160.0,
      "grad_norm": 2.0362146685051843,
      "language_loss": 0.81996644,
      "learning_rate": 3.193645963899858e-06,
      "loss": 0.84218907,
      "num_input_tokens_seen": 56880625,
      "step": 2641,
      "time_per_iteration": 2.5452685356140137
    },
    {
      "auxiliary_loss_clip": 0.01161364,
      "auxiliary_loss_mlp": 0.01027953,
      "balance_loss_clip": 1.05526114,
      "balance_loss_mlp": 1.01926208,
      "epoch": 0.3176817170684783,
      "flos": 25481673267840.0,
      "grad_norm": 1.7722952033375337,
      "language_loss": 0.84076703,
      "learning_rate": 3.193020847610479e-06,
      "loss": 0.86266017,
      "num_input_tokens_seen": 56900945,
      "step": 2642,
      "time_per_iteration": 2.5563437938690186
    },
    {
      "auxiliary_loss_clip": 0.01163302,
      "auxiliary_loss_mlp": 0.01031854,
      "balance_loss_clip": 1.05836082,
      "balance_loss_mlp": 1.02279413,
      "epoch": 0.3178019599591174,
      "flos": 24972603765120.0,
      "grad_norm": 2.0340945530910575,
      "language_loss": 0.71406412,
      "learning_rate": 3.192395550340787e-06,
      "loss": 0.73601568,
      "num_input_tokens_seen": 56918895,
      "step": 2643,
      "time_per_iteration": 2.5983755588531494
    },
    {
      "auxiliary_loss_clip": 0.01182187,
      "auxiliary_loss_mlp": 0.01029862,
      "balance_loss_clip": 1.05884242,
      "balance_loss_mlp": 1.02141619,
      "epoch": 0.31792220284975653,
      "flos": 12422220019200.0,
      "grad_norm": 1.852321342700682,
      "language_loss": 0.76504725,
      "learning_rate": 3.191770072185638e-06,
      "loss": 0.78716767,
      "num_input_tokens_seen": 56935890,
      "step": 2644,
      "time_per_iteration": 2.5009708404541016
    },
    {
      "auxiliary_loss_clip": 0.01180588,
      "auxiliary_loss_mlp": 0.0103713,
      "balance_loss_clip": 1.05742192,
      "balance_loss_mlp": 1.0278852,
      "epoch": 0.3180424457403956,
      "flos": 15485759089920.0,
      "grad_norm": 2.395451684155793,
      "language_loss": 0.72973293,
      "learning_rate": 3.191144413239916e-06,
      "loss": 0.75191009,
      "num_input_tokens_seen": 56952460,
      "step": 2645,
      "time_per_iteration": 2.4313387870788574
    },
    {
      "auxiliary_loss_clip": 0.01169339,
      "auxiliary_loss_mlp": 0.01036704,
      "balance_loss_clip": 1.05692148,
      "balance_loss_mlp": 1.02751279,
      "epoch": 0.3181626886310347,
      "flos": 26174964648960.0,
      "grad_norm": 1.983108536040906,
      "language_loss": 0.88373291,
      "learning_rate": 3.190518573598534e-06,
      "loss": 0.90579337,
      "num_input_tokens_seen": 56969065,
      "step": 2646,
      "time_per_iteration": 2.5359935760498047
    },
    {
      "auxiliary_loss_clip": 0.01159165,
      "auxiliary_loss_mlp": 0.01037176,
      "balance_loss_clip": 1.05368567,
      "balance_loss_mlp": 1.02776408,
      "epoch": 0.3182829315216738,
      "flos": 25483109811840.0,
      "grad_norm": 2.122215686681643,
      "language_loss": 0.77337086,
      "learning_rate": 3.1898925533564308e-06,
      "loss": 0.79533428,
      "num_input_tokens_seen": 56990535,
      "step": 2647,
      "time_per_iteration": 2.6039280891418457
    },
    {
      "auxiliary_loss_clip": 0.01142102,
      "auxiliary_loss_mlp": 0.01034459,
      "balance_loss_clip": 1.05286932,
      "balance_loss_mlp": 1.02556622,
      "epoch": 0.31840317441231286,
      "flos": 18113701927680.0,
      "grad_norm": 2.342769357604192,
      "language_loss": 0.64212525,
      "learning_rate": 3.1892663526085733e-06,
      "loss": 0.66389084,
      "num_input_tokens_seen": 57008910,
      "step": 2648,
      "time_per_iteration": 2.537564516067505
    },
    {
      "auxiliary_loss_clip": 0.01088266,
      "auxiliary_loss_mlp": 0.01004107,
      "balance_loss_clip": 1.01677966,
      "balance_loss_mlp": 1.00302792,
      "epoch": 0.31852341730295197,
      "flos": 64741948957440.0,
      "grad_norm": 0.7484988669303206,
      "language_loss": 0.56965578,
      "learning_rate": 3.188639971449956e-06,
      "loss": 0.59057951,
      "num_input_tokens_seen": 57074960,
      "step": 2649,
      "time_per_iteration": 3.0121805667877197
    },
    {
      "auxiliary_loss_clip": 0.01197483,
      "auxiliary_loss_mlp": 0.01028465,
      "balance_loss_clip": 1.05930972,
      "balance_loss_mlp": 1.01973331,
      "epoch": 0.318643660193591,
      "flos": 20668135582080.0,
      "grad_norm": 2.331412653515872,
      "language_loss": 0.72185898,
      "learning_rate": 3.1880134099756e-06,
      "loss": 0.74411851,
      "num_input_tokens_seen": 57094595,
      "step": 2650,
      "time_per_iteration": 2.5160577297210693
    },
    {
      "auxiliary_loss_clip": 0.01177795,
      "auxiliary_loss_mlp": 0.0102325,
      "balance_loss_clip": 1.05430102,
      "balance_loss_mlp": 1.01451826,
      "epoch": 0.31876390308423014,
      "flos": 26943345411840.0,
      "grad_norm": 1.9460425718120204,
      "language_loss": 0.69406295,
      "learning_rate": 3.1873866682805535e-06,
      "loss": 0.71607339,
      "num_input_tokens_seen": 57115290,
      "step": 2651,
      "time_per_iteration": 2.5709779262542725
    },
    {
      "auxiliary_loss_clip": 0.01171562,
      "auxiliary_loss_mlp": 0.01033712,
      "balance_loss_clip": 1.05688798,
      "balance_loss_mlp": 1.02476549,
      "epoch": 0.31888414597486925,
      "flos": 18041916597120.0,
      "grad_norm": 1.938450377907265,
      "language_loss": 0.88781404,
      "learning_rate": 3.186759746459894e-06,
      "loss": 0.90986675,
      "num_input_tokens_seen": 57134400,
      "step": 2652,
      "time_per_iteration": 2.4988715648651123
    },
    {
      "auxiliary_loss_clip": 0.01167836,
      "auxiliary_loss_mlp": 0.01028148,
      "balance_loss_clip": 1.05722857,
      "balance_loss_mlp": 1.01909363,
      "epoch": 0.3190043888655083,
      "flos": 25149319701120.0,
      "grad_norm": 1.7924446837430996,
      "language_loss": 0.79769063,
      "learning_rate": 3.1861326446087246e-06,
      "loss": 0.81965047,
      "num_input_tokens_seen": 57153140,
      "step": 2653,
      "time_per_iteration": 2.556647777557373
    },
    {
      "auxiliary_loss_clip": 0.01183329,
      "auxiliary_loss_mlp": 0.01030442,
      "balance_loss_clip": 1.05700707,
      "balance_loss_mlp": 1.02064347,
      "epoch": 0.3191246317561474,
      "flos": 22053892331520.0,
      "grad_norm": 2.52207124254031,
      "language_loss": 0.71587068,
      "learning_rate": 3.1855053628221763e-06,
      "loss": 0.73800838,
      "num_input_tokens_seen": 57172395,
      "step": 2654,
      "time_per_iteration": 2.5029003620147705
    },
    {
      "auxiliary_loss_clip": 0.01143666,
      "auxiliary_loss_mlp": 0.01030823,
      "balance_loss_clip": 1.04982662,
      "balance_loss_mlp": 1.0214119,
      "epoch": 0.3192448746467865,
      "flos": 14901815687040.0,
      "grad_norm": 2.867701481127341,
      "language_loss": 0.8983897,
      "learning_rate": 3.184877901195407e-06,
      "loss": 0.92013454,
      "num_input_tokens_seen": 57189090,
      "step": 2655,
      "time_per_iteration": 2.5815958976745605
    },
    {
      "auxiliary_loss_clip": 0.01074607,
      "auxiliary_loss_mlp": 0.01010327,
      "balance_loss_clip": 1.0312084,
      "balance_loss_mlp": 1.00793076,
      "epoch": 0.3193651175374256,
      "flos": 67234832657280.0,
      "grad_norm": 0.8000441545952708,
      "language_loss": 0.62858403,
      "learning_rate": 3.184250259823602e-06,
      "loss": 0.64943337,
      "num_input_tokens_seen": 57251620,
      "step": 2656,
      "time_per_iteration": 3.179706573486328
    },
    {
      "auxiliary_loss_clip": 0.01151327,
      "auxiliary_loss_mlp": 0.01032392,
      "balance_loss_clip": 1.05284452,
      "balance_loss_mlp": 1.02278924,
      "epoch": 0.3194853604280647,
      "flos": 12233077977600.0,
      "grad_norm": 2.10505701184891,
      "language_loss": 0.81961387,
      "learning_rate": 3.183622438801974e-06,
      "loss": 0.84145105,
      "num_input_tokens_seen": 57266910,
      "step": 2657,
      "time_per_iteration": 3.364516496658325
    },
    {
      "auxiliary_loss_clip": 0.01198138,
      "auxiliary_loss_mlp": 0.01037331,
      "balance_loss_clip": 1.06061089,
      "balance_loss_mlp": 1.02889752,
      "epoch": 0.3196056033187038,
      "flos": 14939917038720.0,
      "grad_norm": 2.0034519755168665,
      "language_loss": 0.75626612,
      "learning_rate": 3.1829944382257637e-06,
      "loss": 0.77862084,
      "num_input_tokens_seen": 57285040,
      "step": 2658,
      "time_per_iteration": 2.4232571125030518
    },
    {
      "auxiliary_loss_clip": 0.01178779,
      "auxiliary_loss_mlp": 0.01029475,
      "balance_loss_clip": 1.05815458,
      "balance_loss_mlp": 1.02107072,
      "epoch": 0.31972584620934286,
      "flos": 23768878164480.0,
      "grad_norm": 2.5705549797136262,
      "language_loss": 0.81624818,
      "learning_rate": 3.1823662581902373e-06,
      "loss": 0.83833075,
      "num_input_tokens_seen": 57302725,
      "step": 2659,
      "time_per_iteration": 2.488865852355957
    },
    {
      "auxiliary_loss_clip": 0.0113726,
      "auxiliary_loss_mlp": 0.01028363,
      "balance_loss_clip": 1.04705095,
      "balance_loss_mlp": 1.01924896,
      "epoch": 0.31984608909998197,
      "flos": 21251540280960.0,
      "grad_norm": 2.2266351166514236,
      "language_loss": 0.74425817,
      "learning_rate": 3.1817378987906896e-06,
      "loss": 0.76591432,
      "num_input_tokens_seen": 57322230,
      "step": 2660,
      "time_per_iteration": 2.53401255607605
    },
    {
      "auxiliary_loss_clip": 0.0112966,
      "auxiliary_loss_mlp": 0.01037263,
      "balance_loss_clip": 1.05116057,
      "balance_loss_mlp": 1.02820945,
      "epoch": 0.3199663319906211,
      "flos": 18296235866880.0,
      "grad_norm": 1.987371293722694,
      "language_loss": 0.79672003,
      "learning_rate": 3.181109360122442e-06,
      "loss": 0.8183893,
      "num_input_tokens_seen": 57339820,
      "step": 2661,
      "time_per_iteration": 3.4290072917938232
    },
    {
      "auxiliary_loss_clip": 0.01146868,
      "auxiliary_loss_mlp": 0.01029228,
      "balance_loss_clip": 1.05140936,
      "balance_loss_mlp": 1.02025127,
      "epoch": 0.32008657488126013,
      "flos": 18733627779840.0,
      "grad_norm": 2.425880746601867,
      "language_loss": 0.78360391,
      "learning_rate": 3.1804806422808445e-06,
      "loss": 0.80536491,
      "num_input_tokens_seen": 57356955,
      "step": 2662,
      "time_per_iteration": 4.198965787887573
    },
    {
      "auxiliary_loss_clip": 0.01156041,
      "auxiliary_loss_mlp": 0.01036098,
      "balance_loss_clip": 1.05287266,
      "balance_loss_mlp": 1.02707434,
      "epoch": 0.32020681777189924,
      "flos": 20595344670720.0,
      "grad_norm": 1.7860179000476801,
      "language_loss": 0.72918212,
      "learning_rate": 3.1798517453612714e-06,
      "loss": 0.75110352,
      "num_input_tokens_seen": 57376760,
      "step": 2663,
      "time_per_iteration": 2.5562779903411865
    },
    {
      "auxiliary_loss_clip": 0.01179521,
      "auxiliary_loss_mlp": 0.01034504,
      "balance_loss_clip": 1.06063056,
      "balance_loss_mlp": 1.02580774,
      "epoch": 0.32032706066253835,
      "flos": 35261692750080.0,
      "grad_norm": 1.773240893572529,
      "language_loss": 0.74971962,
      "learning_rate": 3.1792226694591265e-06,
      "loss": 0.77185988,
      "num_input_tokens_seen": 57398145,
      "step": 2664,
      "time_per_iteration": 2.658367156982422
    },
    {
      "auxiliary_loss_clip": 0.0114839,
      "auxiliary_loss_mlp": 0.01028116,
      "balance_loss_clip": 1.0551343,
      "balance_loss_mlp": 1.01968193,
      "epoch": 0.3204473035531774,
      "flos": 15304230731520.0,
      "grad_norm": 2.03338043129447,
      "language_loss": 0.80644727,
      "learning_rate": 3.178593414669841e-06,
      "loss": 0.82821238,
      "num_input_tokens_seen": 57416730,
      "step": 2665,
      "time_per_iteration": 2.598355293273926
    },
    {
      "auxiliary_loss_clip": 0.01183026,
      "auxiliary_loss_mlp": 0.01026295,
      "balance_loss_clip": 1.05799913,
      "balance_loss_mlp": 1.01747918,
      "epoch": 0.3205675464438165,
      "flos": 24462564595200.0,
      "grad_norm": 2.8005865739918296,
      "language_loss": 0.7041229,
      "learning_rate": 3.1779639810888707e-06,
      "loss": 0.72621614,
      "num_input_tokens_seen": 57436325,
      "step": 2666,
      "time_per_iteration": 2.521804094314575
    },
    {
      "auxiliary_loss_clip": 0.01179358,
      "auxiliary_loss_mlp": 0.01031687,
      "balance_loss_clip": 1.05889463,
      "balance_loss_mlp": 1.02268636,
      "epoch": 0.3206877893344556,
      "flos": 22456235548800.0,
      "grad_norm": 1.786391339346807,
      "language_loss": 0.75545996,
      "learning_rate": 3.1773343688117013e-06,
      "loss": 0.77757043,
      "num_input_tokens_seen": 57457235,
      "step": 2667,
      "time_per_iteration": 2.5075602531433105
    },
    {
      "auxiliary_loss_clip": 0.01169584,
      "auxiliary_loss_mlp": 0.00763307,
      "balance_loss_clip": 1.05522203,
      "balance_loss_mlp": 1.00067472,
      "epoch": 0.3208080322250947,
      "flos": 20412236113920.0,
      "grad_norm": 4.225036790900256,
      "language_loss": 0.84068,
      "learning_rate": 3.1767045779338445e-06,
      "loss": 0.8600089,
      "num_input_tokens_seen": 57474895,
      "step": 2668,
      "time_per_iteration": 2.5326688289642334
    },
    {
      "auxiliary_loss_clip": 0.01176118,
      "auxiliary_loss_mlp": 0.01026263,
      "balance_loss_clip": 1.05298924,
      "balance_loss_mlp": 1.01857424,
      "epoch": 0.3209282751157338,
      "flos": 21762118154880.0,
      "grad_norm": 1.941818888435655,
      "language_loss": 0.91274035,
      "learning_rate": 3.176074608550839e-06,
      "loss": 0.93476427,
      "num_input_tokens_seen": 57490715,
      "step": 2669,
      "time_per_iteration": 2.5570433139801025
    },
    {
      "auxiliary_loss_clip": 0.01123878,
      "auxiliary_loss_mlp": 0.01034385,
      "balance_loss_clip": 1.05132246,
      "balance_loss_mlp": 1.02551031,
      "epoch": 0.32104851800637285,
      "flos": 22055041566720.0,
      "grad_norm": 2.381875996007557,
      "language_loss": 0.82460862,
      "learning_rate": 3.17544446075825e-06,
      "loss": 0.84619123,
      "num_input_tokens_seen": 57509880,
      "step": 2670,
      "time_per_iteration": 2.651054859161377
    },
    {
      "auxiliary_loss_clip": 0.01168614,
      "auxiliary_loss_mlp": 0.0103203,
      "balance_loss_clip": 1.05406141,
      "balance_loss_mlp": 1.02391744,
      "epoch": 0.32116876089701196,
      "flos": 37012301896320.0,
      "grad_norm": 1.597889973427214,
      "language_loss": 0.70859063,
      "learning_rate": 3.174814134651671e-06,
      "loss": 0.73059708,
      "num_input_tokens_seen": 57532430,
      "step": 2671,
      "time_per_iteration": 2.6753392219543457
    },
    {
      "auxiliary_loss_clip": 0.0119022,
      "auxiliary_loss_mlp": 0.01029587,
      "balance_loss_clip": 1.05826557,
      "balance_loss_mlp": 1.0209918,
      "epoch": 0.3212890037876511,
      "flos": 21979233912960.0,
      "grad_norm": 1.9115989746728388,
      "language_loss": 0.80422068,
      "learning_rate": 3.1741836303267215e-06,
      "loss": 0.82641876,
      "num_input_tokens_seen": 57551965,
      "step": 2672,
      "time_per_iteration": 2.563977003097534
    },
    {
      "auxiliary_loss_clip": 0.01192452,
      "auxiliary_loss_mlp": 0.01029348,
      "balance_loss_clip": 1.05839014,
      "balance_loss_mlp": 1.02112865,
      "epoch": 0.32140924667829013,
      "flos": 10342345875840.0,
      "grad_norm": 1.8069210179413981,
      "language_loss": 0.74914384,
      "learning_rate": 3.1735529478790496e-06,
      "loss": 0.77136183,
      "num_input_tokens_seen": 57569955,
      "step": 2673,
      "time_per_iteration": 2.4567229747772217
    },
    {
      "auxiliary_loss_clip": 0.01181318,
      "auxiliary_loss_mlp": 0.01036346,
      "balance_loss_clip": 1.05734229,
      "balance_loss_mlp": 1.02689862,
      "epoch": 0.32152948956892924,
      "flos": 50798910072960.0,
      "grad_norm": 2.121830759928693,
      "language_loss": 0.79461741,
      "learning_rate": 3.172922087404328e-06,
      "loss": 0.81679416,
      "num_input_tokens_seen": 57592215,
      "step": 2674,
      "time_per_iteration": 2.7686643600463867
    },
    {
      "auxiliary_loss_clip": 0.01091215,
      "auxiliary_loss_mlp": 0.01002056,
      "balance_loss_clip": 1.02087653,
      "balance_loss_mlp": 1.00110841,
      "epoch": 0.32164973245956835,
      "flos": 63863250549120.0,
      "grad_norm": 0.7683724866515861,
      "language_loss": 0.55282241,
      "learning_rate": 3.1722910489982586e-06,
      "loss": 0.57375509,
      "num_input_tokens_seen": 57652575,
      "step": 2675,
      "time_per_iteration": 3.1243398189544678
    },
    {
      "auxiliary_loss_clip": 0.01159543,
      "auxiliary_loss_mlp": 0.01033774,
      "balance_loss_clip": 1.05268073,
      "balance_loss_mlp": 1.02466655,
      "epoch": 0.3217699753502074,
      "flos": 23513948363520.0,
      "grad_norm": 1.482625143549675,
      "language_loss": 0.80013525,
      "learning_rate": 3.1716598327565694e-06,
      "loss": 0.82206845,
      "num_input_tokens_seen": 57672215,
      "step": 2676,
      "time_per_iteration": 2.5619418621063232
    },
    {
      "auxiliary_loss_clip": 0.01190618,
      "auxiliary_loss_mlp": 0.01030638,
      "balance_loss_clip": 1.05724406,
      "balance_loss_mlp": 1.02173281,
      "epoch": 0.3218902182408465,
      "flos": 19062533640960.0,
      "grad_norm": 1.5643073483380618,
      "language_loss": 0.84217495,
      "learning_rate": 3.171028438775015e-06,
      "loss": 0.86438751,
      "num_input_tokens_seen": 57691410,
      "step": 2677,
      "time_per_iteration": 2.4965391159057617
    },
    {
      "auxiliary_loss_clip": 0.01192187,
      "auxiliary_loss_mlp": 0.01029001,
      "balance_loss_clip": 1.05769992,
      "balance_loss_mlp": 1.02070999,
      "epoch": 0.3220104611314856,
      "flos": 20375571306240.0,
      "grad_norm": 1.858680983702436,
      "language_loss": 0.84208,
      "learning_rate": 3.170396867149377e-06,
      "loss": 0.86429185,
      "num_input_tokens_seen": 57709415,
      "step": 2678,
      "time_per_iteration": 2.4523134231567383
    },
    {
      "auxiliary_loss_clip": 0.01130078,
      "auxiliary_loss_mlp": 0.01036336,
      "balance_loss_clip": 1.05200458,
      "balance_loss_mlp": 1.02751482,
      "epoch": 0.3221307040221247,
      "flos": 20117014231680.0,
      "grad_norm": 1.7938712894813198,
      "language_loss": 0.86117458,
      "learning_rate": 3.1697651179754653e-06,
      "loss": 0.88283873,
      "num_input_tokens_seen": 57728075,
      "step": 2679,
      "time_per_iteration": 2.612335205078125
    },
    {
      "auxiliary_loss_clip": 0.01152921,
      "auxiliary_loss_mlp": 0.01031333,
      "balance_loss_clip": 1.05902421,
      "balance_loss_mlp": 1.02272058,
      "epoch": 0.3222509469127638,
      "flos": 23987789602560.0,
      "grad_norm": 1.782525146767077,
      "language_loss": 0.73276806,
      "learning_rate": 3.1691331913491153e-06,
      "loss": 0.75461054,
      "num_input_tokens_seen": 57750645,
      "step": 2680,
      "time_per_iteration": 2.628188371658325
    },
    {
      "auxiliary_loss_clip": 0.0119234,
      "auxiliary_loss_mlp": 0.01027515,
      "balance_loss_clip": 1.05584121,
      "balance_loss_mlp": 1.01950371,
      "epoch": 0.32237118980340285,
      "flos": 17675735397120.0,
      "grad_norm": 1.990361221118503,
      "language_loss": 0.84645855,
      "learning_rate": 3.1685010873661898e-06,
      "loss": 0.86865705,
      "num_input_tokens_seen": 57769820,
      "step": 2681,
      "time_per_iteration": 2.4477040767669678
    },
    {
      "auxiliary_loss_clip": 0.01177423,
      "auxiliary_loss_mlp": 0.01032369,
      "balance_loss_clip": 1.05705523,
      "balance_loss_mlp": 1.0229274,
      "epoch": 0.32249143269404196,
      "flos": 23147982645120.0,
      "grad_norm": 1.9073689939583216,
      "language_loss": 0.79787964,
      "learning_rate": 3.167868806122578e-06,
      "loss": 0.81997758,
      "num_input_tokens_seen": 57788870,
      "step": 2682,
      "time_per_iteration": 2.5127296447753906
    },
    {
      "auxiliary_loss_clip": 0.01170678,
      "auxiliary_loss_mlp": 0.010305,
      "balance_loss_clip": 1.0571593,
      "balance_loss_mlp": 1.02176201,
      "epoch": 0.32261167558468107,
      "flos": 24422308427520.0,
      "grad_norm": 1.7893641358590515,
      "language_loss": 0.66334766,
      "learning_rate": 3.1672363477141968e-06,
      "loss": 0.68535942,
      "num_input_tokens_seen": 57808165,
      "step": 2683,
      "time_per_iteration": 3.403834581375122
    },
    {
      "auxiliary_loss_clip": 0.01167707,
      "auxiliary_loss_mlp": 0.01034313,
      "balance_loss_clip": 1.05328894,
      "balance_loss_mlp": 1.02533042,
      "epoch": 0.3227319184753201,
      "flos": 30367175852160.0,
      "grad_norm": 1.7531527269139193,
      "language_loss": 0.85104823,
      "learning_rate": 3.1666037122369903e-06,
      "loss": 0.87306839,
      "num_input_tokens_seen": 57828825,
      "step": 2684,
      "time_per_iteration": 2.5829105377197266
    },
    {
      "auxiliary_loss_clip": 0.01177012,
      "auxiliary_loss_mlp": 0.01028924,
      "balance_loss_clip": 1.05405569,
      "balance_loss_mlp": 1.02074575,
      "epoch": 0.32285216136595923,
      "flos": 16946174257920.0,
      "grad_norm": 2.0704149156554,
      "language_loss": 0.86442816,
      "learning_rate": 3.165970899786928e-06,
      "loss": 0.8864876,
      "num_input_tokens_seen": 57846740,
      "step": 2685,
      "time_per_iteration": 2.5022048950195312
    },
    {
      "auxiliary_loss_clip": 0.01154311,
      "auxiliary_loss_mlp": 0.01028351,
      "balance_loss_clip": 1.05366087,
      "balance_loss_mlp": 1.02008939,
      "epoch": 0.32297240425659834,
      "flos": 21981532383360.0,
      "grad_norm": 1.6916691925977374,
      "language_loss": 0.75412464,
      "learning_rate": 3.1653379104600067e-06,
      "loss": 0.77595121,
      "num_input_tokens_seen": 57866885,
      "step": 2686,
      "time_per_iteration": 2.5646004676818848
    },
    {
      "auxiliary_loss_clip": 0.01176373,
      "auxiliary_loss_mlp": 0.01029921,
      "balance_loss_clip": 1.05480671,
      "balance_loss_mlp": 1.02155828,
      "epoch": 0.3230926471472374,
      "flos": 22748045639040.0,
      "grad_norm": 2.331889449696419,
      "language_loss": 0.69665074,
      "learning_rate": 3.164704744352251e-06,
      "loss": 0.7187137,
      "num_input_tokens_seen": 57887690,
      "step": 2687,
      "time_per_iteration": 2.5375418663024902
    },
    {
      "auxiliary_loss_clip": 0.011763,
      "auxiliary_loss_mlp": 0.01032012,
      "balance_loss_clip": 1.05488431,
      "balance_loss_mlp": 1.02422166,
      "epoch": 0.3232128900378765,
      "flos": 16942977947520.0,
      "grad_norm": 2.105842305246496,
      "language_loss": 0.80590993,
      "learning_rate": 3.164071401559713e-06,
      "loss": 0.82799304,
      "num_input_tokens_seen": 57905090,
      "step": 2688,
      "time_per_iteration": 3.3633596897125244
    },
    {
      "auxiliary_loss_clip": 0.01164644,
      "auxiliary_loss_mlp": 0.01032487,
      "balance_loss_clip": 1.054299,
      "balance_loss_mlp": 1.02410626,
      "epoch": 0.3233331329285156,
      "flos": 24023736138240.0,
      "grad_norm": 1.6717343602788888,
      "language_loss": 0.70919359,
      "learning_rate": 3.1634378821784674e-06,
      "loss": 0.73116487,
      "num_input_tokens_seen": 57925305,
      "step": 2689,
      "time_per_iteration": 4.2222888469696045
    },
    {
      "auxiliary_loss_clip": 0.01153321,
      "auxiliary_loss_mlp": 0.01034456,
      "balance_loss_clip": 1.05504024,
      "balance_loss_mlp": 1.02609301,
      "epoch": 0.3234533758191547,
      "flos": 18113845582080.0,
      "grad_norm": 2.4492995973323644,
      "language_loss": 0.74481857,
      "learning_rate": 3.1628041863046208e-06,
      "loss": 0.76669633,
      "num_input_tokens_seen": 57942720,
      "step": 2690,
      "time_per_iteration": 2.558166265487671
    },
    {
      "auxiliary_loss_clip": 0.01196038,
      "auxiliary_loss_mlp": 0.01032378,
      "balance_loss_clip": 1.05574393,
      "balance_loss_mlp": 1.02316892,
      "epoch": 0.3235736187097938,
      "flos": 16946138344320.0,
      "grad_norm": 2.21581339529008,
      "language_loss": 0.90908998,
      "learning_rate": 3.162170314034304e-06,
      "loss": 0.93137413,
      "num_input_tokens_seen": 57960135,
      "step": 2691,
      "time_per_iteration": 2.4545295238494873
    },
    {
      "auxiliary_loss_clip": 0.01194613,
      "auxiliary_loss_mlp": 0.01032584,
      "balance_loss_clip": 1.05597401,
      "balance_loss_mlp": 1.02337503,
      "epoch": 0.3236938616004329,
      "flos": 22127150119680.0,
      "grad_norm": 1.69881504924168,
      "language_loss": 0.80864364,
      "learning_rate": 3.1615362654636738e-06,
      "loss": 0.83091563,
      "num_input_tokens_seen": 57980875,
      "step": 2692,
      "time_per_iteration": 2.4704015254974365
    },
    {
      "auxiliary_loss_clip": 0.01143242,
      "auxiliary_loss_mlp": 0.01033969,
      "balance_loss_clip": 1.05474329,
      "balance_loss_mlp": 1.02554703,
      "epoch": 0.32381410449107195,
      "flos": 17164618819200.0,
      "grad_norm": 1.6989526200457512,
      "language_loss": 0.87463701,
      "learning_rate": 3.1609020406889163e-06,
      "loss": 0.89640915,
      "num_input_tokens_seen": 57998310,
      "step": 2693,
      "time_per_iteration": 2.5160059928894043
    },
    {
      "auxiliary_loss_clip": 0.01167696,
      "auxiliary_loss_mlp": 0.01033677,
      "balance_loss_clip": 1.05462456,
      "balance_loss_mlp": 1.02396131,
      "epoch": 0.32393434738171106,
      "flos": 16578125550720.0,
      "grad_norm": 1.6470579250626114,
      "language_loss": 0.85137844,
      "learning_rate": 3.1602676398062416e-06,
      "loss": 0.87339216,
      "num_input_tokens_seen": 58017220,
      "step": 2694,
      "time_per_iteration": 2.51727294921875
    },
    {
      "auxiliary_loss_clip": 0.0117627,
      "auxiliary_loss_mlp": 0.01028129,
      "balance_loss_clip": 1.05586851,
      "balance_loss_mlp": 1.0191462,
      "epoch": 0.3240545902723502,
      "flos": 25483612602240.0,
      "grad_norm": 2.051156024763937,
      "language_loss": 0.61586684,
      "learning_rate": 3.1596330629118886e-06,
      "loss": 0.63791084,
      "num_input_tokens_seen": 58037190,
      "step": 2695,
      "time_per_iteration": 2.571971893310547
    },
    {
      "auxiliary_loss_clip": 0.01131357,
      "auxiliary_loss_mlp": 0.01031594,
      "balance_loss_clip": 1.0513432,
      "balance_loss_mlp": 1.02285552,
      "epoch": 0.32417483316298923,
      "flos": 35845851634560.0,
      "grad_norm": 1.9871282258802616,
      "language_loss": 0.73243213,
      "learning_rate": 3.1589983101021223e-06,
      "loss": 0.75406164,
      "num_input_tokens_seen": 58055820,
      "step": 2696,
      "time_per_iteration": 2.725529909133911
    },
    {
      "auxiliary_loss_clip": 0.01165154,
      "auxiliary_loss_mlp": 0.01028393,
      "balance_loss_clip": 1.05454111,
      "balance_loss_mlp": 1.01994681,
      "epoch": 0.32429507605362834,
      "flos": 30080501406720.0,
      "grad_norm": 2.3291311033144093,
      "language_loss": 0.85074127,
      "learning_rate": 3.1583633814732337e-06,
      "loss": 0.87267673,
      "num_input_tokens_seen": 58075340,
      "step": 2697,
      "time_per_iteration": 2.5923664569854736
    },
    {
      "auxiliary_loss_clip": 0.01190134,
      "auxiliary_loss_mlp": 0.01035893,
      "balance_loss_clip": 1.05493593,
      "balance_loss_mlp": 1.02710152,
      "epoch": 0.3244153189442674,
      "flos": 18223265387520.0,
      "grad_norm": 2.7287229313480172,
      "language_loss": 0.71703166,
      "learning_rate": 3.157728277121541e-06,
      "loss": 0.73929197,
      "num_input_tokens_seen": 58093515,
      "step": 2698,
      "time_per_iteration": 2.4554076194763184
    },
    {
      "auxiliary_loss_clip": 0.01190914,
      "auxiliary_loss_mlp": 0.01028687,
      "balance_loss_clip": 1.05290318,
      "balance_loss_mlp": 1.01988959,
      "epoch": 0.3245355618349065,
      "flos": 17710317216000.0,
      "grad_norm": 2.557176730795796,
      "language_loss": 0.78560084,
      "learning_rate": 3.1570929971433897e-06,
      "loss": 0.80779684,
      "num_input_tokens_seen": 58109300,
      "step": 2699,
      "time_per_iteration": 2.4922051429748535
    },
    {
      "auxiliary_loss_clip": 0.01178222,
      "auxiliary_loss_mlp": 0.01034697,
      "balance_loss_clip": 1.05738342,
      "balance_loss_mlp": 1.02638793,
      "epoch": 0.3246558047255456,
      "flos": 23440798316160.0,
      "grad_norm": 2.3486320779730523,
      "language_loss": 0.83945686,
      "learning_rate": 3.1564575416351504e-06,
      "loss": 0.86158609,
      "num_input_tokens_seen": 58128000,
      "step": 2700,
      "time_per_iteration": 2.5409977436065674
    },
    {
      "auxiliary_loss_clip": 0.01192635,
      "auxiliary_loss_mlp": 0.01028964,
      "balance_loss_clip": 1.05661488,
      "balance_loss_mlp": 1.01995742,
      "epoch": 0.32477604761618467,
      "flos": 21760861178880.0,
      "grad_norm": 1.8382055656498018,
      "language_loss": 0.74573141,
      "learning_rate": 3.155821910693221e-06,
      "loss": 0.76794744,
      "num_input_tokens_seen": 58147415,
      "step": 2701,
      "time_per_iteration": 2.4598922729492188
    },
    {
      "auxiliary_loss_clip": 0.01161604,
      "auxiliary_loss_mlp": 0.01030281,
      "balance_loss_clip": 1.05148828,
      "balance_loss_mlp": 1.02149582,
      "epoch": 0.3248962905068238,
      "flos": 19828328624640.0,
      "grad_norm": 1.77561343765119,
      "language_loss": 0.85860306,
      "learning_rate": 3.1551861044140275e-06,
      "loss": 0.88052189,
      "num_input_tokens_seen": 58167050,
      "step": 2702,
      "time_per_iteration": 2.545163869857788
    },
    {
      "auxiliary_loss_clip": 0.01130465,
      "auxiliary_loss_mlp": 0.01030933,
      "balance_loss_clip": 1.0518229,
      "balance_loss_mlp": 1.0226537,
      "epoch": 0.3250165333974629,
      "flos": 23948215793280.0,
      "grad_norm": 2.518944028727769,
      "language_loss": 0.77937633,
      "learning_rate": 3.15455012289402e-06,
      "loss": 0.80099034,
      "num_input_tokens_seen": 58186695,
      "step": 2703,
      "time_per_iteration": 2.6312096118927
    },
    {
      "auxiliary_loss_clip": 0.01179474,
      "auxiliary_loss_mlp": 0.01029946,
      "balance_loss_clip": 1.05674911,
      "balance_loss_mlp": 1.02092755,
      "epoch": 0.32513677628810195,
      "flos": 23989333887360.0,
      "grad_norm": 1.8105436482824648,
      "language_loss": 0.84446788,
      "learning_rate": 3.153913966229677e-06,
      "loss": 0.86656213,
      "num_input_tokens_seen": 58205815,
      "step": 2704,
      "time_per_iteration": 2.520097255706787
    },
    {
      "auxiliary_loss_clip": 0.01082358,
      "auxiliary_loss_mlp": 0.01004924,
      "balance_loss_clip": 1.02118278,
      "balance_loss_mlp": 1.00370789,
      "epoch": 0.32525701917874106,
      "flos": 70655790009600.0,
      "grad_norm": 0.6400157666945211,
      "language_loss": 0.50297797,
      "learning_rate": 3.1532776345175027e-06,
      "loss": 0.5238508,
      "num_input_tokens_seen": 58270960,
      "step": 2705,
      "time_per_iteration": 3.094642162322998
    },
    {
      "auxiliary_loss_clip": 0.01190474,
      "auxiliary_loss_mlp": 0.01034261,
      "balance_loss_clip": 1.05572164,
      "balance_loss_mlp": 1.02524841,
      "epoch": 0.32537726206938017,
      "flos": 19682639061120.0,
      "grad_norm": 1.9398741628678617,
      "language_loss": 0.78877556,
      "learning_rate": 3.1526411278540285e-06,
      "loss": 0.81102294,
      "num_input_tokens_seen": 58289390,
      "step": 2706,
      "time_per_iteration": 2.4604310989379883
    },
    {
      "auxiliary_loss_clip": 0.01170815,
      "auxiliary_loss_mlp": 0.01033581,
      "balance_loss_clip": 1.05296147,
      "balance_loss_mlp": 1.02412748,
      "epoch": 0.3254975049600192,
      "flos": 28760999293440.0,
      "grad_norm": 2.6649028613281573,
      "language_loss": 0.8103748,
      "learning_rate": 3.1520044463358116e-06,
      "loss": 0.83241868,
      "num_input_tokens_seen": 58306120,
      "step": 2707,
      "time_per_iteration": 2.578207015991211
    },
    {
      "auxiliary_loss_clip": 0.01177063,
      "auxiliary_loss_mlp": 0.01028943,
      "balance_loss_clip": 1.05536747,
      "balance_loss_mlp": 1.01990354,
      "epoch": 0.32561774785065833,
      "flos": 18877378008960.0,
      "grad_norm": 1.5542523578386402,
      "language_loss": 0.80113983,
      "learning_rate": 3.151367590059436e-06,
      "loss": 0.82319987,
      "num_input_tokens_seen": 58324545,
      "step": 2708,
      "time_per_iteration": 2.4830994606018066
    },
    {
      "auxiliary_loss_clip": 0.01193839,
      "auxiliary_loss_mlp": 0.00763984,
      "balance_loss_clip": 1.05644178,
      "balance_loss_mlp": 1.00061584,
      "epoch": 0.32573799074129745,
      "flos": 23112107936640.0,
      "grad_norm": 1.9158896819042956,
      "language_loss": 0.86814183,
      "learning_rate": 3.1507305591215117e-06,
      "loss": 0.88772005,
      "num_input_tokens_seen": 58342455,
      "step": 2709,
      "time_per_iteration": 2.5095303058624268
    },
    {
      "auxiliary_loss_clip": 0.01081599,
      "auxiliary_loss_mlp": 0.01000387,
      "balance_loss_clip": 1.02072513,
      "balance_loss_mlp": 0.99936146,
      "epoch": 0.3258582336319365,
      "flos": 71237650423680.0,
      "grad_norm": 0.6710497388181448,
      "language_loss": 0.55728269,
      "learning_rate": 3.150093353618677e-06,
      "loss": 0.57810253,
      "num_input_tokens_seen": 58407185,
      "step": 2710,
      "time_per_iteration": 3.9900460243225098
    },
    {
      "auxiliary_loss_clip": 0.01179995,
      "auxiliary_loss_mlp": 0.01028208,
      "balance_loss_clip": 1.05258071,
      "balance_loss_mlp": 1.01936841,
      "epoch": 0.3259784765225756,
      "flos": 22456020067200.0,
      "grad_norm": 2.3641429077897764,
      "language_loss": 0.88119578,
      "learning_rate": 3.149455973647596e-06,
      "loss": 0.90327775,
      "num_input_tokens_seen": 58425245,
      "step": 2711,
      "time_per_iteration": 2.496067762374878
    },
    {
      "auxiliary_loss_clip": 0.01139049,
      "auxiliary_loss_mlp": 0.01028339,
      "balance_loss_clip": 1.04591322,
      "balance_loss_mlp": 1.01880252,
      "epoch": 0.32609871941321467,
      "flos": 20484811543680.0,
      "grad_norm": 1.7644762406713717,
      "language_loss": 0.77162302,
      "learning_rate": 3.1488184193049563e-06,
      "loss": 0.79329693,
      "num_input_tokens_seen": 58444780,
      "step": 2712,
      "time_per_iteration": 2.5653631687164307
    },
    {
      "auxiliary_loss_clip": 0.0119313,
      "auxiliary_loss_mlp": 0.01029374,
      "balance_loss_clip": 1.05860734,
      "balance_loss_mlp": 1.02093983,
      "epoch": 0.3262189623038538,
      "flos": 22416805393920.0,
      "grad_norm": 1.5234617118566582,
      "language_loss": 0.72071731,
      "learning_rate": 3.1481806906874767e-06,
      "loss": 0.74294233,
      "num_input_tokens_seen": 58466090,
      "step": 2713,
      "time_per_iteration": 2.537545680999756
    },
    {
      "auxiliary_loss_clip": 0.01191152,
      "auxiliary_loss_mlp": 0.01029131,
      "balance_loss_clip": 1.05639124,
      "balance_loss_mlp": 1.02097738,
      "epoch": 0.3263392051944929,
      "flos": 20923496346240.0,
      "grad_norm": 1.5222323815950405,
      "language_loss": 0.87659419,
      "learning_rate": 3.147542787891899e-06,
      "loss": 0.89879698,
      "num_input_tokens_seen": 58485435,
      "step": 2714,
      "time_per_iteration": 2.4770402908325195
    },
    {
      "auxiliary_loss_clip": 0.01165721,
      "auxiliary_loss_mlp": 0.01034286,
      "balance_loss_clip": 1.05646634,
      "balance_loss_mlp": 1.02497625,
      "epoch": 0.32645944808513194,
      "flos": 24025172682240.0,
      "grad_norm": 2.226836444538743,
      "language_loss": 0.75280476,
      "learning_rate": 3.1469047110149926e-06,
      "loss": 0.77480483,
      "num_input_tokens_seen": 58504175,
      "step": 2715,
      "time_per_iteration": 3.446967363357544
    },
    {
      "auxiliary_loss_clip": 0.01131137,
      "auxiliary_loss_mlp": 0.01028571,
      "balance_loss_clip": 1.05160069,
      "balance_loss_mlp": 1.02020264,
      "epoch": 0.32657969097577105,
      "flos": 21032413361280.0,
      "grad_norm": 1.9436120964076988,
      "language_loss": 0.85275984,
      "learning_rate": 3.146266460153554e-06,
      "loss": 0.87435687,
      "num_input_tokens_seen": 58523885,
      "step": 2716,
      "time_per_iteration": 3.363856077194214
    },
    {
      "auxiliary_loss_clip": 0.01162522,
      "auxiliary_loss_mlp": 0.00763834,
      "balance_loss_clip": 1.05383217,
      "balance_loss_mlp": 1.00063705,
      "epoch": 0.32669993386641016,
      "flos": 22710267509760.0,
      "grad_norm": 1.8365836094478587,
      "language_loss": 0.80179358,
      "learning_rate": 3.145628035404404e-06,
      "loss": 0.82105708,
      "num_input_tokens_seen": 58543085,
      "step": 2717,
      "time_per_iteration": 2.5573678016662598
    },
    {
      "auxiliary_loss_clip": 0.01079784,
      "auxiliary_loss_mlp": 0.01004353,
      "balance_loss_clip": 1.02022147,
      "balance_loss_mlp": 1.00322008,
      "epoch": 0.3268201767570492,
      "flos": 72105718406400.0,
      "grad_norm": 0.8800989363533821,
      "language_loss": 0.57526404,
      "learning_rate": 3.1449894368643922e-06,
      "loss": 0.5961054,
      "num_input_tokens_seen": 58605400,
      "step": 2718,
      "time_per_iteration": 3.1740405559539795
    },
    {
      "auxiliary_loss_clip": 0.01149127,
      "auxiliary_loss_mlp": 0.0103224,
      "balance_loss_clip": 1.05382383,
      "balance_loss_mlp": 1.02353776,
      "epoch": 0.32694041964768833,
      "flos": 24535175938560.0,
      "grad_norm": 1.4842212095007867,
      "language_loss": 0.71396226,
      "learning_rate": 3.1443506646303934e-06,
      "loss": 0.73577595,
      "num_input_tokens_seen": 58626700,
      "step": 2719,
      "time_per_iteration": 2.5880401134490967
    },
    {
      "auxiliary_loss_clip": 0.01181125,
      "auxiliary_loss_mlp": 0.0102859,
      "balance_loss_clip": 1.05421257,
      "balance_loss_mlp": 1.01957762,
      "epoch": 0.32706066253832744,
      "flos": 33183003755520.0,
      "grad_norm": 3.096262231864864,
      "language_loss": 0.66281915,
      "learning_rate": 3.1437117187993086e-06,
      "loss": 0.68491626,
      "num_input_tokens_seen": 58649020,
      "step": 2720,
      "time_per_iteration": 2.5920658111572266
    },
    {
      "auxiliary_loss_clip": 0.01142924,
      "auxiliary_loss_mlp": 0.01036316,
      "balance_loss_clip": 1.04939818,
      "balance_loss_mlp": 1.0276618,
      "epoch": 0.3271809054289665,
      "flos": 24061622008320.0,
      "grad_norm": 1.6352889588221644,
      "language_loss": 0.79710084,
      "learning_rate": 3.143072599468065e-06,
      "loss": 0.81889319,
      "num_input_tokens_seen": 58668845,
      "step": 2721,
      "time_per_iteration": 2.5829741954803467
    },
    {
      "auxiliary_loss_clip": 0.01162481,
      "auxiliary_loss_mlp": 0.01026562,
      "balance_loss_clip": 1.05459476,
      "balance_loss_mlp": 1.01813388,
      "epoch": 0.3273011483196056,
      "flos": 38253769712640.0,
      "grad_norm": 1.5041987902718992,
      "language_loss": 0.75658202,
      "learning_rate": 3.1424333067336174e-06,
      "loss": 0.77847242,
      "num_input_tokens_seen": 58691610,
      "step": 2722,
      "time_per_iteration": 2.6655757427215576
    },
    {
      "auxiliary_loss_clip": 0.01182258,
      "auxiliary_loss_mlp": 0.01034431,
      "balance_loss_clip": 1.05412769,
      "balance_loss_mlp": 1.02504921,
      "epoch": 0.3274213912102447,
      "flos": 29054389582080.0,
      "grad_norm": 1.7790416177480768,
      "language_loss": 0.78194147,
      "learning_rate": 3.141793840692945e-06,
      "loss": 0.80410838,
      "num_input_tokens_seen": 58712360,
      "step": 2723,
      "time_per_iteration": 2.5465664863586426
    },
    {
      "auxiliary_loss_clip": 0.01155013,
      "auxiliary_loss_mlp": 0.01031634,
      "balance_loss_clip": 1.05115497,
      "balance_loss_mlp": 1.02207315,
      "epoch": 0.32754163410088377,
      "flos": 29133249891840.0,
      "grad_norm": 2.141415948292631,
      "language_loss": 0.61650538,
      "learning_rate": 3.1411542014430553e-06,
      "loss": 0.63837183,
      "num_input_tokens_seen": 58733440,
      "step": 2724,
      "time_per_iteration": 2.591707706451416
    },
    {
      "auxiliary_loss_clip": 0.01145698,
      "auxiliary_loss_mlp": 0.01031293,
      "balance_loss_clip": 1.04682374,
      "balance_loss_mlp": 1.02284694,
      "epoch": 0.3276618769915229,
      "flos": 20631075724800.0,
      "grad_norm": 1.6898622287564375,
      "language_loss": 0.8172431,
      "learning_rate": 3.1405143890809804e-06,
      "loss": 0.83901304,
      "num_input_tokens_seen": 58752735,
      "step": 2725,
      "time_per_iteration": 2.615417003631592
    },
    {
      "auxiliary_loss_clip": 0.01161647,
      "auxiliary_loss_mlp": 0.01027014,
      "balance_loss_clip": 1.05358028,
      "balance_loss_mlp": 1.01898193,
      "epoch": 0.327782119882162,
      "flos": 18657425076480.0,
      "grad_norm": 1.7748133468683849,
      "language_loss": 0.69959855,
      "learning_rate": 3.1398744037037796e-06,
      "loss": 0.72148514,
      "num_input_tokens_seen": 58772070,
      "step": 2726,
      "time_per_iteration": 2.5282788276672363
    },
    {
      "auxiliary_loss_clip": 0.01162873,
      "auxiliary_loss_mlp": 0.01029437,
      "balance_loss_clip": 1.05430639,
      "balance_loss_mlp": 1.02123523,
      "epoch": 0.32790236277280105,
      "flos": 21795802133760.0,
      "grad_norm": 3.5058638702400886,
      "language_loss": 0.84428805,
      "learning_rate": 3.139234245408538e-06,
      "loss": 0.86621118,
      "num_input_tokens_seen": 58790950,
      "step": 2727,
      "time_per_iteration": 2.5232949256896973
    },
    {
      "auxiliary_loss_clip": 0.01150785,
      "auxiliary_loss_mlp": 0.00763251,
      "balance_loss_clip": 1.0541718,
      "balance_loss_mlp": 1.0006001,
      "epoch": 0.32802260566344016,
      "flos": 23331414424320.0,
      "grad_norm": 1.4027983021834594,
      "language_loss": 0.76069951,
      "learning_rate": 3.1385939142923666e-06,
      "loss": 0.77983987,
      "num_input_tokens_seen": 58813340,
      "step": 2728,
      "time_per_iteration": 2.567537784576416
    },
    {
      "auxiliary_loss_clip": 0.01164984,
      "auxiliary_loss_mlp": 0.01032246,
      "balance_loss_clip": 1.05145645,
      "balance_loss_mlp": 1.02325797,
      "epoch": 0.3281428485540792,
      "flos": 24206988349440.0,
      "grad_norm": 1.961203247980162,
      "language_loss": 0.78408444,
      "learning_rate": 3.137953410452405e-06,
      "loss": 0.80605674,
      "num_input_tokens_seen": 58833610,
      "step": 2729,
      "time_per_iteration": 2.545344829559326
    },
    {
      "auxiliary_loss_clip": 0.01158294,
      "auxiliary_loss_mlp": 0.01033674,
      "balance_loss_clip": 1.0496335,
      "balance_loss_mlp": 1.02546036,
      "epoch": 0.3282630914447183,
      "flos": 34128962380800.0,
      "grad_norm": 1.6421596659617923,
      "language_loss": 0.74445462,
      "learning_rate": 3.1373127339858146e-06,
      "loss": 0.76637435,
      "num_input_tokens_seen": 58856210,
      "step": 2730,
      "time_per_iteration": 2.644753932952881
    },
    {
      "auxiliary_loss_clip": 0.01142428,
      "auxiliary_loss_mlp": 0.01027631,
      "balance_loss_clip": 1.04760504,
      "balance_loss_mlp": 1.02037692,
      "epoch": 0.32838333433535744,
      "flos": 27600726170880.0,
      "grad_norm": 1.8103094504922241,
      "language_loss": 0.74423331,
      "learning_rate": 3.136671884989787e-06,
      "loss": 0.76593387,
      "num_input_tokens_seen": 58876120,
      "step": 2731,
      "time_per_iteration": 2.620925188064575
    },
    {
      "auxiliary_loss_clip": 0.0112396,
      "auxiliary_loss_mlp": 0.01029609,
      "balance_loss_clip": 1.04842675,
      "balance_loss_mlp": 1.02072811,
      "epoch": 0.3285035772259965,
      "flos": 12349500935040.0,
      "grad_norm": 2.1177205652376565,
      "language_loss": 0.87565303,
      "learning_rate": 3.1360308635615383e-06,
      "loss": 0.89718866,
      "num_input_tokens_seen": 58894660,
      "step": 2732,
      "time_per_iteration": 2.6295647621154785
    },
    {
      "auxiliary_loss_clip": 0.0116973,
      "auxiliary_loss_mlp": 0.0102941,
      "balance_loss_clip": 1.05275822,
      "balance_loss_mlp": 1.01976609,
      "epoch": 0.3286238201166356,
      "flos": 24316084932480.0,
      "grad_norm": 1.9301541430485614,
      "language_loss": 0.78520918,
      "learning_rate": 3.135389669798311e-06,
      "loss": 0.80720055,
      "num_input_tokens_seen": 58912720,
      "step": 2733,
      "time_per_iteration": 2.5923287868499756
    },
    {
      "auxiliary_loss_clip": 0.0117521,
      "auxiliary_loss_mlp": 0.00763201,
      "balance_loss_clip": 1.05306721,
      "balance_loss_mlp": 1.00057244,
      "epoch": 0.3287440630072747,
      "flos": 21392812471680.0,
      "grad_norm": 1.7195785694770358,
      "language_loss": 0.80045044,
      "learning_rate": 3.134748303797373e-06,
      "loss": 0.81983453,
      "num_input_tokens_seen": 58930090,
      "step": 2734,
      "time_per_iteration": 2.5224335193634033
    },
    {
      "auxiliary_loss_clip": 0.01136794,
      "auxiliary_loss_mlp": 0.01032068,
      "balance_loss_clip": 1.04885435,
      "balance_loss_mlp": 1.02257895,
      "epoch": 0.32886430589791377,
      "flos": 23732536579200.0,
      "grad_norm": 1.789089162321881,
      "language_loss": 0.81015676,
      "learning_rate": 3.1341067656560203e-06,
      "loss": 0.8318454,
      "num_input_tokens_seen": 58947935,
      "step": 2735,
      "time_per_iteration": 2.609121084213257
    },
    {
      "auxiliary_loss_clip": 0.01173713,
      "auxiliary_loss_mlp": 0.01030072,
      "balance_loss_clip": 1.05412519,
      "balance_loss_mlp": 1.02098203,
      "epoch": 0.3289845487885529,
      "flos": 22418708814720.0,
      "grad_norm": 2.005109041925455,
      "language_loss": 0.86447185,
      "learning_rate": 3.133465055471572e-06,
      "loss": 0.88650972,
      "num_input_tokens_seen": 58967720,
      "step": 2736,
      "time_per_iteration": 3.4520626068115234
    },
    {
      "auxiliary_loss_clip": 0.01143588,
      "auxiliary_loss_mlp": 0.01028348,
      "balance_loss_clip": 1.04840648,
      "balance_loss_mlp": 1.01983643,
      "epoch": 0.329104791679192,
      "flos": 19682603147520.0,
      "grad_norm": 3.9249019448475315,
      "language_loss": 0.66175836,
      "learning_rate": 3.1328231733413767e-06,
      "loss": 0.68347776,
      "num_input_tokens_seen": 58984360,
      "step": 2737,
      "time_per_iteration": 2.5393660068511963
    },
    {
      "auxiliary_loss_clip": 0.0117341,
      "auxiliary_loss_mlp": 0.01035398,
      "balance_loss_clip": 1.0530076,
      "balance_loss_mlp": 1.02602851,
      "epoch": 0.32922503456983104,
      "flos": 15997234803840.0,
      "grad_norm": 2.517977310543383,
      "language_loss": 0.91078639,
      "learning_rate": 3.1321811193628067e-06,
      "loss": 0.93287444,
      "num_input_tokens_seen": 59002505,
      "step": 2738,
      "time_per_iteration": 2.5558552742004395
    },
    {
      "auxiliary_loss_clip": 0.01180166,
      "auxiliary_loss_mlp": 0.00764131,
      "balance_loss_clip": 1.0562675,
      "balance_loss_mlp": 1.00049877,
      "epoch": 0.32934527746047015,
      "flos": 26834069260800.0,
      "grad_norm": 1.692209889409344,
      "language_loss": 0.69912708,
      "learning_rate": 3.131538893633261e-06,
      "loss": 0.71856999,
      "num_input_tokens_seen": 59022065,
      "step": 2739,
      "time_per_iteration": 2.58483624458313
    },
    {
      "auxiliary_loss_clip": 0.01193664,
      "auxiliary_loss_mlp": 0.01031653,
      "balance_loss_clip": 1.05687737,
      "balance_loss_mlp": 1.02320099,
      "epoch": 0.32946552035110926,
      "flos": 23403774372480.0,
      "grad_norm": 2.0316138901416703,
      "language_loss": 0.78344131,
      "learning_rate": 3.130896496250165e-06,
      "loss": 0.80569446,
      "num_input_tokens_seen": 59041890,
      "step": 2740,
      "time_per_iteration": 2.502199411392212
    },
    {
      "auxiliary_loss_clip": 0.01193416,
      "auxiliary_loss_mlp": 0.01030122,
      "balance_loss_clip": 1.05590367,
      "balance_loss_mlp": 1.02155089,
      "epoch": 0.3295857632417483,
      "flos": 14172470029440.0,
      "grad_norm": 2.057239755727636,
      "language_loss": 0.86492032,
      "learning_rate": 3.1302539273109693e-06,
      "loss": 0.88715571,
      "num_input_tokens_seen": 59058715,
      "step": 2741,
      "time_per_iteration": 4.060149669647217
    },
    {
      "auxiliary_loss_clip": 0.01158753,
      "auxiliary_loss_mlp": 0.01034995,
      "balance_loss_clip": 1.05486703,
      "balance_loss_mlp": 1.02543414,
      "epoch": 0.32970600613238743,
      "flos": 22196708807040.0,
      "grad_norm": 1.6136790967051624,
      "language_loss": 0.803855,
      "learning_rate": 3.1296111869131513e-06,
      "loss": 0.82579249,
      "num_input_tokens_seen": 59076140,
      "step": 2742,
      "time_per_iteration": 3.3554325103759766
    },
    {
      "auxiliary_loss_clip": 0.01192164,
      "auxiliary_loss_mlp": 0.01030759,
      "balance_loss_clip": 1.05575359,
      "balance_loss_mlp": 1.02224731,
      "epoch": 0.32982624902302654,
      "flos": 22053784590720.0,
      "grad_norm": 1.8611195951837967,
      "language_loss": 0.85763323,
      "learning_rate": 3.1289682751542153e-06,
      "loss": 0.87986243,
      "num_input_tokens_seen": 59095700,
      "step": 2743,
      "time_per_iteration": 2.464463233947754
    },
    {
      "auxiliary_loss_clip": 0.01176162,
      "auxiliary_loss_mlp": 0.0102958,
      "balance_loss_clip": 1.05371094,
      "balance_loss_mlp": 1.020872,
      "epoch": 0.3299464919136656,
      "flos": 18661626967680.0,
      "grad_norm": 1.9409555574641155,
      "language_loss": 0.71338344,
      "learning_rate": 3.1283251921316883e-06,
      "loss": 0.73544085,
      "num_input_tokens_seen": 59113445,
      "step": 2744,
      "time_per_iteration": 2.4867451190948486
    },
    {
      "auxiliary_loss_clip": 0.01134698,
      "auxiliary_loss_mlp": 0.01034318,
      "balance_loss_clip": 1.05118132,
      "balance_loss_mlp": 1.0244832,
      "epoch": 0.3300667348043047,
      "flos": 13407357404160.0,
      "grad_norm": 2.913798876066495,
      "language_loss": 0.81023258,
      "learning_rate": 3.1276819379431277e-06,
      "loss": 0.83192277,
      "num_input_tokens_seen": 59131535,
      "step": 2745,
      "time_per_iteration": 2.5624399185180664
    },
    {
      "auxiliary_loss_clip": 0.01173215,
      "auxiliary_loss_mlp": 0.00764189,
      "balance_loss_clip": 1.05453289,
      "balance_loss_mlp": 1.00064063,
      "epoch": 0.33018697769494376,
      "flos": 15742556398080.0,
      "grad_norm": 2.0936198590434225,
      "language_loss": 0.75473702,
      "learning_rate": 3.1270385126861134e-06,
      "loss": 0.77411103,
      "num_input_tokens_seen": 59149520,
      "step": 2746,
      "time_per_iteration": 2.4873788356781006
    },
    {
      "auxiliary_loss_clip": 0.01195702,
      "auxiliary_loss_mlp": 0.01034259,
      "balance_loss_clip": 1.0569973,
      "balance_loss_mlp": 1.02432871,
      "epoch": 0.3303072205855829,
      "flos": 18258601392000.0,
      "grad_norm": 1.8831606611305816,
      "language_loss": 0.82025903,
      "learning_rate": 3.1263949164582533e-06,
      "loss": 0.84255868,
      "num_input_tokens_seen": 59169170,
      "step": 2747,
      "time_per_iteration": 2.4668898582458496
    },
    {
      "auxiliary_loss_clip": 0.01192521,
      "auxiliary_loss_mlp": 0.01031519,
      "balance_loss_clip": 1.05309093,
      "balance_loss_mlp": 1.02266765,
      "epoch": 0.330427463476222,
      "flos": 17749424148480.0,
      "grad_norm": 4.539365216926317,
      "language_loss": 0.78308821,
      "learning_rate": 3.1257511493571797e-06,
      "loss": 0.80532861,
      "num_input_tokens_seen": 59187675,
      "step": 2748,
      "time_per_iteration": 2.4170444011688232
    },
    {
      "auxiliary_loss_clip": 0.01150153,
      "auxiliary_loss_mlp": 0.01031558,
      "balance_loss_clip": 1.05056405,
      "balance_loss_mlp": 1.02291489,
      "epoch": 0.33054770636686104,
      "flos": 27162580072320.0,
      "grad_norm": 1.9490683824747084,
      "language_loss": 0.78467047,
      "learning_rate": 3.125107211480552e-06,
      "loss": 0.80648762,
      "num_input_tokens_seen": 59207610,
      "step": 2749,
      "time_per_iteration": 2.579108476638794
    },
    {
      "auxiliary_loss_clip": 0.01119104,
      "auxiliary_loss_mlp": 0.01037455,
      "balance_loss_clip": 1.04829955,
      "balance_loss_mlp": 1.02853203,
      "epoch": 0.33066794925750015,
      "flos": 20117193799680.0,
      "grad_norm": 1.7075292249442584,
      "language_loss": 0.79815769,
      "learning_rate": 3.124463102926054e-06,
      "loss": 0.81972325,
      "num_input_tokens_seen": 59226945,
      "step": 2750,
      "time_per_iteration": 2.5984702110290527
    },
    {
      "auxiliary_loss_clip": 0.0107862,
      "auxiliary_loss_mlp": 0.01002748,
      "balance_loss_clip": 1.02483416,
      "balance_loss_mlp": 1.00163329,
      "epoch": 0.33078819214813926,
      "flos": 70642609718400.0,
      "grad_norm": 0.7729465978551084,
      "language_loss": 0.6163817,
      "learning_rate": 3.1238188237913984e-06,
      "loss": 0.63719541,
      "num_input_tokens_seen": 59291485,
      "step": 2751,
      "time_per_iteration": 3.1397552490234375
    },
    {
      "auxiliary_loss_clip": 0.01199678,
      "auxiliary_loss_mlp": 0.01034006,
      "balance_loss_clip": 1.05940342,
      "balance_loss_mlp": 1.02455831,
      "epoch": 0.3309084350387783,
      "flos": 21141940907520.0,
      "grad_norm": 2.5169710047634193,
      "language_loss": 0.76210451,
      "learning_rate": 3.1231743741743202e-06,
      "loss": 0.78444135,
      "num_input_tokens_seen": 59310990,
      "step": 2752,
      "time_per_iteration": 2.4696764945983887
    },
    {
      "auxiliary_loss_clip": 0.01172715,
      "auxiliary_loss_mlp": 0.01033895,
      "balance_loss_clip": 1.05092049,
      "balance_loss_mlp": 1.02516294,
      "epoch": 0.3310286779294174,
      "flos": 14209350318720.0,
      "grad_norm": 2.4352076894350305,
      "language_loss": 0.83444971,
      "learning_rate": 3.122529754172582e-06,
      "loss": 0.85651577,
      "num_input_tokens_seen": 59327875,
      "step": 2753,
      "time_per_iteration": 2.48481822013855
    },
    {
      "auxiliary_loss_clip": 0.0118076,
      "auxiliary_loss_mlp": 0.01035805,
      "balance_loss_clip": 1.05689394,
      "balance_loss_mlp": 1.02652478,
      "epoch": 0.33114892082005654,
      "flos": 20778130005120.0,
      "grad_norm": 1.9102499010941272,
      "language_loss": 0.73010021,
      "learning_rate": 3.1218849638839736e-06,
      "loss": 0.75226587,
      "num_input_tokens_seen": 59347135,
      "step": 2754,
      "time_per_iteration": 2.5567264556884766
    },
    {
      "auxiliary_loss_clip": 0.01137548,
      "auxiliary_loss_mlp": 0.01036108,
      "balance_loss_clip": 1.04534566,
      "balance_loss_mlp": 1.0262078,
      "epoch": 0.3312691637106956,
      "flos": 17090750499840.0,
      "grad_norm": 1.6547585653169383,
      "language_loss": 0.785092,
      "learning_rate": 3.121240003406307e-06,
      "loss": 0.80682856,
      "num_input_tokens_seen": 59365985,
      "step": 2755,
      "time_per_iteration": 2.542545795440674
    },
    {
      "auxiliary_loss_clip": 0.01153884,
      "auxiliary_loss_mlp": 0.01031601,
      "balance_loss_clip": 1.05309296,
      "balance_loss_mlp": 1.02196908,
      "epoch": 0.3313894066013347,
      "flos": 29456230008960.0,
      "grad_norm": 2.5690401948586534,
      "language_loss": 0.72341871,
      "learning_rate": 3.120594872837425e-06,
      "loss": 0.74527359,
      "num_input_tokens_seen": 59384655,
      "step": 2756,
      "time_per_iteration": 2.600107192993164
    },
    {
      "auxiliary_loss_clip": 0.01078267,
      "auxiliary_loss_mlp": 0.00754039,
      "balance_loss_clip": 1.02038193,
      "balance_loss_mlp": 1.00154066,
      "epoch": 0.3315096494919738,
      "flos": 61419242280960.0,
      "grad_norm": 0.8304241367334306,
      "language_loss": 0.62385881,
      "learning_rate": 3.1199495722751906e-06,
      "loss": 0.64218187,
      "num_input_tokens_seen": 59444185,
      "step": 2757,
      "time_per_iteration": 3.1084578037261963
    },
    {
      "auxiliary_loss_clip": 0.0113728,
      "auxiliary_loss_mlp": 0.01035406,
      "balance_loss_clip": 1.04860616,
      "balance_loss_mlp": 1.02584529,
      "epoch": 0.33162989238261287,
      "flos": 21653057485440.0,
      "grad_norm": 1.5867036401819143,
      "language_loss": 0.84134471,
      "learning_rate": 3.1193041018174972e-06,
      "loss": 0.86307156,
      "num_input_tokens_seen": 59464900,
      "step": 2758,
      "time_per_iteration": 2.5904479026794434
    },
    {
      "auxiliary_loss_clip": 0.01182796,
      "auxiliary_loss_mlp": 0.01028725,
      "balance_loss_clip": 1.05649698,
      "balance_loss_mlp": 1.01961744,
      "epoch": 0.331750135273252,
      "flos": 22674787850880.0,
      "grad_norm": 2.518464440665844,
      "language_loss": 0.94788033,
      "learning_rate": 3.118658461562261e-06,
      "loss": 0.96999556,
      "num_input_tokens_seen": 59481000,
      "step": 2759,
      "time_per_iteration": 2.5496444702148438
    },
    {
      "auxiliary_loss_clip": 0.0116562,
      "auxiliary_loss_mlp": 0.01035404,
      "balance_loss_clip": 1.05583405,
      "balance_loss_mlp": 1.02552152,
      "epoch": 0.33187037816389103,
      "flos": 22746896403840.0,
      "grad_norm": 1.6533177678466706,
      "language_loss": 0.8507998,
      "learning_rate": 3.118012651607426e-06,
      "loss": 0.87281001,
      "num_input_tokens_seen": 59502605,
      "step": 2760,
      "time_per_iteration": 2.5263097286224365
    },
    {
      "auxiliary_loss_clip": 0.01193611,
      "auxiliary_loss_mlp": 0.01037843,
      "balance_loss_clip": 1.05622172,
      "balance_loss_mlp": 1.02765107,
      "epoch": 0.33199062105453014,
      "flos": 19203769918080.0,
      "grad_norm": 2.4254441292340485,
      "language_loss": 0.83478636,
      "learning_rate": 3.1173666720509603e-06,
      "loss": 0.85710084,
      "num_input_tokens_seen": 59519540,
      "step": 2761,
      "time_per_iteration": 2.4219202995300293
    },
    {
      "auxiliary_loss_clip": 0.01168581,
      "auxiliary_loss_mlp": 0.01030734,
      "balance_loss_clip": 1.05312145,
      "balance_loss_mlp": 1.02172756,
      "epoch": 0.33211086394516925,
      "flos": 31577006764800.0,
      "grad_norm": 1.6923913410383216,
      "language_loss": 0.68148553,
      "learning_rate": 3.116720522990859e-06,
      "loss": 0.70347857,
      "num_input_tokens_seen": 59540415,
      "step": 2762,
      "time_per_iteration": 2.5745129585266113
    },
    {
      "auxiliary_loss_clip": 0.01125152,
      "auxiliary_loss_mlp": 0.01033436,
      "balance_loss_clip": 1.05001163,
      "balance_loss_mlp": 1.02435267,
      "epoch": 0.3322311068358083,
      "flos": 17932496791680.0,
      "grad_norm": 1.9306088334356015,
      "language_loss": 0.61995107,
      "learning_rate": 3.116074204525142e-06,
      "loss": 0.64153695,
      "num_input_tokens_seen": 59558590,
      "step": 2763,
      "time_per_iteration": 3.3974854946136475
    },
    {
      "auxiliary_loss_clip": 0.0117086,
      "auxiliary_loss_mlp": 0.01029229,
      "balance_loss_clip": 1.05436826,
      "balance_loss_mlp": 1.02089024,
      "epoch": 0.3323513497264474,
      "flos": 32269831269120.0,
      "grad_norm": 1.4331940432253774,
      "language_loss": 0.83622736,
      "learning_rate": 3.1154277167518553e-06,
      "loss": 0.85822821,
      "num_input_tokens_seen": 59580205,
      "step": 2764,
      "time_per_iteration": 2.5662119388580322
    },
    {
      "auxiliary_loss_clip": 0.01062631,
      "auxiliary_loss_mlp": 0.01001813,
      "balance_loss_clip": 1.01838255,
      "balance_loss_mlp": 1.00066233,
      "epoch": 0.33247159261708653,
      "flos": 52668674588160.0,
      "grad_norm": 0.7804633477763853,
      "language_loss": 0.59487081,
      "learning_rate": 3.114781059769072e-06,
      "loss": 0.61551523,
      "num_input_tokens_seen": 59631530,
      "step": 2765,
      "time_per_iteration": 2.976964235305786
    },
    {
      "auxiliary_loss_clip": 0.01162952,
      "auxiliary_loss_mlp": 0.01030335,
      "balance_loss_clip": 1.05310428,
      "balance_loss_mlp": 1.02079248,
      "epoch": 0.3325918355077256,
      "flos": 27125232906240.0,
      "grad_norm": 2.743449929994885,
      "language_loss": 0.67059427,
      "learning_rate": 3.1141342336748874e-06,
      "loss": 0.69252712,
      "num_input_tokens_seen": 59651090,
      "step": 2766,
      "time_per_iteration": 2.6151795387268066
    },
    {
      "auxiliary_loss_clip": 0.01177667,
      "auxiliary_loss_mlp": 0.01034036,
      "balance_loss_clip": 1.05664277,
      "balance_loss_mlp": 1.02500534,
      "epoch": 0.3327120783983647,
      "flos": 23664414435840.0,
      "grad_norm": 1.4137262903912304,
      "language_loss": 0.82222188,
      "learning_rate": 3.1134872385674253e-06,
      "loss": 0.84433889,
      "num_input_tokens_seen": 59675245,
      "step": 2767,
      "time_per_iteration": 2.5766637325286865
    },
    {
      "auxiliary_loss_clip": 0.01167436,
      "auxiliary_loss_mlp": 0.01034197,
      "balance_loss_clip": 1.05121827,
      "balance_loss_mlp": 1.02471995,
      "epoch": 0.3328323212890038,
      "flos": 19171378828800.0,
      "grad_norm": 1.6515907177798275,
      "language_loss": 0.85343361,
      "learning_rate": 3.1128400745448353e-06,
      "loss": 0.8754499,
      "num_input_tokens_seen": 59694625,
      "step": 2768,
      "time_per_iteration": 3.3697237968444824
    },
    {
      "auxiliary_loss_clip": 0.01180835,
      "auxiliary_loss_mlp": 0.01032246,
      "balance_loss_clip": 1.05591393,
      "balance_loss_mlp": 1.02310896,
      "epoch": 0.33295256417964286,
      "flos": 37706347463040.0,
      "grad_norm": 2.169925744566654,
      "language_loss": 0.62822759,
      "learning_rate": 3.11219274170529e-06,
      "loss": 0.65035838,
      "num_input_tokens_seen": 59716435,
      "step": 2769,
      "time_per_iteration": 3.335397720336914
    },
    {
      "auxiliary_loss_clip": 0.01157557,
      "auxiliary_loss_mlp": 0.01034031,
      "balance_loss_clip": 1.05101037,
      "balance_loss_mlp": 1.02562094,
      "epoch": 0.333072807070282,
      "flos": 26505989412480.0,
      "grad_norm": 14.509614954692156,
      "language_loss": 0.81849802,
      "learning_rate": 3.1115452401469903e-06,
      "loss": 0.84041399,
      "num_input_tokens_seen": 59736835,
      "step": 2770,
      "time_per_iteration": 2.556903123855591
    },
    {
      "auxiliary_loss_clip": 0.01125532,
      "auxiliary_loss_mlp": 0.01033786,
      "balance_loss_clip": 1.04597116,
      "balance_loss_mlp": 1.02481604,
      "epoch": 0.3331930499609211,
      "flos": 21430913823360.0,
      "grad_norm": 2.1390266964790836,
      "language_loss": 0.86334103,
      "learning_rate": 3.1108975699681613e-06,
      "loss": 0.88493425,
      "num_input_tokens_seen": 59754230,
      "step": 2771,
      "time_per_iteration": 2.5740745067596436
    },
    {
      "auxiliary_loss_clip": 0.01147929,
      "auxiliary_loss_mlp": 0.01036992,
      "balance_loss_clip": 1.05150056,
      "balance_loss_mlp": 1.02792573,
      "epoch": 0.33331329285156014,
      "flos": 20659947281280.0,
      "grad_norm": 1.874550957165198,
      "language_loss": 0.71346569,
      "learning_rate": 3.1102497312670542e-06,
      "loss": 0.73531485,
      "num_input_tokens_seen": 59772235,
      "step": 2772,
      "time_per_iteration": 2.539773464202881
    },
    {
      "auxiliary_loss_clip": 0.01153109,
      "auxiliary_loss_mlp": 0.01038073,
      "balance_loss_clip": 1.05132282,
      "balance_loss_mlp": 1.02885818,
      "epoch": 0.33343353574219925,
      "flos": 28001596930560.0,
      "grad_norm": 2.8438995920134076,
      "language_loss": 0.8049711,
      "learning_rate": 3.109601724141946e-06,
      "loss": 0.8268829,
      "num_input_tokens_seen": 59791230,
      "step": 2773,
      "time_per_iteration": 2.547266721725464
    },
    {
      "auxiliary_loss_clip": 0.01161656,
      "auxiliary_loss_mlp": 0.01029194,
      "balance_loss_clip": 1.05396748,
      "balance_loss_mlp": 1.02053893,
      "epoch": 0.33355377863283836,
      "flos": 23764963582080.0,
      "grad_norm": 1.7184250157778536,
      "language_loss": 0.68006015,
      "learning_rate": 3.108953548691138e-06,
      "loss": 0.70196861,
      "num_input_tokens_seen": 59811315,
      "step": 2774,
      "time_per_iteration": 2.53174090385437
    },
    {
      "auxiliary_loss_clip": 0.01195884,
      "auxiliary_loss_mlp": 0.01029654,
      "balance_loss_clip": 1.05773067,
      "balance_loss_mlp": 1.02047515,
      "epoch": 0.3336740215234774,
      "flos": 37779677078400.0,
      "grad_norm": 2.1293331143453287,
      "language_loss": 0.72533524,
      "learning_rate": 3.108305205012959e-06,
      "loss": 0.74759054,
      "num_input_tokens_seen": 59832010,
      "step": 2775,
      "time_per_iteration": 2.579028844833374
    },
    {
      "auxiliary_loss_clip": 0.01161332,
      "auxiliary_loss_mlp": 0.01029798,
      "balance_loss_clip": 1.05292702,
      "balance_loss_mlp": 1.02032089,
      "epoch": 0.3337942644141165,
      "flos": 25519056347520.0,
      "grad_norm": 2.3667259888482577,
      "language_loss": 0.87442344,
      "learning_rate": 3.107656693205761e-06,
      "loss": 0.89633477,
      "num_input_tokens_seen": 59851450,
      "step": 2776,
      "time_per_iteration": 2.5330100059509277
    },
    {
      "auxiliary_loss_clip": 0.01197016,
      "auxiliary_loss_mlp": 0.01035707,
      "balance_loss_clip": 1.05699849,
      "balance_loss_mlp": 1.0258956,
      "epoch": 0.3339145073047556,
      "flos": 25989844930560.0,
      "grad_norm": 2.9084244473816225,
      "language_loss": 0.70411718,
      "learning_rate": 3.107008013367924e-06,
      "loss": 0.72644442,
      "num_input_tokens_seen": 59870245,
      "step": 2777,
      "time_per_iteration": 2.4657881259918213
    },
    {
      "auxiliary_loss_clip": 0.01148842,
      "auxiliary_loss_mlp": 0.01031527,
      "balance_loss_clip": 1.05172324,
      "balance_loss_mlp": 1.02202642,
      "epoch": 0.3340347501953947,
      "flos": 19062569554560.0,
      "grad_norm": 2.0964109295983344,
      "language_loss": 0.86988473,
      "learning_rate": 3.1063591655978507e-06,
      "loss": 0.89168835,
      "num_input_tokens_seen": 59886195,
      "step": 2778,
      "time_per_iteration": 2.5176336765289307
    },
    {
      "auxiliary_loss_clip": 0.01122109,
      "auxiliary_loss_mlp": 0.0103238,
      "balance_loss_clip": 1.0455277,
      "balance_loss_mlp": 1.02304578,
      "epoch": 0.3341549930860338,
      "flos": 18109715518080.0,
      "grad_norm": 1.728711784970176,
      "language_loss": 0.79506189,
      "learning_rate": 3.105710149993972e-06,
      "loss": 0.81660682,
      "num_input_tokens_seen": 59905525,
      "step": 2779,
      "time_per_iteration": 2.5531671047210693
    },
    {
      "auxiliary_loss_clip": 0.01196644,
      "auxiliary_loss_mlp": 0.01030354,
      "balance_loss_clip": 1.0569284,
      "balance_loss_mlp": 1.02102017,
      "epoch": 0.33427523597667286,
      "flos": 22674967418880.0,
      "grad_norm": 1.7292401895339276,
      "language_loss": 0.85241753,
      "learning_rate": 3.1050609666547427e-06,
      "loss": 0.87468749,
      "num_input_tokens_seen": 59925085,
      "step": 2780,
      "time_per_iteration": 2.4527883529663086
    },
    {
      "auxiliary_loss_clip": 0.01158706,
      "auxiliary_loss_mlp": 0.01039448,
      "balance_loss_clip": 1.05448353,
      "balance_loss_mlp": 1.03050768,
      "epoch": 0.33439547886731197,
      "flos": 22638338524800.0,
      "grad_norm": 1.8283661288281738,
      "language_loss": 0.77297407,
      "learning_rate": 3.104411615678644e-06,
      "loss": 0.79495561,
      "num_input_tokens_seen": 59943935,
      "step": 2781,
      "time_per_iteration": 2.5432119369506836
    },
    {
      "auxiliary_loss_clip": 0.0116099,
      "auxiliary_loss_mlp": 0.0103476,
      "balance_loss_clip": 1.05300725,
      "balance_loss_mlp": 1.02479994,
      "epoch": 0.3345157217579511,
      "flos": 24096383395200.0,
      "grad_norm": 4.294028980692174,
      "language_loss": 0.73754764,
      "learning_rate": 3.1037620971641803e-06,
      "loss": 0.75950515,
      "num_input_tokens_seen": 59963725,
      "step": 2782,
      "time_per_iteration": 2.5091447830200195
    },
    {
      "auxiliary_loss_clip": 0.0119616,
      "auxiliary_loss_mlp": 0.01036294,
      "balance_loss_clip": 1.05814433,
      "balance_loss_mlp": 1.02685869,
      "epoch": 0.33463596464859013,
      "flos": 18989491334400.0,
      "grad_norm": 2.1889981021101548,
      "language_loss": 0.64687735,
      "learning_rate": 3.1031124112098844e-06,
      "loss": 0.66920185,
      "num_input_tokens_seen": 59981935,
      "step": 2783,
      "time_per_iteration": 2.4288344383239746
    },
    {
      "auxiliary_loss_clip": 0.0116814,
      "auxiliary_loss_mlp": 0.01026885,
      "balance_loss_clip": 1.05526125,
      "balance_loss_mlp": 1.01790881,
      "epoch": 0.33475620753922924,
      "flos": 20375607219840.0,
      "grad_norm": 2.059659066138216,
      "language_loss": 0.72374785,
      "learning_rate": 3.1024625579143127e-06,
      "loss": 0.74569809,
      "num_input_tokens_seen": 59999455,
      "step": 2784,
      "time_per_iteration": 2.489598035812378
    },
    {
      "auxiliary_loss_clip": 0.01193086,
      "auxiliary_loss_mlp": 0.0103534,
      "balance_loss_clip": 1.05673635,
      "balance_loss_mlp": 1.02617896,
      "epoch": 0.33487645042986836,
      "flos": 18182578256640.0,
      "grad_norm": 1.808290904659665,
      "language_loss": 0.73041236,
      "learning_rate": 3.101812537376048e-06,
      "loss": 0.75269663,
      "num_input_tokens_seen": 60018475,
      "step": 2785,
      "time_per_iteration": 2.443201780319214
    },
    {
      "auxiliary_loss_clip": 0.01158049,
      "auxiliary_loss_mlp": 0.00763856,
      "balance_loss_clip": 1.05207539,
      "balance_loss_mlp": 1.00059342,
      "epoch": 0.3349966933205074,
      "flos": 25848824135040.0,
      "grad_norm": 1.9729312251774827,
      "language_loss": 0.84468871,
      "learning_rate": 3.1011623496936973e-06,
      "loss": 0.86390775,
      "num_input_tokens_seen": 60036770,
      "step": 2786,
      "time_per_iteration": 2.5408260822296143
    },
    {
      "auxiliary_loss_clip": 0.01193375,
      "auxiliary_loss_mlp": 0.01031668,
      "balance_loss_clip": 1.05780411,
      "balance_loss_mlp": 1.02285278,
      "epoch": 0.3351169362111465,
      "flos": 28111447699200.0,
      "grad_norm": 1.7474483151917581,
      "language_loss": 0.6976797,
      "learning_rate": 3.100511994965893e-06,
      "loss": 0.71993011,
      "num_input_tokens_seen": 60056725,
      "step": 2787,
      "time_per_iteration": 2.5296106338500977
    },
    {
      "auxiliary_loss_clip": 0.01174502,
      "auxiliary_loss_mlp": 0.01032182,
      "balance_loss_clip": 1.05610752,
      "balance_loss_mlp": 1.02318156,
      "epoch": 0.33523717910178563,
      "flos": 22673315393280.0,
      "grad_norm": 1.7022789766044617,
      "language_loss": 0.84408545,
      "learning_rate": 3.0998614732912947e-06,
      "loss": 0.86615229,
      "num_input_tokens_seen": 60076100,
      "step": 2788,
      "time_per_iteration": 2.4613378047943115
    },
    {
      "auxiliary_loss_clip": 0.01181942,
      "auxiliary_loss_mlp": 0.01034897,
      "balance_loss_clip": 1.05838513,
      "balance_loss_mlp": 1.02564025,
      "epoch": 0.3353574219924247,
      "flos": 15669801400320.0,
      "grad_norm": 2.4432268464487645,
      "language_loss": 0.67808342,
      "learning_rate": 3.0992107847685855e-06,
      "loss": 0.70025182,
      "num_input_tokens_seen": 60093815,
      "step": 2789,
      "time_per_iteration": 3.2646031379699707
    },
    {
      "auxiliary_loss_clip": 0.01171927,
      "auxiliary_loss_mlp": 0.01040012,
      "balance_loss_clip": 1.05956745,
      "balance_loss_mlp": 1.0304513,
      "epoch": 0.3354776648830638,
      "flos": 24790644443520.0,
      "grad_norm": 1.5583806863305008,
      "language_loss": 0.79321051,
      "learning_rate": 3.0985599294964736e-06,
      "loss": 0.81532985,
      "num_input_tokens_seen": 60113370,
      "step": 2790,
      "time_per_iteration": 2.546794891357422
    },
    {
      "auxiliary_loss_clip": 0.01162839,
      "auxiliary_loss_mlp": 0.01039022,
      "balance_loss_clip": 1.05497122,
      "balance_loss_mlp": 1.02819812,
      "epoch": 0.33559790777370285,
      "flos": 28694852398080.0,
      "grad_norm": 2.081542223193698,
      "language_loss": 0.70130706,
      "learning_rate": 3.097908907573695e-06,
      "loss": 0.72332573,
      "num_input_tokens_seen": 60131350,
      "step": 2791,
      "time_per_iteration": 2.5969111919403076
    },
    {
      "auxiliary_loss_clip": 0.01124099,
      "auxiliary_loss_mlp": 0.01037581,
      "balance_loss_clip": 1.05260944,
      "balance_loss_mlp": 1.02872729,
      "epoch": 0.33571815066434196,
      "flos": 22235779825920.0,
      "grad_norm": 1.9882523961761898,
      "language_loss": 0.89591956,
      "learning_rate": 3.0972577190990067e-06,
      "loss": 0.91753638,
      "num_input_tokens_seen": 60149830,
      "step": 2792,
      "time_per_iteration": 2.596595525741577
    },
    {
      "auxiliary_loss_clip": 0.01156373,
      "auxiliary_loss_mlp": 0.01033906,
      "balance_loss_clip": 1.0543313,
      "balance_loss_mlp": 1.02497685,
      "epoch": 0.3358383935549811,
      "flos": 23842279607040.0,
      "grad_norm": 1.7828988341915757,
      "language_loss": 0.80069876,
      "learning_rate": 3.096606364171196e-06,
      "loss": 0.82260156,
      "num_input_tokens_seen": 60169620,
      "step": 2793,
      "time_per_iteration": 2.5602777004241943
    },
    {
      "auxiliary_loss_clip": 0.01135225,
      "auxiliary_loss_mlp": 0.01033361,
      "balance_loss_clip": 1.04930151,
      "balance_loss_mlp": 1.02370477,
      "epoch": 0.33595863644562013,
      "flos": 22267308988800.0,
      "grad_norm": 2.3224747717232224,
      "language_loss": 0.85101867,
      "learning_rate": 3.0959548428890703e-06,
      "loss": 0.87270457,
      "num_input_tokens_seen": 60188490,
      "step": 2794,
      "time_per_iteration": 2.544663429260254
    },
    {
      "auxiliary_loss_clip": 0.01177984,
      "auxiliary_loss_mlp": 0.01039454,
      "balance_loss_clip": 1.05831838,
      "balance_loss_mlp": 1.03011346,
      "epoch": 0.33607887933625924,
      "flos": 20119779578880.0,
      "grad_norm": 1.5196336030087647,
      "language_loss": 0.84115517,
      "learning_rate": 3.095303155351468e-06,
      "loss": 0.86332953,
      "num_input_tokens_seen": 60208695,
      "step": 2795,
      "time_per_iteration": 3.3538596630096436
    },
    {
      "auxiliary_loss_clip": 0.01132007,
      "auxiliary_loss_mlp": 0.01035944,
      "balance_loss_clip": 1.05377054,
      "balance_loss_mlp": 1.02703881,
      "epoch": 0.33619912222689835,
      "flos": 19318109886720.0,
      "grad_norm": 2.175559617833444,
      "language_loss": 0.79453677,
      "learning_rate": 3.0946513016572464e-06,
      "loss": 0.81621623,
      "num_input_tokens_seen": 60227600,
      "step": 2796,
      "time_per_iteration": 3.410482883453369
    },
    {
      "auxiliary_loss_clip": 0.01184565,
      "auxiliary_loss_mlp": 0.01033796,
      "balance_loss_clip": 1.05628538,
      "balance_loss_mlp": 1.02425289,
      "epoch": 0.3363193651175374,
      "flos": 16800664262400.0,
      "grad_norm": 3.020137812732992,
      "language_loss": 0.76981485,
      "learning_rate": 3.0939992819052938e-06,
      "loss": 0.79199839,
      "num_input_tokens_seen": 60245110,
      "step": 2797,
      "time_per_iteration": 2.4818058013916016
    },
    {
      "auxiliary_loss_clip": 0.01169443,
      "auxiliary_loss_mlp": 0.01030983,
      "balance_loss_clip": 1.0585146,
      "balance_loss_mlp": 1.02157736,
      "epoch": 0.3364396080081765,
      "flos": 23550289948800.0,
      "grad_norm": 2.0811658074619106,
      "language_loss": 0.81283718,
      "learning_rate": 3.0933470961945193e-06,
      "loss": 0.83484143,
      "num_input_tokens_seen": 60263405,
      "step": 2798,
      "time_per_iteration": 2.548593521118164
    },
    {
      "auxiliary_loss_clip": 0.01164715,
      "auxiliary_loss_mlp": 0.01036986,
      "balance_loss_clip": 1.05844295,
      "balance_loss_mlp": 1.02821207,
      "epoch": 0.3365598508988156,
      "flos": 28037902602240.0,
      "grad_norm": 1.5933206515666067,
      "language_loss": 0.68220854,
      "learning_rate": 3.0926947446238597e-06,
      "loss": 0.70422554,
      "num_input_tokens_seen": 60282975,
      "step": 2799,
      "time_per_iteration": 2.603602886199951
    },
    {
      "auxiliary_loss_clip": 0.01183793,
      "auxiliary_loss_mlp": 0.01035064,
      "balance_loss_clip": 1.05471015,
      "balance_loss_mlp": 1.0252049,
      "epoch": 0.3366800937894547,
      "flos": 16982767238400.0,
      "grad_norm": 2.39436592149514,
      "language_loss": 0.82461476,
      "learning_rate": 3.092042227292276e-06,
      "loss": 0.84680331,
      "num_input_tokens_seen": 60299810,
      "step": 2800,
      "time_per_iteration": 2.5505878925323486
    },
    {
      "auxiliary_loss_clip": 0.01191723,
      "auxiliary_loss_mlp": 0.01030951,
      "balance_loss_clip": 1.05854976,
      "balance_loss_mlp": 1.02246976,
      "epoch": 0.3368003366800938,
      "flos": 23915321913600.0,
      "grad_norm": 1.5730277935919512,
      "language_loss": 0.88006878,
      "learning_rate": 3.0913895442987557e-06,
      "loss": 0.90229547,
      "num_input_tokens_seen": 60320775,
      "step": 2801,
      "time_per_iteration": 2.4913220405578613
    },
    {
      "auxiliary_loss_clip": 0.01153269,
      "auxiliary_loss_mlp": 0.00764059,
      "balance_loss_clip": 1.05524945,
      "balance_loss_mlp": 1.00052059,
      "epoch": 0.3369205795707329,
      "flos": 24791219061120.0,
      "grad_norm": 1.5405413196578854,
      "language_loss": 0.86292446,
      "learning_rate": 3.090736695742308e-06,
      "loss": 0.88209772,
      "num_input_tokens_seen": 60341905,
      "step": 2802,
      "time_per_iteration": 2.5626161098480225
    },
    {
      "auxiliary_loss_clip": 0.01131308,
      "auxiliary_loss_mlp": 0.01032308,
      "balance_loss_clip": 1.05089545,
      "balance_loss_mlp": 1.02376688,
      "epoch": 0.33704082246137196,
      "flos": 17931096161280.0,
      "grad_norm": 2.2596717581977095,
      "language_loss": 0.52499092,
      "learning_rate": 3.0900836817219713e-06,
      "loss": 0.5466271,
      "num_input_tokens_seen": 60358335,
      "step": 2803,
      "time_per_iteration": 2.552642583847046
    },
    {
      "auxiliary_loss_clip": 0.01192919,
      "auxiliary_loss_mlp": 0.01032191,
      "balance_loss_clip": 1.05703592,
      "balance_loss_mlp": 1.02365518,
      "epoch": 0.33716106535201107,
      "flos": 21286517149440.0,
      "grad_norm": 1.6339236632469183,
      "language_loss": 0.83554125,
      "learning_rate": 3.089430502336807e-06,
      "loss": 0.85779238,
      "num_input_tokens_seen": 60378305,
      "step": 2804,
      "time_per_iteration": 2.4783382415771484
    },
    {
      "auxiliary_loss_clip": 0.01183403,
      "auxiliary_loss_mlp": 0.01029714,
      "balance_loss_clip": 1.05715764,
      "balance_loss_mlp": 1.02036202,
      "epoch": 0.3372813082426502,
      "flos": 18402962152320.0,
      "grad_norm": 3.299616317513142,
      "language_loss": 0.89962888,
      "learning_rate": 3.088777157685902e-06,
      "loss": 0.92176002,
      "num_input_tokens_seen": 60393895,
      "step": 2805,
      "time_per_iteration": 2.476597547531128
    },
    {
      "auxiliary_loss_clip": 0.01162708,
      "auxiliary_loss_mlp": 0.01027235,
      "balance_loss_clip": 1.05594015,
      "balance_loss_mlp": 1.01829433,
      "epoch": 0.33740155113328923,
      "flos": 17201391367680.0,
      "grad_norm": 1.8817690805679947,
      "language_loss": 0.85918957,
      "learning_rate": 3.088123647868367e-06,
      "loss": 0.88108909,
      "num_input_tokens_seen": 60410445,
      "step": 2806,
      "time_per_iteration": 2.489069700241089
    },
    {
      "auxiliary_loss_clip": 0.01184795,
      "auxiliary_loss_mlp": 0.01033842,
      "balance_loss_clip": 1.05695033,
      "balance_loss_mlp": 1.02505589,
      "epoch": 0.33752179402392835,
      "flos": 29058950609280.0,
      "grad_norm": 1.841258936596911,
      "language_loss": 0.81294703,
      "learning_rate": 3.0874699729833405e-06,
      "loss": 0.83513343,
      "num_input_tokens_seen": 60431815,
      "step": 2807,
      "time_per_iteration": 2.571648120880127
    },
    {
      "auxiliary_loss_clip": 0.01162052,
      "auxiliary_loss_mlp": 0.01027991,
      "balance_loss_clip": 1.05567408,
      "balance_loss_mlp": 1.01874042,
      "epoch": 0.3376420369145674,
      "flos": 25080730680960.0,
      "grad_norm": 1.5942560599041764,
      "language_loss": 0.79916126,
      "learning_rate": 3.086816133129983e-06,
      "loss": 0.82106173,
      "num_input_tokens_seen": 60452075,
      "step": 2808,
      "time_per_iteration": 2.542128801345825
    },
    {
      "auxiliary_loss_clip": 0.01195746,
      "auxiliary_loss_mlp": 0.01028453,
      "balance_loss_clip": 1.06124699,
      "balance_loss_mlp": 1.01963139,
      "epoch": 0.3377622798052065,
      "flos": 27490624007040.0,
      "grad_norm": 1.6209068587423892,
      "language_loss": 0.76131868,
      "learning_rate": 3.0861621284074826e-06,
      "loss": 0.78356069,
      "num_input_tokens_seen": 60472600,
      "step": 2809,
      "time_per_iteration": 2.5211522579193115
    },
    {
      "auxiliary_loss_clip": 0.01173977,
      "auxiliary_loss_mlp": 0.010321,
      "balance_loss_clip": 1.05811119,
      "balance_loss_mlp": 1.02384448,
      "epoch": 0.3378825226958456,
      "flos": 21975211589760.0,
      "grad_norm": 1.4633976848118593,
      "language_loss": 0.73133975,
      "learning_rate": 3.085507958915051e-06,
      "loss": 0.75340056,
      "num_input_tokens_seen": 60491030,
      "step": 2810,
      "time_per_iteration": 2.4954710006713867
    },
    {
      "auxiliary_loss_clip": 0.0116423,
      "auxiliary_loss_mlp": 0.01032624,
      "balance_loss_clip": 1.05806351,
      "balance_loss_mlp": 1.02296162,
      "epoch": 0.3380027655864847,
      "flos": 42523189200000.0,
      "grad_norm": 1.9593642710409174,
      "language_loss": 0.71038061,
      "learning_rate": 3.084853624751925e-06,
      "loss": 0.73234916,
      "num_input_tokens_seen": 60512615,
      "step": 2811,
      "time_per_iteration": 2.677703857421875
    },
    {
      "auxiliary_loss_clip": 0.01156172,
      "auxiliary_loss_mlp": 0.01034031,
      "balance_loss_clip": 1.05782747,
      "balance_loss_mlp": 1.02499461,
      "epoch": 0.3381230084771238,
      "flos": 26725080418560.0,
      "grad_norm": 1.6767827979958827,
      "language_loss": 0.85907954,
      "learning_rate": 3.0841991260173668e-06,
      "loss": 0.88098156,
      "num_input_tokens_seen": 60532520,
      "step": 2812,
      "time_per_iteration": 2.593587875366211
    },
    {
      "auxiliary_loss_clip": 0.01198465,
      "auxiliary_loss_mlp": 0.01029983,
      "balance_loss_clip": 1.0609982,
      "balance_loss_mlp": 1.02039814,
      "epoch": 0.3382432513677629,
      "flos": 22710375250560.0,
      "grad_norm": 1.826504349550865,
      "language_loss": 0.80315924,
      "learning_rate": 3.0835444628106634e-06,
      "loss": 0.82544374,
      "num_input_tokens_seen": 60551500,
      "step": 2813,
      "time_per_iteration": 2.4562697410583496
    },
    {
      "auxiliary_loss_clip": 0.01193363,
      "auxiliary_loss_mlp": 0.00764117,
      "balance_loss_clip": 1.058496,
      "balance_loss_mlp": 1.00055718,
      "epoch": 0.33836349425840195,
      "flos": 22122409524480.0,
      "grad_norm": 1.7304601418190944,
      "language_loss": 0.8320021,
      "learning_rate": 3.082889635231126e-06,
      "loss": 0.85157686,
      "num_input_tokens_seen": 60570160,
      "step": 2814,
      "time_per_iteration": 2.4557278156280518
    },
    {
      "auxiliary_loss_clip": 0.01168663,
      "auxiliary_loss_mlp": 0.01028183,
      "balance_loss_clip": 1.05481958,
      "balance_loss_mlp": 1.01848519,
      "epoch": 0.33848373714904106,
      "flos": 27308090067840.0,
      "grad_norm": 2.672041583900298,
      "language_loss": 0.76411843,
      "learning_rate": 3.0822346433780925e-06,
      "loss": 0.78608686,
      "num_input_tokens_seen": 60590885,
      "step": 2815,
      "time_per_iteration": 2.65100359916687
    },
    {
      "auxiliary_loss_clip": 0.01179063,
      "auxiliary_loss_mlp": 0.01028793,
      "balance_loss_clip": 1.05459321,
      "balance_loss_mlp": 1.01936984,
      "epoch": 0.3386039800396802,
      "flos": 25848716394240.0,
      "grad_norm": 1.9189750251488258,
      "language_loss": 0.87047362,
      "learning_rate": 3.0815794873509237e-06,
      "loss": 0.89255226,
      "num_input_tokens_seen": 60609170,
      "step": 2816,
      "time_per_iteration": 3.373980760574341
    },
    {
      "auxiliary_loss_clip": 0.01196763,
      "auxiliary_loss_mlp": 0.01030862,
      "balance_loss_clip": 1.06155539,
      "balance_loss_mlp": 1.02186131,
      "epoch": 0.33872422293031923,
      "flos": 18880646146560.0,
      "grad_norm": 1.7859634282738086,
      "language_loss": 0.73045981,
      "learning_rate": 3.0809241672490066e-06,
      "loss": 0.75273609,
      "num_input_tokens_seen": 60627340,
      "step": 2817,
      "time_per_iteration": 2.4223473072052
    },
    {
      "auxiliary_loss_clip": 0.01170535,
      "auxiliary_loss_mlp": 0.01028023,
      "balance_loss_clip": 1.05873156,
      "balance_loss_mlp": 1.01927888,
      "epoch": 0.33884446582095834,
      "flos": 23146977064320.0,
      "grad_norm": 1.6399070242588367,
      "language_loss": 0.84991193,
      "learning_rate": 3.080268683171753e-06,
      "loss": 0.87189746,
      "num_input_tokens_seen": 60647630,
      "step": 2818,
      "time_per_iteration": 2.541529893875122
    },
    {
      "auxiliary_loss_clip": 0.01178853,
      "auxiliary_loss_mlp": 0.01028763,
      "balance_loss_clip": 1.05573595,
      "balance_loss_mlp": 1.02034044,
      "epoch": 0.33896470871159745,
      "flos": 15997342544640.0,
      "grad_norm": 3.465952927681238,
      "language_loss": 0.89056385,
      "learning_rate": 3.0796130352185985e-06,
      "loss": 0.91263998,
      "num_input_tokens_seen": 60664485,
      "step": 2819,
      "time_per_iteration": 2.4549951553344727
    },
    {
      "auxiliary_loss_clip": 0.0115203,
      "auxiliary_loss_mlp": 0.00764486,
      "balance_loss_clip": 1.05051506,
      "balance_loss_mlp": 1.00056481,
      "epoch": 0.3390849516022365,
      "flos": 34495754112000.0,
      "grad_norm": 1.7206782321086551,
      "language_loss": 0.66293436,
      "learning_rate": 3.0789572234890057e-06,
      "loss": 0.68209958,
      "num_input_tokens_seen": 60686125,
      "step": 2820,
      "time_per_iteration": 2.6476056575775146
    },
    {
      "auxiliary_loss_clip": 0.01168583,
      "auxiliary_loss_mlp": 0.01031529,
      "balance_loss_clip": 1.05967259,
      "balance_loss_mlp": 1.02205217,
      "epoch": 0.3392051944928756,
      "flos": 16180307447040.0,
      "grad_norm": 1.5980720703876337,
      "language_loss": 0.77722597,
      "learning_rate": 3.0783012480824596e-06,
      "loss": 0.79922712,
      "num_input_tokens_seen": 60705270,
      "step": 2821,
      "time_per_iteration": 3.373331308364868
    },
    {
      "auxiliary_loss_clip": 0.01192974,
      "auxiliary_loss_mlp": 0.01035819,
      "balance_loss_clip": 1.05705953,
      "balance_loss_mlp": 1.02707541,
      "epoch": 0.33932543738351467,
      "flos": 17086656349440.0,
      "grad_norm": 2.05781483801845,
      "language_loss": 0.74176073,
      "learning_rate": 3.077645109098471e-06,
      "loss": 0.7640487,
      "num_input_tokens_seen": 60721540,
      "step": 2822,
      "time_per_iteration": 3.177468776702881
    },
    {
      "auxiliary_loss_clip": 0.01138654,
      "auxiliary_loss_mlp": 0.01031232,
      "balance_loss_clip": 1.05350184,
      "balance_loss_mlp": 1.02254176,
      "epoch": 0.3394456802741538,
      "flos": 22126970551680.0,
      "grad_norm": 1.8771677630903518,
      "language_loss": 0.72189283,
      "learning_rate": 3.076988806636577e-06,
      "loss": 0.74359167,
      "num_input_tokens_seen": 60739300,
      "step": 2823,
      "time_per_iteration": 3.2664270401000977
    },
    {
      "auxiliary_loss_clip": 0.01171885,
      "auxiliary_loss_mlp": 0.00764276,
      "balance_loss_clip": 1.05893779,
      "balance_loss_mlp": 1.00064969,
      "epoch": 0.3395659231647929,
      "flos": 25226887121280.0,
      "grad_norm": 1.8426482199465122,
      "language_loss": 0.88545227,
      "learning_rate": 3.0763323407963377e-06,
      "loss": 0.90481389,
      "num_input_tokens_seen": 60758910,
      "step": 2824,
      "time_per_iteration": 2.5418875217437744
    },
    {
      "auxiliary_loss_clip": 0.01177069,
      "auxiliary_loss_mlp": 0.01030703,
      "balance_loss_clip": 1.0530386,
      "balance_loss_mlp": 1.02266824,
      "epoch": 0.33968616605543195,
      "flos": 29096477343360.0,
      "grad_norm": 1.7419393378233285,
      "language_loss": 0.80139029,
      "learning_rate": 3.075675711677337e-06,
      "loss": 0.82346797,
      "num_input_tokens_seen": 60779005,
      "step": 2825,
      "time_per_iteration": 2.550868034362793
    },
    {
      "auxiliary_loss_clip": 0.01161744,
      "auxiliary_loss_mlp": 0.01038405,
      "balance_loss_clip": 1.05785692,
      "balance_loss_mlp": 1.02902889,
      "epoch": 0.33980640894607106,
      "flos": 21433966479360.0,
      "grad_norm": 1.875569005835158,
      "language_loss": 0.77708316,
      "learning_rate": 3.0750189193791865e-06,
      "loss": 0.7990846,
      "num_input_tokens_seen": 60798590,
      "step": 2826,
      "time_per_iteration": 2.492832660675049
    },
    {
      "auxiliary_loss_clip": 0.01177994,
      "auxiliary_loss_mlp": 0.01028095,
      "balance_loss_clip": 1.05595779,
      "balance_loss_mlp": 1.01919031,
      "epoch": 0.33992665183671017,
      "flos": 32490035596800.0,
      "grad_norm": 2.25449747897589,
      "language_loss": 0.70592326,
      "learning_rate": 3.0743619640015203e-06,
      "loss": 0.72798419,
      "num_input_tokens_seen": 60818840,
      "step": 2827,
      "time_per_iteration": 2.5631790161132812
    },
    {
      "auxiliary_loss_clip": 0.0116979,
      "auxiliary_loss_mlp": 0.01032412,
      "balance_loss_clip": 1.05369449,
      "balance_loss_mlp": 1.02279806,
      "epoch": 0.3400468947273492,
      "flos": 17055414495360.0,
      "grad_norm": 3.858596552100289,
      "language_loss": 0.92709273,
      "learning_rate": 3.073704845643999e-06,
      "loss": 0.9491148,
      "num_input_tokens_seen": 60835965,
      "step": 2828,
      "time_per_iteration": 2.5107100009918213
    },
    {
      "auxiliary_loss_clip": 0.01181236,
      "auxiliary_loss_mlp": 0.01039316,
      "balance_loss_clip": 1.05490315,
      "balance_loss_mlp": 1.02958906,
      "epoch": 0.34016713761798834,
      "flos": 16872988296960.0,
      "grad_norm": 2.7177534372300367,
      "language_loss": 0.77822864,
      "learning_rate": 3.0730475644063063e-06,
      "loss": 0.80043423,
      "num_input_tokens_seen": 60851065,
      "step": 2829,
      "time_per_iteration": 2.4641716480255127
    },
    {
      "auxiliary_loss_clip": 0.01156822,
      "auxiliary_loss_mlp": 0.00763451,
      "balance_loss_clip": 1.05148387,
      "balance_loss_mlp": 1.00060332,
      "epoch": 0.34028738050862745,
      "flos": 21907161273600.0,
      "grad_norm": 1.6332431121648796,
      "language_loss": 0.64891446,
      "learning_rate": 3.072390120388151e-06,
      "loss": 0.66811717,
      "num_input_tokens_seen": 60869390,
      "step": 2830,
      "time_per_iteration": 2.4921460151672363
    },
    {
      "auxiliary_loss_clip": 0.01181327,
      "auxiliary_loss_mlp": 0.01029944,
      "balance_loss_clip": 1.05773854,
      "balance_loss_mlp": 1.02050829,
      "epoch": 0.3404076233992665,
      "flos": 22746034477440.0,
      "grad_norm": 2.7823955540506313,
      "language_loss": 0.70972896,
      "learning_rate": 3.071732513689267e-06,
      "loss": 0.73184162,
      "num_input_tokens_seen": 60887925,
      "step": 2831,
      "time_per_iteration": 2.477276563644409
    },
    {
      "auxiliary_loss_clip": 0.01183593,
      "auxiliary_loss_mlp": 0.01034559,
      "balance_loss_clip": 1.06125474,
      "balance_loss_mlp": 1.02537358,
      "epoch": 0.3405278662899056,
      "flos": 17052361839360.0,
      "grad_norm": 2.119739785314952,
      "language_loss": 0.6717006,
      "learning_rate": 3.0710747444094134e-06,
      "loss": 0.69388211,
      "num_input_tokens_seen": 60905955,
      "step": 2832,
      "time_per_iteration": 2.439854621887207
    },
    {
      "auxiliary_loss_clip": 0.01168702,
      "auxiliary_loss_mlp": 0.01033059,
      "balance_loss_clip": 1.05697453,
      "balance_loss_mlp": 1.02399278,
      "epoch": 0.3406481091805447,
      "flos": 42813131783040.0,
      "grad_norm": 1.7877636726489856,
      "language_loss": 0.64813149,
      "learning_rate": 3.070416812648372e-06,
      "loss": 0.67014909,
      "num_input_tokens_seen": 60929405,
      "step": 2833,
      "time_per_iteration": 2.6894752979278564
    },
    {
      "auxiliary_loss_clip": 0.01147064,
      "auxiliary_loss_mlp": 0.01031857,
      "balance_loss_clip": 1.04871488,
      "balance_loss_mlp": 1.02310157,
      "epoch": 0.3407683520711838,
      "flos": 26761457917440.0,
      "grad_norm": 2.0562421399888353,
      "language_loss": 0.65065026,
      "learning_rate": 3.069758718505951e-06,
      "loss": 0.67243946,
      "num_input_tokens_seen": 60951145,
      "step": 2834,
      "time_per_iteration": 2.5993926525115967
    },
    {
      "auxiliary_loss_clip": 0.01194028,
      "auxiliary_loss_mlp": 0.0103552,
      "balance_loss_clip": 1.05948353,
      "balance_loss_mlp": 1.02622747,
      "epoch": 0.3408885949618229,
      "flos": 28767643309440.0,
      "grad_norm": 1.5872817200681593,
      "language_loss": 0.79973483,
      "learning_rate": 3.0691004620819836e-06,
      "loss": 0.82203031,
      "num_input_tokens_seen": 60971275,
      "step": 2835,
      "time_per_iteration": 2.5071046352386475
    },
    {
      "auxiliary_loss_clip": 0.01047716,
      "auxiliary_loss_mlp": 0.01001519,
      "balance_loss_clip": 1.02581668,
      "balance_loss_mlp": 1.00033307,
      "epoch": 0.341008837852462,
      "flos": 63576252881280.0,
      "grad_norm": 0.794941977036189,
      "language_loss": 0.60230374,
      "learning_rate": 3.0684420434763254e-06,
      "loss": 0.62279606,
      "num_input_tokens_seen": 61037460,
      "step": 2836,
      "time_per_iteration": 3.1616110801696777
    },
    {
      "auxiliary_loss_clip": 0.0114123,
      "auxiliary_loss_mlp": 0.01035426,
      "balance_loss_clip": 1.05393887,
      "balance_loss_mlp": 1.02721214,
      "epoch": 0.34112908074310105,
      "flos": 20812173120000.0,
      "grad_norm": 1.875328642869392,
      "language_loss": 0.77162451,
      "learning_rate": 3.06778346278886e-06,
      "loss": 0.79339111,
      "num_input_tokens_seen": 61056295,
      "step": 2837,
      "time_per_iteration": 2.584921360015869
    },
    {
      "auxiliary_loss_clip": 0.01196858,
      "auxiliary_loss_mlp": 0.01028809,
      "balance_loss_clip": 1.06109297,
      "balance_loss_mlp": 1.01985598,
      "epoch": 0.34124932363374016,
      "flos": 24976446520320.0,
      "grad_norm": 1.6778330485733912,
      "language_loss": 0.78711438,
      "learning_rate": 3.0671247201194906e-06,
      "loss": 0.80937111,
      "num_input_tokens_seen": 61078430,
      "step": 2838,
      "time_per_iteration": 2.5037589073181152
    },
    {
      "auxiliary_loss_clip": 0.01150946,
      "auxiliary_loss_mlp": 0.0103295,
      "balance_loss_clip": 1.05079067,
      "balance_loss_mlp": 1.02421236,
      "epoch": 0.3413695665243792,
      "flos": 28402970480640.0,
      "grad_norm": 1.7403362728454987,
      "language_loss": 0.75793797,
      "learning_rate": 3.066465815568151e-06,
      "loss": 0.77977693,
      "num_input_tokens_seen": 61099260,
      "step": 2839,
      "time_per_iteration": 2.672842264175415
    },
    {
      "auxiliary_loss_clip": 0.01179379,
      "auxiliary_loss_mlp": 0.01027021,
      "balance_loss_clip": 1.05462503,
      "balance_loss_mlp": 1.01860511,
      "epoch": 0.34148980941501833,
      "flos": 25302012416640.0,
      "grad_norm": 1.799863540739093,
      "language_loss": 0.69140643,
      "learning_rate": 3.0658067492347947e-06,
      "loss": 0.7134704,
      "num_input_tokens_seen": 61121900,
      "step": 2840,
      "time_per_iteration": 2.557950258255005
    },
    {
      "auxiliary_loss_clip": 0.01101928,
      "auxiliary_loss_mlp": 0.01031228,
      "balance_loss_clip": 1.04898286,
      "balance_loss_mlp": 1.0218823,
      "epoch": 0.34161005230565744,
      "flos": 17530081747200.0,
      "grad_norm": 3.3886022029106737,
      "language_loss": 0.66399121,
      "learning_rate": 3.065147521219402e-06,
      "loss": 0.68532276,
      "num_input_tokens_seen": 61141155,
      "step": 2841,
      "time_per_iteration": 2.627209186553955
    },
    {
      "auxiliary_loss_clip": 0.01159118,
      "auxiliary_loss_mlp": 0.01035578,
      "balance_loss_clip": 1.05848575,
      "balance_loss_mlp": 1.02666676,
      "epoch": 0.3417302951962965,
      "flos": 43650101566080.0,
      "grad_norm": 1.4637670655832455,
      "language_loss": 0.74448383,
      "learning_rate": 3.064488131621977e-06,
      "loss": 0.7664308,
      "num_input_tokens_seen": 61164480,
      "step": 2842,
      "time_per_iteration": 2.6990067958831787
    },
    {
      "auxiliary_loss_clip": 0.01171966,
      "auxiliary_loss_mlp": 0.01035033,
      "balance_loss_clip": 1.05330873,
      "balance_loss_mlp": 1.02562118,
      "epoch": 0.3418505380869356,
      "flos": 30882207012480.0,
      "grad_norm": 1.8366930298179058,
      "language_loss": 0.73696601,
      "learning_rate": 3.063828580542549e-06,
      "loss": 0.75903594,
      "num_input_tokens_seen": 61185675,
      "step": 2843,
      "time_per_iteration": 3.413469076156616
    },
    {
      "auxiliary_loss_clip": 0.01162901,
      "auxiliary_loss_mlp": 0.01032966,
      "balance_loss_clip": 1.05430579,
      "balance_loss_mlp": 1.02448463,
      "epoch": 0.3419707809775747,
      "flos": 19463871277440.0,
      "grad_norm": 3.4930476029731596,
      "language_loss": 0.73393798,
      "learning_rate": 3.0631688680811706e-06,
      "loss": 0.75589663,
      "num_input_tokens_seen": 61205300,
      "step": 2844,
      "time_per_iteration": 2.522000312805176
    },
    {
      "auxiliary_loss_clip": 0.01195029,
      "auxiliary_loss_mlp": 0.0103834,
      "balance_loss_clip": 1.0588218,
      "balance_loss_mlp": 1.02940512,
      "epoch": 0.3420910238682138,
      "flos": 28727818104960.0,
      "grad_norm": 2.5380852845665056,
      "language_loss": 0.75948596,
      "learning_rate": 3.062508994337921e-06,
      "loss": 0.7818197,
      "num_input_tokens_seen": 61224905,
      "step": 2845,
      "time_per_iteration": 2.5299503803253174
    },
    {
      "auxiliary_loss_clip": 0.01179607,
      "auxiliary_loss_mlp": 0.01030705,
      "balance_loss_clip": 1.05426919,
      "balance_loss_mlp": 1.02167535,
      "epoch": 0.3422112667588529,
      "flos": 21397265758080.0,
      "grad_norm": 1.9335868565547332,
      "language_loss": 0.79656082,
      "learning_rate": 3.0618489594129013e-06,
      "loss": 0.81866395,
      "num_input_tokens_seen": 61243045,
      "step": 2846,
      "time_per_iteration": 2.472214460372925
    },
    {
      "auxiliary_loss_clip": 0.01156333,
      "auxiliary_loss_mlp": 0.01030199,
      "balance_loss_clip": 1.05567932,
      "balance_loss_mlp": 1.02195549,
      "epoch": 0.342331509649492,
      "flos": 13881450038400.0,
      "grad_norm": 2.8792717115825157,
      "language_loss": 0.71314698,
      "learning_rate": 3.061188763406239e-06,
      "loss": 0.73501229,
      "num_input_tokens_seen": 61259190,
      "step": 2847,
      "time_per_iteration": 2.5289528369903564
    },
    {
      "auxiliary_loss_clip": 0.01159031,
      "auxiliary_loss_mlp": 0.01036551,
      "balance_loss_clip": 1.05303383,
      "balance_loss_mlp": 1.02731848,
      "epoch": 0.34245175254013105,
      "flos": 28621450955520.0,
      "grad_norm": 2.0738726317178307,
      "language_loss": 0.82062089,
      "learning_rate": 3.060528406418085e-06,
      "loss": 0.84257674,
      "num_input_tokens_seen": 61279040,
      "step": 2848,
      "time_per_iteration": 3.411067485809326
    },
    {
      "auxiliary_loss_clip": 0.01156205,
      "auxiliary_loss_mlp": 0.01032724,
      "balance_loss_clip": 1.05335414,
      "balance_loss_mlp": 1.02458823,
      "epoch": 0.34257199543077016,
      "flos": 34127058960000.0,
      "grad_norm": 1.7329179094990828,
      "language_loss": 0.61690629,
      "learning_rate": 3.0598678885486145e-06,
      "loss": 0.63879555,
      "num_input_tokens_seen": 61301580,
      "step": 2849,
      "time_per_iteration": 4.286153793334961
    },
    {
      "auxiliary_loss_clip": 0.01152153,
      "auxiliary_loss_mlp": 0.00763574,
      "balance_loss_clip": 1.05148506,
      "balance_loss_mlp": 1.00064421,
      "epoch": 0.34269223832140927,
      "flos": 19974018188160.0,
      "grad_norm": 1.7104306537581189,
      "language_loss": 0.74597412,
      "learning_rate": 3.0592072098980282e-06,
      "loss": 0.76513135,
      "num_input_tokens_seen": 61321240,
      "step": 2850,
      "time_per_iteration": 2.5296194553375244
    },
    {
      "auxiliary_loss_clip": 0.01156511,
      "auxiliary_loss_mlp": 0.01032452,
      "balance_loss_clip": 1.05220044,
      "balance_loss_mlp": 1.02336204,
      "epoch": 0.3428124812120483,
      "flos": 27235658292480.0,
      "grad_norm": 2.0927009317449117,
      "language_loss": 0.72429907,
      "learning_rate": 3.0585463705665514e-06,
      "loss": 0.7461887,
      "num_input_tokens_seen": 61341615,
      "step": 2851,
      "time_per_iteration": 2.558152437210083
    },
    {
      "auxiliary_loss_clip": 0.01147678,
      "auxiliary_loss_mlp": 0.0103319,
      "balance_loss_clip": 1.05035079,
      "balance_loss_mlp": 1.02469015,
      "epoch": 0.34293272410268744,
      "flos": 24570871079040.0,
      "grad_norm": 2.3775123087698216,
      "language_loss": 0.7083534,
      "learning_rate": 3.0578853706544304e-06,
      "loss": 0.73016202,
      "num_input_tokens_seen": 61359005,
      "step": 2852,
      "time_per_iteration": 2.5719010829925537
    },
    {
      "auxiliary_loss_clip": 0.01151815,
      "auxiliary_loss_mlp": 0.00764177,
      "balance_loss_clip": 1.05299461,
      "balance_loss_mlp": 1.00068247,
      "epoch": 0.34305296699332655,
      "flos": 21506865131520.0,
      "grad_norm": 2.533326538591344,
      "language_loss": 0.65476978,
      "learning_rate": 3.0572242102619404e-06,
      "loss": 0.67392975,
      "num_input_tokens_seen": 61376160,
      "step": 2853,
      "time_per_iteration": 2.5469963550567627
    },
    {
      "auxiliary_loss_clip": 0.01161826,
      "auxiliary_loss_mlp": 0.01033595,
      "balance_loss_clip": 1.05493999,
      "balance_loss_mlp": 1.02486849,
      "epoch": 0.3431732098839656,
      "flos": 24056665931520.0,
      "grad_norm": 1.7055883971682353,
      "language_loss": 0.80535883,
      "learning_rate": 3.0565628894893784e-06,
      "loss": 0.82731307,
      "num_input_tokens_seen": 61396795,
      "step": 2854,
      "time_per_iteration": 2.5753142833709717
    },
    {
      "auxiliary_loss_clip": 0.01171231,
      "auxiliary_loss_mlp": 0.01031877,
      "balance_loss_clip": 1.05597413,
      "balance_loss_mlp": 1.0231812,
      "epoch": 0.3432934527746047,
      "flos": 16800879744000.0,
      "grad_norm": 1.6235873889745867,
      "language_loss": 0.74764627,
      "learning_rate": 3.0559014084370655e-06,
      "loss": 0.76967728,
      "num_input_tokens_seen": 61415320,
      "step": 2855,
      "time_per_iteration": 2.5247013568878174
    },
    {
      "auxiliary_loss_clip": 0.01168402,
      "auxiliary_loss_mlp": 0.01029502,
      "balance_loss_clip": 1.05393612,
      "balance_loss_mlp": 1.02011454,
      "epoch": 0.34341369566524377,
      "flos": 23439720908160.0,
      "grad_norm": 1.6235014409123967,
      "language_loss": 0.78580523,
      "learning_rate": 3.055239767205349e-06,
      "loss": 0.80778426,
      "num_input_tokens_seen": 61437070,
      "step": 2856,
      "time_per_iteration": 2.5534403324127197
    },
    {
      "auxiliary_loss_clip": 0.01179992,
      "auxiliary_loss_mlp": 0.01033105,
      "balance_loss_clip": 1.0611037,
      "balance_loss_mlp": 1.02428365,
      "epoch": 0.3435339385558829,
      "flos": 17267466435840.0,
      "grad_norm": 1.8040581741170942,
      "language_loss": 0.78219175,
      "learning_rate": 3.054577965894599e-06,
      "loss": 0.80432272,
      "num_input_tokens_seen": 61453215,
      "step": 2857,
      "time_per_iteration": 2.477968454360962
    },
    {
      "auxiliary_loss_clip": 0.01175765,
      "auxiliary_loss_mlp": 0.01033708,
      "balance_loss_clip": 1.06136703,
      "balance_loss_mlp": 1.02341461,
      "epoch": 0.343654181446522,
      "flos": 22199366413440.0,
      "grad_norm": 1.7485965374897743,
      "language_loss": 0.70266128,
      "learning_rate": 3.0539160046052094e-06,
      "loss": 0.724756,
      "num_input_tokens_seen": 61472915,
      "step": 2858,
      "time_per_iteration": 2.531754493713379
    },
    {
      "auxiliary_loss_clip": 0.01156407,
      "auxiliary_loss_mlp": 0.01036703,
      "balance_loss_clip": 1.05174041,
      "balance_loss_mlp": 1.02662969,
      "epoch": 0.34377442433716104,
      "flos": 19901801894400.0,
      "grad_norm": 2.3387116561969172,
      "language_loss": 0.70100939,
      "learning_rate": 3.0532538834376003e-06,
      "loss": 0.72294044,
      "num_input_tokens_seen": 61492475,
      "step": 2859,
      "time_per_iteration": 2.5214338302612305
    },
    {
      "auxiliary_loss_clip": 0.01181855,
      "auxiliary_loss_mlp": 0.01035845,
      "balance_loss_clip": 1.05591464,
      "balance_loss_mlp": 1.02688098,
      "epoch": 0.34389466722780015,
      "flos": 22197678474240.0,
      "grad_norm": 1.7691035631254464,
      "language_loss": 0.78658879,
      "learning_rate": 3.0525916024922143e-06,
      "loss": 0.80876577,
      "num_input_tokens_seen": 61511660,
      "step": 2860,
      "time_per_iteration": 2.461674213409424
    },
    {
      "auxiliary_loss_clip": 0.01162466,
      "auxiliary_loss_mlp": 0.0103288,
      "balance_loss_clip": 1.05317581,
      "balance_loss_mlp": 1.02409387,
      "epoch": 0.34401491011843927,
      "flos": 18624567110400.0,
      "grad_norm": 2.8709792150611073,
      "language_loss": 0.84490848,
      "learning_rate": 3.0519291618695193e-06,
      "loss": 0.86686194,
      "num_input_tokens_seen": 61529060,
      "step": 2861,
      "time_per_iteration": 2.4676291942596436
    },
    {
      "auxiliary_loss_clip": 0.01140195,
      "auxiliary_loss_mlp": 0.01034243,
      "balance_loss_clip": 1.0478785,
      "balance_loss_mlp": 1.02580857,
      "epoch": 0.3441351530090783,
      "flos": 17858197509120.0,
      "grad_norm": 1.5933669564669064,
      "language_loss": 0.75363326,
      "learning_rate": 3.0512665616700065e-06,
      "loss": 0.77537769,
      "num_input_tokens_seen": 61548125,
      "step": 2862,
      "time_per_iteration": 2.509248971939087
    },
    {
      "auxiliary_loss_clip": 0.01126235,
      "auxiliary_loss_mlp": 0.01035955,
      "balance_loss_clip": 1.04855323,
      "balance_loss_mlp": 1.02740145,
      "epoch": 0.34425539589971743,
      "flos": 23112754381440.0,
      "grad_norm": 1.8061499002611905,
      "language_loss": 0.89301401,
      "learning_rate": 3.0506038019941933e-06,
      "loss": 0.91463596,
      "num_input_tokens_seen": 61568135,
      "step": 2863,
      "time_per_iteration": 2.658419132232666
    },
    {
      "auxiliary_loss_clip": 0.01149556,
      "auxiliary_loss_mlp": 0.01027002,
      "balance_loss_clip": 1.05410445,
      "balance_loss_mlp": 1.01834762,
      "epoch": 0.34437563879035654,
      "flos": 21907699977600.0,
      "grad_norm": 3.3593188937805154,
      "language_loss": 0.67939389,
      "learning_rate": 3.049940882942617e-06,
      "loss": 0.70115948,
      "num_input_tokens_seen": 61586920,
      "step": 2864,
      "time_per_iteration": 2.616769790649414
    },
    {
      "auxiliary_loss_clip": 0.01191192,
      "auxiliary_loss_mlp": 0.01031707,
      "balance_loss_clip": 1.05491972,
      "balance_loss_mlp": 1.02242041,
      "epoch": 0.3444958816809956,
      "flos": 23076915586560.0,
      "grad_norm": 10.237397370839172,
      "language_loss": 0.80480927,
      "learning_rate": 3.0492778046158448e-06,
      "loss": 0.82703823,
      "num_input_tokens_seen": 61608340,
      "step": 2865,
      "time_per_iteration": 2.4431538581848145
    },
    {
      "auxiliary_loss_clip": 0.01178974,
      "auxiliary_loss_mlp": 0.01032186,
      "balance_loss_clip": 1.05908966,
      "balance_loss_mlp": 1.0240171,
      "epoch": 0.3446161245716347,
      "flos": 21908633731200.0,
      "grad_norm": 1.9731235851958284,
      "language_loss": 0.77011299,
      "learning_rate": 3.0486145671144633e-06,
      "loss": 0.79222465,
      "num_input_tokens_seen": 61628130,
      "step": 2866,
      "time_per_iteration": 2.4715888500213623
    },
    {
      "auxiliary_loss_clip": 0.01099033,
      "auxiliary_loss_mlp": 0.01035573,
      "balance_loss_clip": 1.0438087,
      "balance_loss_mlp": 1.0259831,
      "epoch": 0.3447363674622738,
      "flos": 25112834461440.0,
      "grad_norm": 2.543567343040239,
      "language_loss": 0.76915431,
      "learning_rate": 3.047951170539086e-06,
      "loss": 0.7905004,
      "num_input_tokens_seen": 61647755,
      "step": 2867,
      "time_per_iteration": 2.6193974018096924
    },
    {
      "auxiliary_loss_clip": 0.01148248,
      "auxiliary_loss_mlp": 0.01039723,
      "balance_loss_clip": 1.05726612,
      "balance_loss_mlp": 1.03155136,
      "epoch": 0.3448566103529129,
      "flos": 11984684451840.0,
      "grad_norm": 1.8275632005147604,
      "language_loss": 0.83997416,
      "learning_rate": 3.047287614990349e-06,
      "loss": 0.8618539,
      "num_input_tokens_seen": 61665675,
      "step": 2868,
      "time_per_iteration": 2.513073205947876
    },
    {
      "auxiliary_loss_clip": 0.01158404,
      "auxiliary_loss_mlp": 0.01031191,
      "balance_loss_clip": 1.05368495,
      "balance_loss_mlp": 1.02187431,
      "epoch": 0.344976853243552,
      "flos": 40187882465280.0,
      "grad_norm": 2.342040551024905,
      "language_loss": 0.62204897,
      "learning_rate": 3.046623900568914e-06,
      "loss": 0.64394498,
      "num_input_tokens_seen": 61688240,
      "step": 2869,
      "time_per_iteration": 2.650275468826294
    },
    {
      "auxiliary_loss_clip": 0.01159393,
      "auxiliary_loss_mlp": 0.0103366,
      "balance_loss_clip": 1.05224645,
      "balance_loss_mlp": 1.0245111,
      "epoch": 0.34509709613419104,
      "flos": 28723652127360.0,
      "grad_norm": 5.212375758726025,
      "language_loss": 0.69895619,
      "learning_rate": 3.045960027375465e-06,
      "loss": 0.72088671,
      "num_input_tokens_seen": 61706075,
      "step": 2870,
      "time_per_iteration": 3.398531675338745
    },
    {
      "auxiliary_loss_clip": 0.01182474,
      "auxiliary_loss_mlp": 0.01032133,
      "balance_loss_clip": 1.05527139,
      "balance_loss_mlp": 1.02251267,
      "epoch": 0.34521733902483015,
      "flos": 29967597982080.0,
      "grad_norm": 4.708531668857249,
      "language_loss": 0.82405508,
      "learning_rate": 3.045295995510711e-06,
      "loss": 0.84620118,
      "num_input_tokens_seen": 61723045,
      "step": 2871,
      "time_per_iteration": 2.5406501293182373
    },
    {
      "auxiliary_loss_clip": 0.01159498,
      "auxiliary_loss_mlp": 0.01031837,
      "balance_loss_clip": 1.05462241,
      "balance_loss_mlp": 1.02331328,
      "epoch": 0.34533758191546926,
      "flos": 27923059843200.0,
      "grad_norm": 2.0246450591988254,
      "language_loss": 0.7362324,
      "learning_rate": 3.0446318050753865e-06,
      "loss": 0.75814581,
      "num_input_tokens_seen": 61743525,
      "step": 2872,
      "time_per_iteration": 2.5530900955200195
    },
    {
      "auxiliary_loss_clip": 0.01171955,
      "auxiliary_loss_mlp": 0.01031196,
      "balance_loss_clip": 1.05520427,
      "balance_loss_mlp": 1.02305448,
      "epoch": 0.3454578248061083,
      "flos": 27125879351040.0,
      "grad_norm": 2.013350686971414,
      "language_loss": 0.77449453,
      "learning_rate": 3.0439674561702474e-06,
      "loss": 0.79652596,
      "num_input_tokens_seen": 61763025,
      "step": 2873,
      "time_per_iteration": 2.594564914703369
    },
    {
      "auxiliary_loss_clip": 0.01174804,
      "auxiliary_loss_mlp": 0.01031586,
      "balance_loss_clip": 1.05588913,
      "balance_loss_mlp": 1.02320588,
      "epoch": 0.3455780676967474,
      "flos": 19024899166080.0,
      "grad_norm": 2.378899857448509,
      "language_loss": 0.88083708,
      "learning_rate": 3.043302948896076e-06,
      "loss": 0.90290093,
      "num_input_tokens_seen": 61781630,
      "step": 2874,
      "time_per_iteration": 2.458533525466919
    },
    {
      "auxiliary_loss_clip": 0.01124766,
      "auxiliary_loss_mlp": 0.01033048,
      "balance_loss_clip": 1.05099654,
      "balance_loss_mlp": 1.02399421,
      "epoch": 0.34569831058738654,
      "flos": 34496005507200.0,
      "grad_norm": 2.1854991874801555,
      "language_loss": 0.60771257,
      "learning_rate": 3.0426382833536756e-06,
      "loss": 0.6292907,
      "num_input_tokens_seen": 61804985,
      "step": 2875,
      "time_per_iteration": 4.31379508972168
    },
    {
      "auxiliary_loss_clip": 0.01143287,
      "auxiliary_loss_mlp": 0.01030138,
      "balance_loss_clip": 1.04923487,
      "balance_loss_mlp": 1.02156055,
      "epoch": 0.3458185534780256,
      "flos": 31138681098240.0,
      "grad_norm": 2.264687999635517,
      "language_loss": 0.77959049,
      "learning_rate": 3.041973459643877e-06,
      "loss": 0.80132473,
      "num_input_tokens_seen": 61824440,
      "step": 2876,
      "time_per_iteration": 3.3671629428863525
    },
    {
      "auxiliary_loss_clip": 0.0112489,
      "auxiliary_loss_mlp": 0.01030125,
      "balance_loss_clip": 1.04592729,
      "balance_loss_mlp": 1.02159607,
      "epoch": 0.3459387963686647,
      "flos": 32452508862720.0,
      "grad_norm": 2.1421198147396643,
      "language_loss": 0.67067444,
      "learning_rate": 3.0413084778675334e-06,
      "loss": 0.69222462,
      "num_input_tokens_seen": 61845690,
      "step": 2877,
      "time_per_iteration": 2.657008409500122
    },
    {
      "auxiliary_loss_clip": 0.01153671,
      "auxiliary_loss_mlp": 0.00763408,
      "balance_loss_clip": 1.04934549,
      "balance_loss_mlp": 1.00064945,
      "epoch": 0.3460590392593038,
      "flos": 24675658030080.0,
      "grad_norm": 2.0852301773977917,
      "language_loss": 0.8401655,
      "learning_rate": 3.0406433381255214e-06,
      "loss": 0.85933626,
      "num_input_tokens_seen": 61863725,
      "step": 2878,
      "time_per_iteration": 2.548107385635376
    },
    {
      "auxiliary_loss_clip": 0.01175807,
      "auxiliary_loss_mlp": 0.01027541,
      "balance_loss_clip": 1.05775738,
      "balance_loss_mlp": 1.01937246,
      "epoch": 0.34617928214994287,
      "flos": 18807316531200.0,
      "grad_norm": 3.048187612612579,
      "language_loss": 0.82509327,
      "learning_rate": 3.0399780405187425e-06,
      "loss": 0.84712672,
      "num_input_tokens_seen": 61882720,
      "step": 2879,
      "time_per_iteration": 2.463813066482544
    },
    {
      "auxiliary_loss_clip": 0.01174518,
      "auxiliary_loss_mlp": 0.01027048,
      "balance_loss_clip": 1.05523729,
      "balance_loss_mlp": 1.01909065,
      "epoch": 0.346299525040582,
      "flos": 24857653265280.0,
      "grad_norm": 2.4315994336817637,
      "language_loss": 0.78618073,
      "learning_rate": 3.0393125851481216e-06,
      "loss": 0.80819643,
      "num_input_tokens_seen": 61902595,
      "step": 2880,
      "time_per_iteration": 2.514599084854126
    },
    {
      "auxiliary_loss_clip": 0.01146137,
      "auxiliary_loss_mlp": 0.01024417,
      "balance_loss_clip": 1.05446315,
      "balance_loss_mlp": 1.01624501,
      "epoch": 0.3464197679312211,
      "flos": 16434914025600.0,
      "grad_norm": 2.2383979210936613,
      "language_loss": 0.86767185,
      "learning_rate": 3.038646972114608e-06,
      "loss": 0.88937736,
      "num_input_tokens_seen": 61918920,
      "step": 2881,
      "time_per_iteration": 2.5247910022735596
    },
    {
      "auxiliary_loss_clip": 0.0114468,
      "auxiliary_loss_mlp": 0.01041295,
      "balance_loss_clip": 1.05382347,
      "balance_loss_mlp": 1.03275335,
      "epoch": 0.34654001082186014,
      "flos": 22382474970240.0,
      "grad_norm": 1.7562835057200905,
      "language_loss": 0.67518997,
      "learning_rate": 3.037981201519174e-06,
      "loss": 0.69704974,
      "num_input_tokens_seen": 61939520,
      "step": 2882,
      "time_per_iteration": 2.561526298522949
    },
    {
      "auxiliary_loss_clip": 0.01178516,
      "auxiliary_loss_mlp": 0.01033042,
      "balance_loss_clip": 1.05910099,
      "balance_loss_mlp": 1.02461433,
      "epoch": 0.34666025371249926,
      "flos": 19573901614080.0,
      "grad_norm": 2.7169571989647916,
      "language_loss": 0.71286345,
      "learning_rate": 3.0373152734628175e-06,
      "loss": 0.73497903,
      "num_input_tokens_seen": 61957800,
      "step": 2883,
      "time_per_iteration": 2.4900712966918945
    },
    {
      "auxiliary_loss_clip": 0.01172466,
      "auxiliary_loss_mlp": 0.01027372,
      "balance_loss_clip": 1.0540055,
      "balance_loss_mlp": 1.01847339,
      "epoch": 0.34678049660313837,
      "flos": 15267637751040.0,
      "grad_norm": 2.152190947245872,
      "language_loss": 0.76223826,
      "learning_rate": 3.0366491880465584e-06,
      "loss": 0.78423667,
      "num_input_tokens_seen": 61975820,
      "step": 2884,
      "time_per_iteration": 2.4612512588500977
    },
    {
      "auxiliary_loss_clip": 0.01195421,
      "auxiliary_loss_mlp": 0.0103248,
      "balance_loss_clip": 1.059726,
      "balance_loss_mlp": 1.02328932,
      "epoch": 0.3469007394937774,
      "flos": 21181550630400.0,
      "grad_norm": 1.613784129173304,
      "language_loss": 0.82047951,
      "learning_rate": 3.035982945371443e-06,
      "loss": 0.84275854,
      "num_input_tokens_seen": 61997515,
      "step": 2885,
      "time_per_iteration": 2.4990787506103516
    },
    {
      "auxiliary_loss_clip": 0.01170348,
      "auxiliary_loss_mlp": 0.01029612,
      "balance_loss_clip": 1.05553651,
      "balance_loss_mlp": 1.02049804,
      "epoch": 0.34702098238441653,
      "flos": 22375471818240.0,
      "grad_norm": 2.055747682232124,
      "language_loss": 0.85413849,
      "learning_rate": 3.035316545538537e-06,
      "loss": 0.87613809,
      "num_input_tokens_seen": 62016310,
      "step": 2886,
      "time_per_iteration": 2.521435260772705
    },
    {
      "auxiliary_loss_clip": 0.01165092,
      "auxiliary_loss_mlp": 0.01031167,
      "balance_loss_clip": 1.06063163,
      "balance_loss_mlp": 1.02253628,
      "epoch": 0.3471412252750556,
      "flos": 22929430343040.0,
      "grad_norm": 2.649077036405507,
      "language_loss": 0.7893368,
      "learning_rate": 3.034649988648935e-06,
      "loss": 0.81129944,
      "num_input_tokens_seen": 62036075,
      "step": 2887,
      "time_per_iteration": 2.516091823577881
    },
    {
      "auxiliary_loss_clip": 0.01167655,
      "auxiliary_loss_mlp": 0.01026649,
      "balance_loss_clip": 1.0553093,
      "balance_loss_mlp": 1.01782775,
      "epoch": 0.3472614681656947,
      "flos": 21324259365120.0,
      "grad_norm": 1.682397715866157,
      "language_loss": 0.8068493,
      "learning_rate": 3.033983274803752e-06,
      "loss": 0.82879233,
      "num_input_tokens_seen": 62055865,
      "step": 2888,
      "time_per_iteration": 2.536759853363037
    },
    {
      "auxiliary_loss_clip": 0.01160019,
      "auxiliary_loss_mlp": 0.01031533,
      "balance_loss_clip": 1.05383372,
      "balance_loss_mlp": 1.02261615,
      "epoch": 0.3473817110563338,
      "flos": 23475739271040.0,
      "grad_norm": 4.246333451645234,
      "language_loss": 0.7272895,
      "learning_rate": 3.0333164041041283e-06,
      "loss": 0.74920505,
      "num_input_tokens_seen": 62072180,
      "step": 2889,
      "time_per_iteration": 2.5323047637939453
    },
    {
      "auxiliary_loss_clip": 0.01121529,
      "auxiliary_loss_mlp": 0.01025175,
      "balance_loss_clip": 1.04906702,
      "balance_loss_mlp": 1.01687253,
      "epoch": 0.34750195394697286,
      "flos": 22346025644160.0,
      "grad_norm": 2.175184053088743,
      "language_loss": 0.71810549,
      "learning_rate": 3.032649376651228e-06,
      "loss": 0.73957253,
      "num_input_tokens_seen": 62091600,
      "step": 2890,
      "time_per_iteration": 2.6042559146881104
    },
    {
      "auxiliary_loss_clip": 0.01150244,
      "auxiliary_loss_mlp": 0.01027838,
      "balance_loss_clip": 1.05282807,
      "balance_loss_mlp": 1.01841486,
      "epoch": 0.347622196837612,
      "flos": 29095004885760.0,
      "grad_norm": 3.4002649271359715,
      "language_loss": 0.75715297,
      "learning_rate": 3.031982192546238e-06,
      "loss": 0.77893382,
      "num_input_tokens_seen": 62114695,
      "step": 2891,
      "time_per_iteration": 2.593083381652832
    },
    {
      "auxiliary_loss_clip": 0.01178937,
      "auxiliary_loss_mlp": 0.01032835,
      "balance_loss_clip": 1.05558956,
      "balance_loss_mlp": 1.02431178,
      "epoch": 0.3477424397282511,
      "flos": 22455732758400.0,
      "grad_norm": 1.996430127765452,
      "language_loss": 0.94606566,
      "learning_rate": 3.0313148518903696e-06,
      "loss": 0.9681834,
      "num_input_tokens_seen": 62134520,
      "step": 2892,
      "time_per_iteration": 2.484234571456909
    },
    {
      "auxiliary_loss_clip": 0.01167416,
      "auxiliary_loss_mlp": 0.01028234,
      "balance_loss_clip": 1.05686927,
      "balance_loss_mlp": 1.01939476,
      "epoch": 0.34786268261889014,
      "flos": 15778790242560.0,
      "grad_norm": 2.0046582628429657,
      "language_loss": 0.81266308,
      "learning_rate": 3.030647354784859e-06,
      "loss": 0.83461958,
      "num_input_tokens_seen": 62151560,
      "step": 2893,
      "time_per_iteration": 2.498206377029419
    },
    {
      "auxiliary_loss_clip": 0.01150919,
      "auxiliary_loss_mlp": 0.01033242,
      "balance_loss_clip": 1.05435252,
      "balance_loss_mlp": 1.02456331,
      "epoch": 0.34798292550952925,
      "flos": 20777627214720.0,
      "grad_norm": 1.7405890773186343,
      "language_loss": 0.76891041,
      "learning_rate": 3.029979701330964e-06,
      "loss": 0.79075205,
      "num_input_tokens_seen": 62170985,
      "step": 2894,
      "time_per_iteration": 2.5509817600250244
    },
    {
      "auxiliary_loss_clip": 0.0116866,
      "auxiliary_loss_mlp": 0.01032342,
      "balance_loss_clip": 1.05564237,
      "balance_loss_mlp": 1.023646,
      "epoch": 0.34810316840016836,
      "flos": 19937820257280.0,
      "grad_norm": 2.271278945655884,
      "language_loss": 0.80084401,
      "learning_rate": 3.029311891629966e-06,
      "loss": 0.82285404,
      "num_input_tokens_seen": 62189440,
      "step": 2895,
      "time_per_iteration": 2.5065693855285645
    },
    {
      "auxiliary_loss_clip": 0.01159761,
      "auxiliary_loss_mlp": 0.01036377,
      "balance_loss_clip": 1.05475485,
      "balance_loss_mlp": 1.02717996,
      "epoch": 0.3482234112908074,
      "flos": 23623296341760.0,
      "grad_norm": 1.7273877714629253,
      "language_loss": 0.73963678,
      "learning_rate": 3.0286439257831744e-06,
      "loss": 0.76159817,
      "num_input_tokens_seen": 62208910,
      "step": 2896,
      "time_per_iteration": 3.395796298980713
    },
    {
      "auxiliary_loss_clip": 0.01198129,
      "auxiliary_loss_mlp": 0.01037214,
      "balance_loss_clip": 1.05938578,
      "balance_loss_mlp": 1.02655709,
      "epoch": 0.3483436541814465,
      "flos": 23986712194560.0,
      "grad_norm": 3.16386010005365,
      "language_loss": 0.71699721,
      "learning_rate": 3.0279758038919156e-06,
      "loss": 0.73935062,
      "num_input_tokens_seen": 62227135,
      "step": 2897,
      "time_per_iteration": 2.4587204456329346
    },
    {
      "auxiliary_loss_clip": 0.01179106,
      "auxiliary_loss_mlp": 0.01032963,
      "balance_loss_clip": 1.05689836,
      "balance_loss_mlp": 1.02384305,
      "epoch": 0.34846389707208564,
      "flos": 22638338524800.0,
      "grad_norm": 1.6537049686488743,
      "language_loss": 0.77854073,
      "learning_rate": 3.0273075260575455e-06,
      "loss": 0.80066139,
      "num_input_tokens_seen": 62246035,
      "step": 2898,
      "time_per_iteration": 2.4945077896118164
    },
    {
      "auxiliary_loss_clip": 0.01166885,
      "auxiliary_loss_mlp": 0.01035079,
      "balance_loss_clip": 1.05392015,
      "balance_loss_mlp": 1.02496982,
      "epoch": 0.3485841399627247,
      "flos": 21792857218560.0,
      "grad_norm": 1.8870563275997037,
      "language_loss": 0.80711806,
      "learning_rate": 3.0266390923814396e-06,
      "loss": 0.82913768,
      "num_input_tokens_seen": 62264095,
      "step": 2899,
      "time_per_iteration": 2.508171319961548
    },
    {
      "auxiliary_loss_clip": 0.01170637,
      "auxiliary_loss_mlp": 0.01037961,
      "balance_loss_clip": 1.06040359,
      "balance_loss_mlp": 1.02800679,
      "epoch": 0.3487043828533638,
      "flos": 17019036996480.0,
      "grad_norm": 1.8152172148819326,
      "language_loss": 0.82229769,
      "learning_rate": 3.0259705029650008e-06,
      "loss": 0.8443836,
      "num_input_tokens_seen": 62282025,
      "step": 2900,
      "time_per_iteration": 2.485881805419922
    },
    {
      "auxiliary_loss_clip": 0.01180616,
      "auxiliary_loss_mlp": 0.01026141,
      "balance_loss_clip": 1.05577207,
      "balance_loss_mlp": 1.01788545,
      "epoch": 0.34882462574400286,
      "flos": 22601135013120.0,
      "grad_norm": 1.6252599133425325,
      "language_loss": 0.72880208,
      "learning_rate": 3.025301757909652e-06,
      "loss": 0.75086969,
      "num_input_tokens_seen": 62302220,
      "step": 2901,
      "time_per_iteration": 2.4926414489746094
    },
    {
      "auxiliary_loss_clip": 0.0115377,
      "auxiliary_loss_mlp": 0.00764605,
      "balance_loss_clip": 1.05450368,
      "balance_loss_mlp": 1.00065589,
      "epoch": 0.34894486863464197,
      "flos": 29861518141440.0,
      "grad_norm": 1.783490674131647,
      "language_loss": 0.80567622,
      "learning_rate": 3.024632857316842e-06,
      "loss": 0.82485998,
      "num_input_tokens_seen": 62323535,
      "step": 2902,
      "time_per_iteration": 4.201333999633789
    },
    {
      "auxiliary_loss_clip": 0.0118364,
      "auxiliary_loss_mlp": 0.01028962,
      "balance_loss_clip": 1.06009841,
      "balance_loss_mlp": 1.01966381,
      "epoch": 0.3490651115252811,
      "flos": 22122265870080.0,
      "grad_norm": 1.919950527440679,
      "language_loss": 0.77397877,
      "learning_rate": 3.0239638012880412e-06,
      "loss": 0.79610479,
      "num_input_tokens_seen": 62343430,
      "step": 2903,
      "time_per_iteration": 3.2788820266723633
    },
    {
      "auxiliary_loss_clip": 0.0112799,
      "auxiliary_loss_mlp": 0.01031676,
      "balance_loss_clip": 1.04885983,
      "balance_loss_mlp": 1.02212143,
      "epoch": 0.34918535441592014,
      "flos": 12676682943360.0,
      "grad_norm": 2.3410715635561457,
      "language_loss": 0.81419742,
      "learning_rate": 3.0232945899247466e-06,
      "loss": 0.83579409,
      "num_input_tokens_seen": 62360365,
      "step": 2904,
      "time_per_iteration": 2.7015492916107178
    },
    {
      "auxiliary_loss_clip": 0.01178911,
      "auxiliary_loss_mlp": 0.01037029,
      "balance_loss_clip": 1.05403042,
      "balance_loss_mlp": 1.02729511,
      "epoch": 0.34930559730655925,
      "flos": 23185617120000.0,
      "grad_norm": 1.776881880303038,
      "language_loss": 0.77339458,
      "learning_rate": 3.022625223328476e-06,
      "loss": 0.79555392,
      "num_input_tokens_seen": 62382105,
      "step": 2905,
      "time_per_iteration": 2.6516213417053223
    },
    {
      "auxiliary_loss_clip": 0.01186029,
      "auxiliary_loss_mlp": 0.01035203,
      "balance_loss_clip": 1.05825877,
      "balance_loss_mlp": 1.02438486,
      "epoch": 0.34942584019719836,
      "flos": 22855023319680.0,
      "grad_norm": 1.393738410815465,
      "language_loss": 0.69099772,
      "learning_rate": 3.0219557016007723e-06,
      "loss": 0.71321005,
      "num_input_tokens_seen": 62402235,
      "step": 2906,
      "time_per_iteration": 2.589181661605835
    },
    {
      "auxiliary_loss_clip": 0.01176101,
      "auxiliary_loss_mlp": 0.01030392,
      "balance_loss_clip": 1.05693066,
      "balance_loss_mlp": 1.02091467,
      "epoch": 0.3495460830878374,
      "flos": 24426043441920.0,
      "grad_norm": 1.8369676914442659,
      "language_loss": 0.6969316,
      "learning_rate": 3.021286024843202e-06,
      "loss": 0.71899647,
      "num_input_tokens_seen": 62420430,
      "step": 2907,
      "time_per_iteration": 2.490055799484253
    },
    {
      "auxiliary_loss_clip": 0.0109323,
      "auxiliary_loss_mlp": 0.01002864,
      "balance_loss_clip": 1.02513075,
      "balance_loss_mlp": 1.00180876,
      "epoch": 0.3496663259784765,
      "flos": 70008749389440.0,
      "grad_norm": 1.0726586884330434,
      "language_loss": 0.6485374,
      "learning_rate": 3.0206161931573526e-06,
      "loss": 0.66949832,
      "num_input_tokens_seen": 62472980,
      "step": 2908,
      "time_per_iteration": 2.960287570953369
    },
    {
      "auxiliary_loss_clip": 0.01160997,
      "auxiliary_loss_mlp": 0.01035073,
      "balance_loss_clip": 1.05074692,
      "balance_loss_mlp": 1.0265795,
      "epoch": 0.34978656886911563,
      "flos": 28692805322880.0,
      "grad_norm": 1.603290085295026,
      "language_loss": 0.92862695,
      "learning_rate": 3.0199462066448388e-06,
      "loss": 0.95058757,
      "num_input_tokens_seen": 62495175,
      "step": 2909,
      "time_per_iteration": 2.577993392944336
    },
    {
      "auxiliary_loss_clip": 0.01181659,
      "auxiliary_loss_mlp": 0.01030685,
      "balance_loss_clip": 1.05983531,
      "balance_loss_mlp": 1.02118409,
      "epoch": 0.3499068117597547,
      "flos": 21142156389120.0,
      "grad_norm": 1.8314536908633274,
      "language_loss": 0.69664884,
      "learning_rate": 3.019276065407296e-06,
      "loss": 0.71877223,
      "num_input_tokens_seen": 62514295,
      "step": 2910,
      "time_per_iteration": 2.47017765045166
    },
    {
      "auxiliary_loss_clip": 0.01138413,
      "auxiliary_loss_mlp": 0.01035426,
      "balance_loss_clip": 1.05168366,
      "balance_loss_mlp": 1.02610993,
      "epoch": 0.3500270546503938,
      "flos": 22782699285120.0,
      "grad_norm": 1.8468371536414188,
      "language_loss": 0.80348551,
      "learning_rate": 3.018605769546385e-06,
      "loss": 0.82522392,
      "num_input_tokens_seen": 62534850,
      "step": 2911,
      "time_per_iteration": 2.575129508972168
    },
    {
      "auxiliary_loss_clip": 0.0117745,
      "auxiliary_loss_mlp": 0.01034137,
      "balance_loss_clip": 1.05382395,
      "balance_loss_mlp": 1.0245049,
      "epoch": 0.3501472975410329,
      "flos": 22894058424960.0,
      "grad_norm": 1.804305570846599,
      "language_loss": 0.79593498,
      "learning_rate": 3.017935319163788e-06,
      "loss": 0.81805086,
      "num_input_tokens_seen": 62553810,
      "step": 2912,
      "time_per_iteration": 2.48646879196167
    },
    {
      "auxiliary_loss_clip": 0.01180931,
      "auxiliary_loss_mlp": 0.01034433,
      "balance_loss_clip": 1.05763423,
      "balance_loss_mlp": 1.0240376,
      "epoch": 0.35026754043167196,
      "flos": 25446588658560.0,
      "grad_norm": 3.3857023562322914,
      "language_loss": 0.70608127,
      "learning_rate": 3.017264714361213e-06,
      "loss": 0.72823489,
      "num_input_tokens_seen": 62573460,
      "step": 2913,
      "time_per_iteration": 2.4986729621887207
    },
    {
      "auxiliary_loss_clip": 0.01164549,
      "auxiliary_loss_mlp": 0.00764494,
      "balance_loss_clip": 1.05506444,
      "balance_loss_mlp": 1.00079739,
      "epoch": 0.3503877833223111,
      "flos": 19573757959680.0,
      "grad_norm": 1.8962038542750148,
      "language_loss": 0.82142198,
      "learning_rate": 3.016593955240389e-06,
      "loss": 0.84071243,
      "num_input_tokens_seen": 62592150,
      "step": 2914,
      "time_per_iteration": 2.5073235034942627
    },
    {
      "auxiliary_loss_clip": 0.01079893,
      "auxiliary_loss_mlp": 0.010011,
      "balance_loss_clip": 1.02345228,
      "balance_loss_mlp": 1.0000447,
      "epoch": 0.3505080262129502,
      "flos": 65072075880960.0,
      "grad_norm": 0.822748475037358,
      "language_loss": 0.63760668,
      "learning_rate": 3.015923041903071e-06,
      "loss": 0.65841663,
      "num_input_tokens_seen": 62658275,
      "step": 2915,
      "time_per_iteration": 3.100501537322998
    },
    {
      "auxiliary_loss_clip": 0.01181317,
      "auxiliary_loss_mlp": 0.01030381,
      "balance_loss_clip": 1.06030369,
      "balance_loss_mlp": 1.02055824,
      "epoch": 0.35062826910358924,
      "flos": 29314562768640.0,
      "grad_norm": 2.3166085606717055,
      "language_loss": 0.8339988,
      "learning_rate": 3.0152519744510347e-06,
      "loss": 0.85611582,
      "num_input_tokens_seen": 62678075,
      "step": 2916,
      "time_per_iteration": 2.5334339141845703
    },
    {
      "auxiliary_loss_clip": 0.01152459,
      "auxiliary_loss_mlp": 0.01033822,
      "balance_loss_clip": 1.05300522,
      "balance_loss_mlp": 1.0243094,
      "epoch": 0.35074851199422835,
      "flos": 23987717775360.0,
      "grad_norm": 1.8136290003301474,
      "language_loss": 0.82546228,
      "learning_rate": 3.014580752986081e-06,
      "loss": 0.84732509,
      "num_input_tokens_seen": 62696950,
      "step": 2917,
      "time_per_iteration": 2.5631556510925293
    },
    {
      "auxiliary_loss_clip": 0.01139605,
      "auxiliary_loss_mlp": 0.01035894,
      "balance_loss_clip": 1.05364418,
      "balance_loss_mlp": 1.02705419,
      "epoch": 0.3508687548848674,
      "flos": 15224436668160.0,
      "grad_norm": 2.3067430055861204,
      "language_loss": 0.78285944,
      "learning_rate": 3.0139093776100345e-06,
      "loss": 0.80461442,
      "num_input_tokens_seen": 62713540,
      "step": 2918,
      "time_per_iteration": 2.5434465408325195
    },
    {
      "auxiliary_loss_clip": 0.01190736,
      "auxiliary_loss_mlp": 0.01029792,
      "balance_loss_clip": 1.05679631,
      "balance_loss_mlp": 1.02061892,
      "epoch": 0.3509889977755065,
      "flos": 21361750185600.0,
      "grad_norm": 1.7939210097053386,
      "language_loss": 0.75197208,
      "learning_rate": 3.013237848424741e-06,
      "loss": 0.77417737,
      "num_input_tokens_seen": 62732925,
      "step": 2919,
      "time_per_iteration": 2.4598734378814697
    },
    {
      "auxiliary_loss_clip": 0.01166497,
      "auxiliary_loss_mlp": 0.01034552,
      "balance_loss_clip": 1.05639613,
      "balance_loss_mlp": 1.0256592,
      "epoch": 0.35110924066614563,
      "flos": 19135360465920.0,
      "grad_norm": 4.647650362100503,
      "language_loss": 0.75478858,
      "learning_rate": 3.012566165532072e-06,
      "loss": 0.77679908,
      "num_input_tokens_seen": 62751715,
      "step": 2920,
      "time_per_iteration": 2.514832019805908
    },
    {
      "auxiliary_loss_clip": 0.01129612,
      "auxiliary_loss_mlp": 0.01034139,
      "balance_loss_clip": 1.05160165,
      "balance_loss_mlp": 1.02517498,
      "epoch": 0.3512294835567847,
      "flos": 21980885938560.0,
      "grad_norm": 1.9924305951716799,
      "language_loss": 0.76581919,
      "learning_rate": 3.0118943290339207e-06,
      "loss": 0.78745663,
      "num_input_tokens_seen": 62771925,
      "step": 2921,
      "time_per_iteration": 2.6265459060668945
    },
    {
      "auxiliary_loss_clip": 0.01142973,
      "auxiliary_loss_mlp": 0.01031763,
      "balance_loss_clip": 1.0510267,
      "balance_loss_mlp": 1.02229762,
      "epoch": 0.3513497264474238,
      "flos": 17817294896640.0,
      "grad_norm": 2.1042732354260294,
      "language_loss": 0.68369895,
      "learning_rate": 3.011222339032204e-06,
      "loss": 0.7054463,
      "num_input_tokens_seen": 62790075,
      "step": 2922,
      "time_per_iteration": 2.5352516174316406
    },
    {
      "auxiliary_loss_clip": 0.01191693,
      "auxiliary_loss_mlp": 0.01037479,
      "balance_loss_clip": 1.05668116,
      "balance_loss_mlp": 1.02787066,
      "epoch": 0.3514699693380629,
      "flos": 26943417239040.0,
      "grad_norm": 1.7984535259243666,
      "language_loss": 0.69699448,
      "learning_rate": 3.0105501956288626e-06,
      "loss": 0.7192862,
      "num_input_tokens_seen": 62810545,
      "step": 2923,
      "time_per_iteration": 3.3468363285064697
    },
    {
      "auxiliary_loss_clip": 0.0118585,
      "auxiliary_loss_mlp": 0.0103405,
      "balance_loss_clip": 1.05740762,
      "balance_loss_mlp": 1.02428055,
      "epoch": 0.35159021222870196,
      "flos": 15267565923840.0,
      "grad_norm": 1.9478105332390803,
      "language_loss": 0.72798163,
      "learning_rate": 3.0098778989258602e-06,
      "loss": 0.7501806,
      "num_input_tokens_seen": 62829155,
      "step": 2924,
      "time_per_iteration": 2.454888105392456
    },
    {
      "auxiliary_loss_clip": 0.01145254,
      "auxiliary_loss_mlp": 0.01036472,
      "balance_loss_clip": 1.05203819,
      "balance_loss_mlp": 1.02738249,
      "epoch": 0.35171045511934107,
      "flos": 13984154000640.0,
      "grad_norm": 2.951258086630555,
      "language_loss": 0.8815009,
      "learning_rate": 3.009205449025183e-06,
      "loss": 0.90331817,
      "num_input_tokens_seen": 62845350,
      "step": 2925,
      "time_per_iteration": 2.51007080078125
    },
    {
      "auxiliary_loss_clip": 0.01146694,
      "auxiliary_loss_mlp": 0.01033294,
      "balance_loss_clip": 1.05057228,
      "balance_loss_mlp": 1.02404332,
      "epoch": 0.3518306980099802,
      "flos": 14283434119680.0,
      "grad_norm": 1.9469640914707576,
      "language_loss": 0.63222831,
      "learning_rate": 3.008532846028842e-06,
      "loss": 0.65402818,
      "num_input_tokens_seen": 62862110,
      "step": 2926,
      "time_per_iteration": 2.513120412826538
    },
    {
      "auxiliary_loss_clip": 0.01197455,
      "auxiliary_loss_mlp": 0.01037714,
      "balance_loss_clip": 1.06008434,
      "balance_loss_mlp": 1.02777159,
      "epoch": 0.35195094090061924,
      "flos": 27052872958080.0,
      "grad_norm": 2.6667906545515296,
      "language_loss": 0.72278976,
      "learning_rate": 3.0078600900388694e-06,
      "loss": 0.74514151,
      "num_input_tokens_seen": 62882415,
      "step": 2927,
      "time_per_iteration": 2.4651401042938232
    },
    {
      "auxiliary_loss_clip": 0.01138861,
      "auxiliary_loss_mlp": 0.01031047,
      "balance_loss_clip": 1.04865503,
      "balance_loss_mlp": 1.02143335,
      "epoch": 0.35207118379125835,
      "flos": 25629266252160.0,
      "grad_norm": 1.8224458948767925,
      "language_loss": 0.73845279,
      "learning_rate": 3.007187181157323e-06,
      "loss": 0.76015192,
      "num_input_tokens_seen": 62902425,
      "step": 2928,
      "time_per_iteration": 2.5711920261383057
    },
    {
      "auxiliary_loss_clip": 0.0111219,
      "auxiliary_loss_mlp": 0.01028789,
      "balance_loss_clip": 1.04720783,
      "balance_loss_mlp": 1.01982498,
      "epoch": 0.35219142668189746,
      "flos": 18004713085440.0,
      "grad_norm": 2.219903385204011,
      "language_loss": 0.68447924,
      "learning_rate": 3.006514119486282e-06,
      "loss": 0.70588905,
      "num_input_tokens_seen": 62919255,
      "step": 2929,
      "time_per_iteration": 4.204680681228638
    },
    {
      "auxiliary_loss_clip": 0.01142461,
      "auxiliary_loss_mlp": 0.01027718,
      "balance_loss_clip": 1.05109572,
      "balance_loss_mlp": 1.01898611,
      "epoch": 0.3523116695725365,
      "flos": 14028109269120.0,
      "grad_norm": 1.7345613065517642,
      "language_loss": 0.69631529,
      "learning_rate": 3.005840905127849e-06,
      "loss": 0.7180171,
      "num_input_tokens_seen": 62936160,
      "step": 2930,
      "time_per_iteration": 2.516960382461548
    },
    {
      "auxiliary_loss_clip": 0.01193654,
      "auxiliary_loss_mlp": 0.00764166,
      "balance_loss_clip": 1.06096816,
      "balance_loss_mlp": 1.00087798,
      "epoch": 0.3524319124631756,
      "flos": 21433966479360.0,
      "grad_norm": 2.3048826319079865,
      "language_loss": 0.87105119,
      "learning_rate": 3.0051675381841516e-06,
      "loss": 0.89062941,
      "num_input_tokens_seen": 62953470,
      "step": 2931,
      "time_per_iteration": 2.4634788036346436
    },
    {
      "auxiliary_loss_clip": 0.01106299,
      "auxiliary_loss_mlp": 0.01035957,
      "balance_loss_clip": 1.04673111,
      "balance_loss_mlp": 1.02665281,
      "epoch": 0.3525521553538147,
      "flos": 26322773114880.0,
      "grad_norm": 1.600391766366554,
      "language_loss": 0.76800299,
      "learning_rate": 3.0044940187573363e-06,
      "loss": 0.78942549,
      "num_input_tokens_seen": 62974480,
      "step": 2932,
      "time_per_iteration": 2.670949935913086
    },
    {
      "auxiliary_loss_clip": 0.01182458,
      "auxiliary_loss_mlp": 0.01034208,
      "balance_loss_clip": 1.05614138,
      "balance_loss_mlp": 1.02508903,
      "epoch": 0.3526723982444538,
      "flos": 21543314457600.0,
      "grad_norm": 1.7523460779030162,
      "language_loss": 0.65343928,
      "learning_rate": 3.003820346949578e-06,
      "loss": 0.67560595,
      "num_input_tokens_seen": 62992560,
      "step": 2933,
      "time_per_iteration": 2.576967477798462
    },
    {
      "auxiliary_loss_clip": 0.01192051,
      "auxiliary_loss_mlp": 0.01033635,
      "balance_loss_clip": 1.05648708,
      "balance_loss_mlp": 1.02417004,
      "epoch": 0.3527926411350929,
      "flos": 23733649900800.0,
      "grad_norm": 1.8096476756193962,
      "language_loss": 0.79540652,
      "learning_rate": 3.003146522863071e-06,
      "loss": 0.81766337,
      "num_input_tokens_seen": 63013445,
      "step": 2934,
      "time_per_iteration": 2.5080907344818115
    },
    {
      "auxiliary_loss_clip": 0.0116387,
      "auxiliary_loss_mlp": 0.010317,
      "balance_loss_clip": 1.05665207,
      "balance_loss_mlp": 1.02315843,
      "epoch": 0.35291288402573195,
      "flos": 30445461544320.0,
      "grad_norm": 2.1890740253179786,
      "language_loss": 0.85630852,
      "learning_rate": 3.0024725466000345e-06,
      "loss": 0.87826431,
      "num_input_tokens_seen": 63033400,
      "step": 2935,
      "time_per_iteration": 2.5992751121520996
    },
    {
      "auxiliary_loss_clip": 0.01179127,
      "auxiliary_loss_mlp": 0.01027721,
      "balance_loss_clip": 1.05850887,
      "balance_loss_mlp": 1.01947129,
      "epoch": 0.35303312691637107,
      "flos": 23112179763840.0,
      "grad_norm": 1.662825815363589,
      "language_loss": 0.78672743,
      "learning_rate": 3.0017984182627087e-06,
      "loss": 0.80879587,
      "num_input_tokens_seen": 63052725,
      "step": 2936,
      "time_per_iteration": 2.496722459793091
    },
    {
      "auxiliary_loss_clip": 0.01148166,
      "auxiliary_loss_mlp": 0.00764392,
      "balance_loss_clip": 1.0506084,
      "balance_loss_mlp": 1.00084865,
      "epoch": 0.3531533698070102,
      "flos": 21835699165440.0,
      "grad_norm": 18.760500707031348,
      "language_loss": 0.82230437,
      "learning_rate": 3.00112413795336e-06,
      "loss": 0.84143001,
      "num_input_tokens_seen": 63072560,
      "step": 2937,
      "time_per_iteration": 2.68031644821167
    },
    {
      "auxiliary_loss_clip": 0.01159717,
      "auxiliary_loss_mlp": 0.01032299,
      "balance_loss_clip": 1.05035639,
      "balance_loss_mlp": 1.02319145,
      "epoch": 0.35327361269764923,
      "flos": 15778969810560.0,
      "grad_norm": 1.8526473721634917,
      "language_loss": 0.8030709,
      "learning_rate": 3.000449705774275e-06,
      "loss": 0.82499111,
      "num_input_tokens_seen": 63090800,
      "step": 2938,
      "time_per_iteration": 2.502084732055664
    },
    {
      "auxiliary_loss_clip": 0.01179959,
      "auxiliary_loss_mlp": 0.01027559,
      "balance_loss_clip": 1.05810738,
      "balance_loss_mlp": 1.01852274,
      "epoch": 0.35339385558828834,
      "flos": 22090413484800.0,
      "grad_norm": 2.5738479897557567,
      "language_loss": 0.71376145,
      "learning_rate": 2.9997751218277654e-06,
      "loss": 0.73583663,
      "num_input_tokens_seen": 63108955,
      "step": 2939,
      "time_per_iteration": 2.479315996170044
    },
    {
      "auxiliary_loss_clip": 0.01195111,
      "auxiliary_loss_mlp": 0.01031689,
      "balance_loss_clip": 1.06016338,
      "balance_loss_mlp": 1.02243233,
      "epoch": 0.35351409847892745,
      "flos": 24165008328960.0,
      "grad_norm": 1.883787022771459,
      "language_loss": 0.77750146,
      "learning_rate": 2.999100386216166e-06,
      "loss": 0.79976946,
      "num_input_tokens_seen": 63127895,
      "step": 2940,
      "time_per_iteration": 2.445446729660034
    },
    {
      "auxiliary_loss_clip": 0.01164061,
      "auxiliary_loss_mlp": 0.01030784,
      "balance_loss_clip": 1.05529094,
      "balance_loss_mlp": 1.02202201,
      "epoch": 0.3536343413695665,
      "flos": 27052298340480.0,
      "grad_norm": 1.8192266325998778,
      "language_loss": 0.7428627,
      "learning_rate": 2.998425499041831e-06,
      "loss": 0.76481116,
      "num_input_tokens_seen": 63148410,
      "step": 2941,
      "time_per_iteration": 2.552973508834839
    },
    {
      "auxiliary_loss_clip": 0.01077976,
      "auxiliary_loss_mlp": 0.01001691,
      "balance_loss_clip": 1.02188063,
      "balance_loss_mlp": 1.00062418,
      "epoch": 0.3537545842602056,
      "flos": 65991066370560.0,
      "grad_norm": 1.3645667159315482,
      "language_loss": 0.64607334,
      "learning_rate": 2.997750460407142e-06,
      "loss": 0.66687,
      "num_input_tokens_seen": 63209765,
      "step": 2942,
      "time_per_iteration": 3.092388868331909
    },
    {
      "auxiliary_loss_clip": 0.01150851,
      "auxiliary_loss_mlp": 0.01028865,
      "balance_loss_clip": 1.04984069,
      "balance_loss_mlp": 1.01954281,
      "epoch": 0.35387482715084473,
      "flos": 18436897526400.0,
      "grad_norm": 2.0288066105800926,
      "language_loss": 0.70046389,
      "learning_rate": 2.997075270414501e-06,
      "loss": 0.72226107,
      "num_input_tokens_seen": 63226980,
      "step": 2943,
      "time_per_iteration": 2.515299081802368
    },
    {
      "auxiliary_loss_clip": 0.01068478,
      "auxiliary_loss_mlp": 0.01000876,
      "balance_loss_clip": 1.02372503,
      "balance_loss_mlp": 0.9996959,
      "epoch": 0.3539950700414838,
      "flos": 65588579498880.0,
      "grad_norm": 0.6984994446207212,
      "language_loss": 0.57756382,
      "learning_rate": 2.9963999291663347e-06,
      "loss": 0.59825736,
      "num_input_tokens_seen": 63292760,
      "step": 2944,
      "time_per_iteration": 3.091273784637451
    },
    {
      "auxiliary_loss_clip": 0.01138614,
      "auxiliary_loss_mlp": 0.01031943,
      "balance_loss_clip": 1.0551548,
      "balance_loss_mlp": 1.02364039,
      "epoch": 0.3541153129321229,
      "flos": 20521655919360.0,
      "grad_norm": 2.0815134501411916,
      "language_loss": 0.73934811,
      "learning_rate": 2.9957244367650915e-06,
      "loss": 0.76105368,
      "num_input_tokens_seen": 63309005,
      "step": 2945,
      "time_per_iteration": 2.538546323776245
    },
    {
      "auxiliary_loss_clip": 0.01131511,
      "auxiliary_loss_mlp": 0.01029031,
      "balance_loss_clip": 1.05332553,
      "balance_loss_mlp": 1.0197742,
      "epoch": 0.354235555822762,
      "flos": 19573578391680.0,
      "grad_norm": 1.7673821560908987,
      "language_loss": 0.83683157,
      "learning_rate": 2.9950487933132425e-06,
      "loss": 0.85843694,
      "num_input_tokens_seen": 63326420,
      "step": 2946,
      "time_per_iteration": 2.5497591495513916
    },
    {
      "auxiliary_loss_clip": 0.01180689,
      "auxiliary_loss_mlp": 0.01030523,
      "balance_loss_clip": 1.05506551,
      "balance_loss_mlp": 1.02214837,
      "epoch": 0.35435579871340106,
      "flos": 20777268078720.0,
      "grad_norm": 2.542723600213033,
      "language_loss": 0.71537042,
      "learning_rate": 2.994372998913283e-06,
      "loss": 0.73748249,
      "num_input_tokens_seen": 63344925,
      "step": 2947,
      "time_per_iteration": 2.4698736667633057
    },
    {
      "auxiliary_loss_clip": 0.01165576,
      "auxiliary_loss_mlp": 0.0103328,
      "balance_loss_clip": 1.05880082,
      "balance_loss_mlp": 1.02449429,
      "epoch": 0.35447604160404017,
      "flos": 23951807153280.0,
      "grad_norm": 2.848637827391769,
      "language_loss": 0.62397897,
      "learning_rate": 2.99369705366773e-06,
      "loss": 0.64596748,
      "num_input_tokens_seen": 63365170,
      "step": 2948,
      "time_per_iteration": 2.5240981578826904
    },
    {
      "auxiliary_loss_clip": 0.01161566,
      "auxiliary_loss_mlp": 0.01028654,
      "balance_loss_clip": 1.05478275,
      "balance_loss_mlp": 1.01985049,
      "epoch": 0.3545962844946792,
      "flos": 23435662671360.0,
      "grad_norm": 2.057421355231936,
      "language_loss": 0.82543683,
      "learning_rate": 2.9930209576791244e-06,
      "loss": 0.84733903,
      "num_input_tokens_seen": 63383645,
      "step": 2949,
      "time_per_iteration": 2.5266072750091553
    },
    {
      "auxiliary_loss_clip": 0.01176316,
      "auxiliary_loss_mlp": 0.01032345,
      "balance_loss_clip": 1.05739665,
      "balance_loss_mlp": 1.02411985,
      "epoch": 0.35471652738531834,
      "flos": 22085134185600.0,
      "grad_norm": 1.9203970493867377,
      "language_loss": 0.63537633,
      "learning_rate": 2.9923447110500285e-06,
      "loss": 0.65746289,
      "num_input_tokens_seen": 63402390,
      "step": 2950,
      "time_per_iteration": 3.2974531650543213
    },
    {
      "auxiliary_loss_clip": 0.01166821,
      "auxiliary_loss_mlp": 0.01037163,
      "balance_loss_clip": 1.05480599,
      "balance_loss_mlp": 1.02830553,
      "epoch": 0.35483677027595745,
      "flos": 27341881787520.0,
      "grad_norm": 2.0659287671097832,
      "language_loss": 0.75342083,
      "learning_rate": 2.9916683138830295e-06,
      "loss": 0.77546066,
      "num_input_tokens_seen": 63423055,
      "step": 2951,
      "time_per_iteration": 2.5069446563720703
    },
    {
      "auxiliary_loss_clip": 0.01161715,
      "auxiliary_loss_mlp": 0.01037036,
      "balance_loss_clip": 1.05561662,
      "balance_loss_mlp": 1.02758908,
      "epoch": 0.3549570131665965,
      "flos": 13516166678400.0,
      "grad_norm": 1.781025458989053,
      "language_loss": 0.80889213,
      "learning_rate": 2.9909917662807353e-06,
      "loss": 0.83087969,
      "num_input_tokens_seen": 63440855,
      "step": 2952,
      "time_per_iteration": 2.465482234954834
    },
    {
      "auxiliary_loss_clip": 0.01174228,
      "auxiliary_loss_mlp": 0.01039138,
      "balance_loss_clip": 1.05458212,
      "balance_loss_mlp": 1.03014016,
      "epoch": 0.3550772560572356,
      "flos": 20887549810560.0,
      "grad_norm": 2.237762260331911,
      "language_loss": 0.69040287,
      "learning_rate": 2.9903150683457783e-06,
      "loss": 0.71253651,
      "num_input_tokens_seen": 63459400,
      "step": 2953,
      "time_per_iteration": 2.4536101818084717
    },
    {
      "auxiliary_loss_clip": 0.01164352,
      "auxiliary_loss_mlp": 0.01030517,
      "balance_loss_clip": 1.05339575,
      "balance_loss_mlp": 1.02217865,
      "epoch": 0.3551974989478747,
      "flos": 20194042947840.0,
      "grad_norm": 1.8619957723256158,
      "language_loss": 0.64979893,
      "learning_rate": 2.9896382201808126e-06,
      "loss": 0.67174768,
      "num_input_tokens_seen": 63476800,
      "step": 2954,
      "time_per_iteration": 2.5076870918273926
    },
    {
      "auxiliary_loss_clip": 0.01192801,
      "auxiliary_loss_mlp": 0.01031973,
      "balance_loss_clip": 1.05734336,
      "balance_loss_mlp": 1.02325916,
      "epoch": 0.3553177418385138,
      "flos": 19828831415040.0,
      "grad_norm": 2.3509888501696268,
      "language_loss": 0.81204355,
      "learning_rate": 2.988961221888516e-06,
      "loss": 0.83429134,
      "num_input_tokens_seen": 63493475,
      "step": 2955,
      "time_per_iteration": 4.845149993896484
    },
    {
      "auxiliary_loss_clip": 0.01137827,
      "auxiliary_loss_mlp": 0.01028094,
      "balance_loss_clip": 1.04986143,
      "balance_loss_mlp": 1.01952302,
      "epoch": 0.3554379847291529,
      "flos": 14829132516480.0,
      "grad_norm": 3.1613313026581826,
      "language_loss": 0.78882468,
      "learning_rate": 2.988284073571589e-06,
      "loss": 0.81048393,
      "num_input_tokens_seen": 63509560,
      "step": 2956,
      "time_per_iteration": 2.496170997619629
    },
    {
      "auxiliary_loss_clip": 0.01180597,
      "auxiliary_loss_mlp": 0.00763554,
      "balance_loss_clip": 1.05845833,
      "balance_loss_mlp": 1.00084758,
      "epoch": 0.355558227619792,
      "flos": 20485350247680.0,
      "grad_norm": 2.170774794026298,
      "language_loss": 0.72462809,
      "learning_rate": 2.9876067753327528e-06,
      "loss": 0.74406964,
      "num_input_tokens_seen": 63527290,
      "step": 2957,
      "time_per_iteration": 2.477116346359253
    },
    {
      "auxiliary_loss_clip": 0.0118025,
      "auxiliary_loss_mlp": 0.01040496,
      "balance_loss_clip": 1.05509901,
      "balance_loss_mlp": 1.03102446,
      "epoch": 0.35567847051043106,
      "flos": 37663613256960.0,
      "grad_norm": 1.9891017048844266,
      "language_loss": 0.80440634,
      "learning_rate": 2.986929327274754e-06,
      "loss": 0.82661378,
      "num_input_tokens_seen": 63547870,
      "step": 2958,
      "time_per_iteration": 2.586879014968872
    },
    {
      "auxiliary_loss_clip": 0.01179266,
      "auxiliary_loss_mlp": 0.01033146,
      "balance_loss_clip": 1.0598433,
      "balance_loss_mlp": 1.02468848,
      "epoch": 0.35579871340107017,
      "flos": 26943058103040.0,
      "grad_norm": 1.682022100539393,
      "language_loss": 0.7870841,
      "learning_rate": 2.9862517295003617e-06,
      "loss": 0.80920827,
      "num_input_tokens_seen": 63568285,
      "step": 2959,
      "time_per_iteration": 2.517982244491577
    },
    {
      "auxiliary_loss_clip": 0.01143212,
      "auxiliary_loss_mlp": 0.01029474,
      "balance_loss_clip": 1.04818618,
      "balance_loss_mlp": 1.02129018,
      "epoch": 0.3559189562917093,
      "flos": 28293335193600.0,
      "grad_norm": 1.6866107263385866,
      "language_loss": 0.7256726,
      "learning_rate": 2.9855739821123654e-06,
      "loss": 0.74739945,
      "num_input_tokens_seen": 63589865,
      "step": 2960,
      "time_per_iteration": 2.5814740657806396
    },
    {
      "auxiliary_loss_clip": 0.01172072,
      "auxiliary_loss_mlp": 0.0102601,
      "balance_loss_clip": 1.05543816,
      "balance_loss_mlp": 1.01732564,
      "epoch": 0.35603919918234833,
      "flos": 25664063552640.0,
      "grad_norm": 1.795013740185745,
      "language_loss": 0.81711829,
      "learning_rate": 2.98489608521358e-06,
      "loss": 0.83909911,
      "num_input_tokens_seen": 63609805,
      "step": 2961,
      "time_per_iteration": 2.533064365386963
    },
    {
      "auxiliary_loss_clip": 0.01181756,
      "auxiliary_loss_mlp": 0.00763806,
      "balance_loss_clip": 1.05550539,
      "balance_loss_mlp": 1.00084913,
      "epoch": 0.35615944207298744,
      "flos": 23000856537600.0,
      "grad_norm": 1.914961886565478,
      "language_loss": 0.79416561,
      "learning_rate": 2.9842180389068425e-06,
      "loss": 0.81362116,
      "num_input_tokens_seen": 63627115,
      "step": 2962,
      "time_per_iteration": 2.484043598175049
    },
    {
      "auxiliary_loss_clip": 0.01058521,
      "auxiliary_loss_mlp": 0.01005315,
      "balance_loss_clip": 1.02967417,
      "balance_loss_mlp": 1.00359869,
      "epoch": 0.35627968496362655,
      "flos": 68251283723520.0,
      "grad_norm": 0.7663418270421275,
      "language_loss": 0.5926497,
      "learning_rate": 2.98353984329501e-06,
      "loss": 0.61328804,
      "num_input_tokens_seen": 63691460,
      "step": 2963,
      "time_per_iteration": 3.149522542953491
    },
    {
      "auxiliary_loss_clip": 0.01163954,
      "auxiliary_loss_mlp": 0.01029451,
      "balance_loss_clip": 1.05596471,
      "balance_loss_mlp": 1.02062416,
      "epoch": 0.3563999278542656,
      "flos": 22641714403200.0,
      "grad_norm": 1.728939880323328,
      "language_loss": 0.70818675,
      "learning_rate": 2.982861498480965e-06,
      "loss": 0.7301209,
      "num_input_tokens_seen": 63713840,
      "step": 2964,
      "time_per_iteration": 2.5470314025878906
    },
    {
      "auxiliary_loss_clip": 0.01143081,
      "auxiliary_loss_mlp": 0.0103367,
      "balance_loss_clip": 1.04901266,
      "balance_loss_mlp": 1.02518821,
      "epoch": 0.3565201707449047,
      "flos": 25952533678080.0,
      "grad_norm": 1.6449619186871334,
      "language_loss": 0.82461715,
      "learning_rate": 2.9821830045676122e-06,
      "loss": 0.84638464,
      "num_input_tokens_seen": 63733540,
      "step": 2965,
      "time_per_iteration": 2.5800325870513916
    },
    {
      "auxiliary_loss_clip": 0.01193302,
      "auxiliary_loss_mlp": 0.01032377,
      "balance_loss_clip": 1.05826831,
      "balance_loss_mlp": 1.02447963,
      "epoch": 0.3566404136355438,
      "flos": 28475725478400.0,
      "grad_norm": 1.9779485650648483,
      "language_loss": 0.72989655,
      "learning_rate": 2.9815043616578793e-06,
      "loss": 0.7521534,
      "num_input_tokens_seen": 63754335,
      "step": 2966,
      "time_per_iteration": 2.4848713874816895
    },
    {
      "auxiliary_loss_clip": 0.01143199,
      "auxiliary_loss_mlp": 0.0103879,
      "balance_loss_clip": 1.04881489,
      "balance_loss_mlp": 1.03006339,
      "epoch": 0.3567606565261829,
      "flos": 38363117690880.0,
      "grad_norm": 2.0899447486052254,
      "language_loss": 0.77082103,
      "learning_rate": 2.9808255698547145e-06,
      "loss": 0.79264092,
      "num_input_tokens_seen": 63777135,
      "step": 2967,
      "time_per_iteration": 2.7020087242126465
    },
    {
      "auxiliary_loss_clip": 0.0117731,
      "auxiliary_loss_mlp": 0.01029971,
      "balance_loss_clip": 1.05788755,
      "balance_loss_mlp": 1.02135789,
      "epoch": 0.356880899416822,
      "flos": 21981029592960.0,
      "grad_norm": 2.0127120577905826,
      "language_loss": 0.79628217,
      "learning_rate": 2.9801466292610913e-06,
      "loss": 0.81835496,
      "num_input_tokens_seen": 63797020,
      "step": 2968,
      "time_per_iteration": 2.4610321521759033
    },
    {
      "auxiliary_loss_clip": 0.01174793,
      "auxiliary_loss_mlp": 0.0102627,
      "balance_loss_clip": 1.05442548,
      "balance_loss_mlp": 1.01826167,
      "epoch": 0.35700114230746105,
      "flos": 18989132198400.0,
      "grad_norm": 2.046760480636488,
      "language_loss": 0.8116557,
      "learning_rate": 2.979467539980003e-06,
      "loss": 0.83366632,
      "num_input_tokens_seen": 63813810,
      "step": 2969,
      "time_per_iteration": 2.444014310836792
    },
    {
      "auxiliary_loss_clip": 0.01179484,
      "auxiliary_loss_mlp": 0.01036124,
      "balance_loss_clip": 1.05624914,
      "balance_loss_mlp": 1.0277735,
      "epoch": 0.35712138519810016,
      "flos": 19756112330880.0,
      "grad_norm": 2.07334719715132,
      "language_loss": 0.7704581,
      "learning_rate": 2.978788302114468e-06,
      "loss": 0.79261422,
      "num_input_tokens_seen": 63830925,
      "step": 2970,
      "time_per_iteration": 2.4472720623016357
    },
    {
      "auxiliary_loss_clip": 0.01174011,
      "auxiliary_loss_mlp": 0.0103306,
      "balance_loss_clip": 1.05488682,
      "balance_loss_mlp": 1.02411962,
      "epoch": 0.35724162808873927,
      "flos": 35183012008320.0,
      "grad_norm": 1.956911213403874,
      "language_loss": 0.83020657,
      "learning_rate": 2.9781089157675255e-06,
      "loss": 0.85227728,
      "num_input_tokens_seen": 63849385,
      "step": 2971,
      "time_per_iteration": 2.5607309341430664
    },
    {
      "auxiliary_loss_clip": 0.01172844,
      "auxiliary_loss_mlp": 0.010338,
      "balance_loss_clip": 1.05708694,
      "balance_loss_mlp": 1.02533662,
      "epoch": 0.3573618709793783,
      "flos": 25556726736000.0,
      "grad_norm": 1.4080445607258598,
      "language_loss": 0.88316917,
      "learning_rate": 2.977429381042238e-06,
      "loss": 0.90523559,
      "num_input_tokens_seen": 63870060,
      "step": 2972,
      "time_per_iteration": 2.490654468536377
    },
    {
      "auxiliary_loss_clip": 0.01162222,
      "auxiliary_loss_mlp": 0.01027474,
      "balance_loss_clip": 1.05405188,
      "balance_loss_mlp": 1.01970732,
      "epoch": 0.35748211387001744,
      "flos": 29132352051840.0,
      "grad_norm": 2.1795992832952686,
      "language_loss": 0.88873875,
      "learning_rate": 2.9767496980416913e-06,
      "loss": 0.91063571,
      "num_input_tokens_seen": 63889355,
      "step": 2973,
      "time_per_iteration": 2.5249698162078857
    },
    {
      "auxiliary_loss_clip": 0.0115881,
      "auxiliary_loss_mlp": 0.01033317,
      "balance_loss_clip": 1.05237639,
      "balance_loss_mlp": 1.02408481,
      "epoch": 0.35760235676065655,
      "flos": 13954169122560.0,
      "grad_norm": 3.159873043654318,
      "language_loss": 0.80567002,
      "learning_rate": 2.9760698668689914e-06,
      "loss": 0.82759136,
      "num_input_tokens_seen": 63905580,
      "step": 2974,
      "time_per_iteration": 2.4530551433563232
    },
    {
      "auxiliary_loss_clip": 0.0117614,
      "auxiliary_loss_mlp": 0.01025103,
      "balance_loss_clip": 1.05403423,
      "balance_loss_mlp": 1.0172708,
      "epoch": 0.3577225996512956,
      "flos": 44018688977280.0,
      "grad_norm": 1.8480005907925756,
      "language_loss": 0.71325302,
      "learning_rate": 2.975389887627269e-06,
      "loss": 0.73526549,
      "num_input_tokens_seen": 63928180,
      "step": 2975,
      "time_per_iteration": 2.646256685256958
    },
    {
      "auxiliary_loss_clip": 0.01152168,
      "auxiliary_loss_mlp": 0.01033488,
      "balance_loss_clip": 1.05320835,
      "balance_loss_mlp": 1.02555513,
      "epoch": 0.3578428425419347,
      "flos": 17055199013760.0,
      "grad_norm": 2.228301652818403,
      "language_loss": 0.90113956,
      "learning_rate": 2.9747097604196764e-06,
      "loss": 0.9229961,
      "num_input_tokens_seen": 63944825,
      "step": 2976,
      "time_per_iteration": 3.3176894187927246
    },
    {
      "auxiliary_loss_clip": 0.01048579,
      "auxiliary_loss_mlp": 0.01002636,
      "balance_loss_clip": 1.02354014,
      "balance_loss_mlp": 1.00149727,
      "epoch": 0.3579630854325738,
      "flos": 71676550707840.0,
      "grad_norm": 0.6782342843040509,
      "language_loss": 0.56674582,
      "learning_rate": 2.9740294853493875e-06,
      "loss": 0.58725792,
      "num_input_tokens_seen": 64016385,
      "step": 2977,
      "time_per_iteration": 3.346116781234741
    },
    {
      "auxiliary_loss_clip": 0.01137286,
      "auxiliary_loss_mlp": 0.0102882,
      "balance_loss_clip": 1.04993582,
      "balance_loss_mlp": 1.02073145,
      "epoch": 0.3580833283232129,
      "flos": 25046651652480.0,
      "grad_norm": 2.0318924444903796,
      "language_loss": 0.67117083,
      "learning_rate": 2.9733490625196008e-06,
      "loss": 0.69283187,
      "num_input_tokens_seen": 64036245,
      "step": 2978,
      "time_per_iteration": 2.5833258628845215
    },
    {
      "auxiliary_loss_clip": 0.01134314,
      "auxiliary_loss_mlp": 0.01030004,
      "balance_loss_clip": 1.04870224,
      "balance_loss_mlp": 1.02261877,
      "epoch": 0.358203571213852,
      "flos": 13953127628160.0,
      "grad_norm": 3.068179925146427,
      "language_loss": 0.75730401,
      "learning_rate": 2.9726684920335353e-06,
      "loss": 0.77894723,
      "num_input_tokens_seen": 64054110,
      "step": 2979,
      "time_per_iteration": 2.497767210006714
    },
    {
      "auxiliary_loss_clip": 0.01192178,
      "auxiliary_loss_mlp": 0.00763745,
      "balance_loss_clip": 1.05593419,
      "balance_loss_mlp": 1.00068951,
      "epoch": 0.35832381410449105,
      "flos": 20302457172480.0,
      "grad_norm": 2.2464413454239467,
      "language_loss": 0.81978083,
      "learning_rate": 2.971987773994432e-06,
      "loss": 0.83934009,
      "num_input_tokens_seen": 64070295,
      "step": 2980,
      "time_per_iteration": 2.4605650901794434
    },
    {
      "auxiliary_loss_clip": 0.0116609,
      "auxiliary_loss_mlp": 0.01024595,
      "balance_loss_clip": 1.05126548,
      "balance_loss_mlp": 1.0166198,
      "epoch": 0.35844405699513016,
      "flos": 16983234115200.0,
      "grad_norm": 2.0029866291361884,
      "language_loss": 0.83118755,
      "learning_rate": 2.9713069085055566e-06,
      "loss": 0.8530944,
      "num_input_tokens_seen": 64088605,
      "step": 2981,
      "time_per_iteration": 2.476473331451416
    },
    {
      "auxiliary_loss_clip": 0.01146202,
      "auxiliary_loss_mlp": 0.01025839,
      "balance_loss_clip": 1.0511632,
      "balance_loss_mlp": 1.01742864,
      "epoch": 0.35856429988576927,
      "flos": 23216858974080.0,
      "grad_norm": 1.9629050954993847,
      "language_loss": 0.78956056,
      "learning_rate": 2.9706258956701958e-06,
      "loss": 0.81128097,
      "num_input_tokens_seen": 64108595,
      "step": 2982,
      "time_per_iteration": 4.155234098434448
    },
    {
      "auxiliary_loss_clip": 0.01177914,
      "auxiliary_loss_mlp": 0.01030959,
      "balance_loss_clip": 1.05516458,
      "balance_loss_mlp": 1.02272201,
      "epoch": 0.3586845427764083,
      "flos": 23034576430080.0,
      "grad_norm": 2.6952859822389312,
      "language_loss": 0.77377182,
      "learning_rate": 2.9699447355916575e-06,
      "loss": 0.79586053,
      "num_input_tokens_seen": 64127405,
      "step": 2983,
      "time_per_iteration": 2.6154489517211914
    },
    {
      "auxiliary_loss_clip": 0.01188941,
      "auxiliary_loss_mlp": 0.00763004,
      "balance_loss_clip": 1.05645323,
      "balance_loss_mlp": 1.00072801,
      "epoch": 0.35880478566704743,
      "flos": 20010682995840.0,
      "grad_norm": 2.0530629640322355,
      "language_loss": 0.74110591,
      "learning_rate": 2.969263428373275e-06,
      "loss": 0.76062536,
      "num_input_tokens_seen": 64145755,
      "step": 2984,
      "time_per_iteration": 2.435987710952759
    },
    {
      "auxiliary_loss_clip": 0.01162489,
      "auxiliary_loss_mlp": 0.01032509,
      "balance_loss_clip": 1.05279374,
      "balance_loss_mlp": 1.02428913,
      "epoch": 0.35892502855768654,
      "flos": 13699095667200.0,
      "grad_norm": 1.996497047089602,
      "language_loss": 0.78986418,
      "learning_rate": 2.9685819741184007e-06,
      "loss": 0.81181413,
      "num_input_tokens_seen": 64164195,
      "step": 2985,
      "time_per_iteration": 2.4754245281219482
    },
    {
      "auxiliary_loss_clip": 0.01140273,
      "auxiliary_loss_mlp": 0.01032343,
      "balance_loss_clip": 1.05050433,
      "balance_loss_mlp": 1.02383435,
      "epoch": 0.3590452714483256,
      "flos": 18114096977280.0,
      "grad_norm": 6.0183160916630785,
      "language_loss": 0.68161905,
      "learning_rate": 2.967900372930411e-06,
      "loss": 0.70334518,
      "num_input_tokens_seen": 64182705,
      "step": 2986,
      "time_per_iteration": 2.6570496559143066
    },
    {
      "auxiliary_loss_clip": 0.01154199,
      "auxiliary_loss_mlp": 0.01032721,
      "balance_loss_clip": 1.05138612,
      "balance_loss_mlp": 1.02426291,
      "epoch": 0.3591655143389647,
      "flos": 17749352321280.0,
      "grad_norm": 3.9236801382411315,
      "language_loss": 0.79233557,
      "learning_rate": 2.9672186249127046e-06,
      "loss": 0.81420475,
      "num_input_tokens_seen": 64202170,
      "step": 2987,
      "time_per_iteration": 2.469970703125
    },
    {
      "auxiliary_loss_clip": 0.01160333,
      "auxiliary_loss_mlp": 0.01036119,
      "balance_loss_clip": 1.05470395,
      "balance_loss_mlp": 1.02813792,
      "epoch": 0.3592857572296038,
      "flos": 25224409082880.0,
      "grad_norm": 1.9482356707264374,
      "language_loss": 0.78970307,
      "learning_rate": 2.9665367301687014e-06,
      "loss": 0.81166756,
      "num_input_tokens_seen": 64220415,
      "step": 2988,
      "time_per_iteration": 2.5062496662139893
    },
    {
      "auxiliary_loss_clip": 0.0115295,
      "auxiliary_loss_mlp": 0.01029108,
      "balance_loss_clip": 1.05173755,
      "balance_loss_mlp": 1.0210495,
      "epoch": 0.3594060001202429,
      "flos": 29384408764800.0,
      "grad_norm": 2.1563760533526044,
      "language_loss": 0.76961029,
      "learning_rate": 2.965854688801845e-06,
      "loss": 0.79143089,
      "num_input_tokens_seen": 64242475,
      "step": 2989,
      "time_per_iteration": 2.5602898597717285
    },
    {
      "auxiliary_loss_clip": 0.01170718,
      "auxiliary_loss_mlp": 0.0102655,
      "balance_loss_clip": 1.04943419,
      "balance_loss_mlp": 1.01852095,
      "epoch": 0.359526243010882,
      "flos": 17052900543360.0,
      "grad_norm": 1.925616359896462,
      "language_loss": 0.76317143,
      "learning_rate": 2.9651725009156005e-06,
      "loss": 0.78514421,
      "num_input_tokens_seen": 64260220,
      "step": 2990,
      "time_per_iteration": 2.4543933868408203
    },
    {
      "auxiliary_loss_clip": 0.0115328,
      "auxiliary_loss_mlp": 0.01032245,
      "balance_loss_clip": 1.04994416,
      "balance_loss_mlp": 1.02341127,
      "epoch": 0.3596464859015211,
      "flos": 22965089569920.0,
      "grad_norm": 1.5976815336725965,
      "language_loss": 0.74403501,
      "learning_rate": 2.964490166613454e-06,
      "loss": 0.76589024,
      "num_input_tokens_seen": 64280145,
      "step": 2991,
      "time_per_iteration": 2.511255979537964
    },
    {
      "auxiliary_loss_clip": 0.0108966,
      "auxiliary_loss_mlp": 0.01000824,
      "balance_loss_clip": 1.02266264,
      "balance_loss_mlp": 0.99978685,
      "epoch": 0.35976672879216015,
      "flos": 54739462590720.0,
      "grad_norm": 0.7590962906847478,
      "language_loss": 0.57717937,
      "learning_rate": 2.963807685998917e-06,
      "loss": 0.59808421,
      "num_input_tokens_seen": 64336010,
      "step": 2992,
      "time_per_iteration": 2.8447141647338867
    },
    {
      "auxiliary_loss_clip": 0.01135906,
      "auxiliary_loss_mlp": 0.01029592,
      "balance_loss_clip": 1.04918718,
      "balance_loss_mlp": 1.02121711,
      "epoch": 0.35988697168279926,
      "flos": 43139020901760.0,
      "grad_norm": 1.469671895656916,
      "language_loss": 0.77936363,
      "learning_rate": 2.9631250591755196e-06,
      "loss": 0.8010186,
      "num_input_tokens_seen": 64358725,
      "step": 2993,
      "time_per_iteration": 2.7509398460388184
    },
    {
      "auxiliary_loss_clip": 0.01156387,
      "auxiliary_loss_mlp": 0.01031661,
      "balance_loss_clip": 1.05406249,
      "balance_loss_mlp": 1.02260745,
      "epoch": 0.36000721457343837,
      "flos": 35845600239360.0,
      "grad_norm": 1.6352496961306806,
      "language_loss": 0.57549322,
      "learning_rate": 2.962442286246817e-06,
      "loss": 0.59737372,
      "num_input_tokens_seen": 64381555,
      "step": 2994,
      "time_per_iteration": 2.613842725753784
    },
    {
      "auxiliary_loss_clip": 0.01164761,
      "auxiliary_loss_mlp": 0.01029919,
      "balance_loss_clip": 1.05320179,
      "balance_loss_mlp": 1.02185488,
      "epoch": 0.3601274574640774,
      "flos": 18291100222080.0,
      "grad_norm": 2.1585049943428656,
      "language_loss": 0.69675332,
      "learning_rate": 2.9617593673163853e-06,
      "loss": 0.71870017,
      "num_input_tokens_seen": 64400375,
      "step": 2995,
      "time_per_iteration": 2.481973648071289
    },
    {
      "auxiliary_loss_clip": 0.01162208,
      "auxiliary_loss_mlp": 0.01024018,
      "balance_loss_clip": 1.0502317,
      "balance_loss_mlp": 1.01640046,
      "epoch": 0.36024770035471654,
      "flos": 13333955961600.0,
      "grad_norm": 2.2758988078744617,
      "language_loss": 0.77393389,
      "learning_rate": 2.9610763024878216e-06,
      "loss": 0.79579616,
      "num_input_tokens_seen": 64415880,
      "step": 2996,
      "time_per_iteration": 2.4539296627044678
    },
    {
      "auxiliary_loss_clip": 0.01153675,
      "auxiliary_loss_mlp": 0.01035445,
      "balance_loss_clip": 1.05115676,
      "balance_loss_mlp": 1.02703476,
      "epoch": 0.3603679432453556,
      "flos": 20267013427200.0,
      "grad_norm": 1.6758543819927192,
      "language_loss": 0.91451299,
      "learning_rate": 2.960393091864747e-06,
      "loss": 0.93640423,
      "num_input_tokens_seen": 64434260,
      "step": 2997,
      "time_per_iteration": 2.489513635635376
    },
    {
      "auxiliary_loss_clip": 0.01162383,
      "auxiliary_loss_mlp": 0.01024365,
      "balance_loss_clip": 1.05561626,
      "balance_loss_mlp": 1.01630664,
      "epoch": 0.3604881861359947,
      "flos": 22451135817600.0,
      "grad_norm": 1.8164362400208431,
      "language_loss": 0.74771369,
      "learning_rate": 2.959709735550804e-06,
      "loss": 0.76958114,
      "num_input_tokens_seen": 64453855,
      "step": 2998,
      "time_per_iteration": 2.49832820892334
    },
    {
      "auxiliary_loss_clip": 0.01133679,
      "auxiliary_loss_mlp": 0.01025359,
      "balance_loss_clip": 1.04901218,
      "balance_loss_mlp": 1.01730657,
      "epoch": 0.3606084290266338,
      "flos": 22054251467520.0,
      "grad_norm": 2.3783355842547937,
      "language_loss": 0.76083314,
      "learning_rate": 2.9590262336496575e-06,
      "loss": 0.78242362,
      "num_input_tokens_seen": 64473585,
      "step": 2999,
      "time_per_iteration": 2.5815391540527344
    },
    {
      "auxiliary_loss_clip": 0.01142755,
      "auxiliary_loss_mlp": 0.01034449,
      "balance_loss_clip": 1.05298042,
      "balance_loss_mlp": 1.02572334,
      "epoch": 0.36072867191727287,
      "flos": 15632921111040.0,
      "grad_norm": 2.0668966412379697,
      "language_loss": 0.85315132,
      "learning_rate": 2.9583425862649936e-06,
      "loss": 0.87492335,
      "num_input_tokens_seen": 64491720,
      "step": 3000,
      "time_per_iteration": 2.5043280124664307
    },
    {
      "auxiliary_loss_clip": 0.01190005,
      "auxiliary_loss_mlp": 0.01028325,
      "balance_loss_clip": 1.05618858,
      "balance_loss_mlp": 1.02005816,
      "epoch": 0.360848914807912,
      "flos": 19677000625920.0,
      "grad_norm": 1.9197797231320495,
      "language_loss": 0.73385042,
      "learning_rate": 2.9576587935005215e-06,
      "loss": 0.75603372,
      "num_input_tokens_seen": 64509800,
      "step": 3001,
      "time_per_iteration": 2.433746337890625
    },
    {
      "auxiliary_loss_clip": 0.01175934,
      "auxiliary_loss_mlp": 0.01028569,
      "balance_loss_clip": 1.05320883,
      "balance_loss_mlp": 1.02019477,
      "epoch": 0.3609691576985511,
      "flos": 18877808972160.0,
      "grad_norm": 2.2343689547062087,
      "language_loss": 0.71778381,
      "learning_rate": 2.9569748554599713e-06,
      "loss": 0.73982882,
      "num_input_tokens_seen": 64525410,
      "step": 3002,
      "time_per_iteration": 2.4360578060150146
    },
    {
      "auxiliary_loss_clip": 0.01161677,
      "auxiliary_loss_mlp": 0.01032589,
      "balance_loss_clip": 1.05496955,
      "balance_loss_mlp": 1.02442932,
      "epoch": 0.36108940058919015,
      "flos": 42224088648960.0,
      "grad_norm": 4.244675395062702,
      "language_loss": 0.73288316,
      "learning_rate": 2.956290772247097e-06,
      "loss": 0.75482583,
      "num_input_tokens_seen": 64544085,
      "step": 3003,
      "time_per_iteration": 3.464754343032837
    },
    {
      "auxiliary_loss_clip": 0.01124095,
      "auxiliary_loss_mlp": 0.01033909,
      "balance_loss_clip": 1.04968941,
      "balance_loss_mlp": 1.0259521,
      "epoch": 0.36120964347982926,
      "flos": 23185150243200.0,
      "grad_norm": 1.6944738342495917,
      "language_loss": 0.73045415,
      "learning_rate": 2.9556065439656724e-06,
      "loss": 0.75203419,
      "num_input_tokens_seen": 64563135,
      "step": 3004,
      "time_per_iteration": 2.5628418922424316
    },
    {
      "auxiliary_loss_clip": 0.01110862,
      "auxiliary_loss_mlp": 0.01029254,
      "balance_loss_clip": 1.04467463,
      "balance_loss_mlp": 1.02067125,
      "epoch": 0.36132988637046837,
      "flos": 18113055482880.0,
      "grad_norm": 2.3794617465351733,
      "language_loss": 0.81456739,
      "learning_rate": 2.9549221707194952e-06,
      "loss": 0.83596849,
      "num_input_tokens_seen": 64581985,
      "step": 3005,
      "time_per_iteration": 2.602739095687866
    },
    {
      "auxiliary_loss_clip": 0.01176126,
      "auxiliary_loss_mlp": 0.01025335,
      "balance_loss_clip": 1.05500448,
      "balance_loss_mlp": 1.01719332,
      "epoch": 0.3614501292611074,
      "flos": 27813101333760.0,
      "grad_norm": 1.8741744879628925,
      "language_loss": 0.72265118,
      "learning_rate": 2.954237652612384e-06,
      "loss": 0.74466574,
      "num_input_tokens_seen": 64601035,
      "step": 3006,
      "time_per_iteration": 2.5227699279785156
    },
    {
      "auxiliary_loss_clip": 0.01155368,
      "auxiliary_loss_mlp": 0.01027096,
      "balance_loss_clip": 1.05020416,
      "balance_loss_mlp": 1.01942468,
      "epoch": 0.36157037215174653,
      "flos": 22634926732800.0,
      "grad_norm": 1.8622469017009153,
      "language_loss": 0.84637547,
      "learning_rate": 2.9535529897481796e-06,
      "loss": 0.86820012,
      "num_input_tokens_seen": 64618580,
      "step": 3007,
      "time_per_iteration": 2.5105478763580322
    },
    {
      "auxiliary_loss_clip": 0.01190097,
      "auxiliary_loss_mlp": 0.0102716,
      "balance_loss_clip": 1.05491745,
      "balance_loss_mlp": 1.01880932,
      "epoch": 0.36169061504238564,
      "flos": 12600839376000.0,
      "grad_norm": 2.3578117338497746,
      "language_loss": 0.76721716,
      "learning_rate": 2.9528681822307446e-06,
      "loss": 0.78938973,
      "num_input_tokens_seen": 64635430,
      "step": 3008,
      "time_per_iteration": 3.230964422225952
    },
    {
      "auxiliary_loss_clip": 0.01172885,
      "auxiliary_loss_mlp": 0.00762667,
      "balance_loss_clip": 1.05761623,
      "balance_loss_mlp": 1.00062132,
      "epoch": 0.3618108579330247,
      "flos": 26684644682880.0,
      "grad_norm": 2.1050139023473884,
      "language_loss": 0.8221277,
      "learning_rate": 2.952183230163964e-06,
      "loss": 0.84148324,
      "num_input_tokens_seen": 64655005,
      "step": 3009,
      "time_per_iteration": 4.012263774871826
    },
    {
      "auxiliary_loss_clip": 0.01142407,
      "auxiliary_loss_mlp": 0.01025574,
      "balance_loss_clip": 1.05066669,
      "balance_loss_mlp": 1.01778948,
      "epoch": 0.3619311008236638,
      "flos": 22817029708800.0,
      "grad_norm": 1.8362056113763947,
      "language_loss": 0.72998989,
      "learning_rate": 2.9514981336517448e-06,
      "loss": 0.75166965,
      "num_input_tokens_seen": 64674775,
      "step": 3010,
      "time_per_iteration": 2.5612552165985107
    },
    {
      "auxiliary_loss_clip": 0.01174296,
      "auxiliary_loss_mlp": 0.01031634,
      "balance_loss_clip": 1.05639648,
      "balance_loss_mlp": 1.02302694,
      "epoch": 0.36205134371430286,
      "flos": 25919603884800.0,
      "grad_norm": 2.1586767954913344,
      "language_loss": 0.81476468,
      "learning_rate": 2.950812892798015e-06,
      "loss": 0.83682394,
      "num_input_tokens_seen": 64695670,
      "step": 3011,
      "time_per_iteration": 2.4976279735565186
    },
    {
      "auxiliary_loss_clip": 0.01129894,
      "auxiliary_loss_mlp": 0.00763099,
      "balance_loss_clip": 1.0525856,
      "balance_loss_mlp": 1.00069559,
      "epoch": 0.362171586604942,
      "flos": 26139592730880.0,
      "grad_norm": 7.740300828705119,
      "language_loss": 0.87423897,
      "learning_rate": 2.9501275077067256e-06,
      "loss": 0.89316893,
      "num_input_tokens_seen": 64716290,
      "step": 3012,
      "time_per_iteration": 2.5927493572235107
    },
    {
      "auxiliary_loss_clip": 0.01101595,
      "auxiliary_loss_mlp": 0.01024346,
      "balance_loss_clip": 1.04460859,
      "balance_loss_mlp": 1.01646662,
      "epoch": 0.3622918294955811,
      "flos": 28074208273920.0,
      "grad_norm": 1.4297100715115676,
      "language_loss": 0.88562202,
      "learning_rate": 2.949441978481848e-06,
      "loss": 0.90688145,
      "num_input_tokens_seen": 64737190,
      "step": 3013,
      "time_per_iteration": 2.665468215942383
    },
    {
      "auxiliary_loss_clip": 0.01149607,
      "auxiliary_loss_mlp": 0.01028959,
      "balance_loss_clip": 1.04914284,
      "balance_loss_mlp": 1.02026808,
      "epoch": 0.36241207238622014,
      "flos": 19828005402240.0,
      "grad_norm": 2.4466537836984323,
      "language_loss": 0.79897869,
      "learning_rate": 2.9487563052273778e-06,
      "loss": 0.8207643,
      "num_input_tokens_seen": 64753950,
      "step": 3014,
      "time_per_iteration": 2.503666877746582
    },
    {
      "auxiliary_loss_clip": 0.01171805,
      "auxiliary_loss_mlp": 0.01031463,
      "balance_loss_clip": 1.05699706,
      "balance_loss_mlp": 1.02327895,
      "epoch": 0.36253231527685925,
      "flos": 21397158017280.0,
      "grad_norm": 1.8469445822922441,
      "language_loss": 0.85834885,
      "learning_rate": 2.94807048804733e-06,
      "loss": 0.88038158,
      "num_input_tokens_seen": 64773570,
      "step": 3015,
      "time_per_iteration": 2.452136278152466
    },
    {
      "auxiliary_loss_clip": 0.01148504,
      "auxiliary_loss_mlp": 0.01027574,
      "balance_loss_clip": 1.0476234,
      "balance_loss_mlp": 1.01873493,
      "epoch": 0.36265255816749836,
      "flos": 18362885552640.0,
      "grad_norm": 1.859501233370567,
      "language_loss": 0.89644897,
      "learning_rate": 2.9473845270457434e-06,
      "loss": 0.91820973,
      "num_input_tokens_seen": 64790385,
      "step": 3016,
      "time_per_iteration": 2.5093281269073486
    },
    {
      "auxiliary_loss_clip": 0.01151882,
      "auxiliary_loss_mlp": 0.01028777,
      "balance_loss_clip": 1.04930615,
      "balance_loss_mlp": 1.02036715,
      "epoch": 0.3627728010581374,
      "flos": 18660046769280.0,
      "grad_norm": 2.2555949725112887,
      "language_loss": 0.70363128,
      "learning_rate": 2.946698422326677e-06,
      "loss": 0.72543782,
      "num_input_tokens_seen": 64807845,
      "step": 3017,
      "time_per_iteration": 2.476022958755493
    },
    {
      "auxiliary_loss_clip": 0.01127007,
      "auxiliary_loss_mlp": 0.01028513,
      "balance_loss_clip": 1.04501915,
      "balance_loss_mlp": 1.02031136,
      "epoch": 0.36289304394877653,
      "flos": 27524272072320.0,
      "grad_norm": 2.2585039808337632,
      "language_loss": 0.79826194,
      "learning_rate": 2.946012173994213e-06,
      "loss": 0.81981713,
      "num_input_tokens_seen": 64827630,
      "step": 3018,
      "time_per_iteration": 2.61464262008667
    },
    {
      "auxiliary_loss_clip": 0.0116869,
      "auxiliary_loss_mlp": 0.01027502,
      "balance_loss_clip": 1.05535173,
      "balance_loss_mlp": 1.01940143,
      "epoch": 0.36301328683941564,
      "flos": 34533244932480.0,
      "grad_norm": 1.5639918027379245,
      "language_loss": 0.67593777,
      "learning_rate": 2.945325782152454e-06,
      "loss": 0.69789964,
      "num_input_tokens_seen": 64850665,
      "step": 3019,
      "time_per_iteration": 2.5781259536743164
    },
    {
      "auxiliary_loss_clip": 0.01159809,
      "auxiliary_loss_mlp": 0.01024673,
      "balance_loss_clip": 1.04933882,
      "balance_loss_mlp": 1.01688886,
      "epoch": 0.3631335297300547,
      "flos": 19025976574080.0,
      "grad_norm": 2.47884559230756,
      "language_loss": 0.78729206,
      "learning_rate": 2.9446392469055257e-06,
      "loss": 0.80913687,
      "num_input_tokens_seen": 64868700,
      "step": 3020,
      "time_per_iteration": 2.483788251876831
    },
    {
      "auxiliary_loss_clip": 0.01140368,
      "auxiliary_loss_mlp": 0.01028218,
      "balance_loss_clip": 1.0542922,
      "balance_loss_mlp": 1.02014184,
      "epoch": 0.3632537726206938,
      "flos": 19536769929600.0,
      "grad_norm": 1.7360396790473371,
      "language_loss": 0.79997796,
      "learning_rate": 2.9439525683575745e-06,
      "loss": 0.82166386,
      "num_input_tokens_seen": 64887620,
      "step": 3021,
      "time_per_iteration": 2.506786823272705
    },
    {
      "auxiliary_loss_clip": 0.011932,
      "auxiliary_loss_mlp": 0.0103072,
      "balance_loss_clip": 1.05840254,
      "balance_loss_mlp": 1.02205336,
      "epoch": 0.3633740155113329,
      "flos": 21068611292160.0,
      "grad_norm": 2.3537419886204636,
      "language_loss": 0.7490477,
      "learning_rate": 2.9432657466127694e-06,
      "loss": 0.77128685,
      "num_input_tokens_seen": 64907190,
      "step": 3022,
      "time_per_iteration": 2.429471254348755
    },
    {
      "auxiliary_loss_clip": 0.01133027,
      "auxiliary_loss_mlp": 0.01027058,
      "balance_loss_clip": 1.05306089,
      "balance_loss_mlp": 1.01889479,
      "epoch": 0.36349425840197197,
      "flos": 20298722158080.0,
      "grad_norm": 1.6351722895222163,
      "language_loss": 0.76420426,
      "learning_rate": 2.9425787817753007e-06,
      "loss": 0.78580511,
      "num_input_tokens_seen": 64925850,
      "step": 3023,
      "time_per_iteration": 2.567260503768921
    },
    {
      "auxiliary_loss_clip": 0.01147785,
      "auxiliary_loss_mlp": 0.01030199,
      "balance_loss_clip": 1.05193591,
      "balance_loss_mlp": 1.02221227,
      "epoch": 0.3636145012926111,
      "flos": 29716762331520.0,
      "grad_norm": 1.9424965757930548,
      "language_loss": 0.71516395,
      "learning_rate": 2.94189167394938e-06,
      "loss": 0.73694378,
      "num_input_tokens_seen": 64948285,
      "step": 3024,
      "time_per_iteration": 2.604113817214966
    },
    {
      "auxiliary_loss_clip": 0.01191209,
      "auxiliary_loss_mlp": 0.0103109,
      "balance_loss_clip": 1.05912459,
      "balance_loss_mlp": 1.02292967,
      "epoch": 0.3637347441832502,
      "flos": 21431847576960.0,
      "grad_norm": 1.8814627423968688,
      "language_loss": 0.81169844,
      "learning_rate": 2.941204423239241e-06,
      "loss": 0.83392143,
      "num_input_tokens_seen": 64967160,
      "step": 3025,
      "time_per_iteration": 2.430171251296997
    },
    {
      "auxiliary_loss_clip": 0.0117084,
      "auxiliary_loss_mlp": 0.01029747,
      "balance_loss_clip": 1.05368423,
      "balance_loss_mlp": 1.02141404,
      "epoch": 0.36385498707388925,
      "flos": 29533941083520.0,
      "grad_norm": 2.0726574161555837,
      "language_loss": 0.75780177,
      "learning_rate": 2.9405170297491395e-06,
      "loss": 0.77980763,
      "num_input_tokens_seen": 64987155,
      "step": 3026,
      "time_per_iteration": 2.5254337787628174
    },
    {
      "auxiliary_loss_clip": 0.01109924,
      "auxiliary_loss_mlp": 0.00763453,
      "balance_loss_clip": 1.05187631,
      "balance_loss_mlp": 1.00072742,
      "epoch": 0.36397522996452836,
      "flos": 22236569925120.0,
      "grad_norm": 2.6072112493235218,
      "language_loss": 0.80024898,
      "learning_rate": 2.939829493583353e-06,
      "loss": 0.81898272,
      "num_input_tokens_seen": 65003800,
      "step": 3027,
      "time_per_iteration": 2.594695806503296
    },
    {
      "auxiliary_loss_clip": 0.01136742,
      "auxiliary_loss_mlp": 0.01028336,
      "balance_loss_clip": 1.0462985,
      "balance_loss_mlp": 1.02015245,
      "epoch": 0.3640954728551674,
      "flos": 21506505995520.0,
      "grad_norm": 2.690110434807627,
      "language_loss": 0.83562315,
      "learning_rate": 2.939141814846179e-06,
      "loss": 0.85727394,
      "num_input_tokens_seen": 65021215,
      "step": 3028,
      "time_per_iteration": 2.5258214473724365
    },
    {
      "auxiliary_loss_clip": 0.01160532,
      "auxiliary_loss_mlp": 0.01025117,
      "balance_loss_clip": 1.05197728,
      "balance_loss_mlp": 1.01651645,
      "epoch": 0.3642157157458065,
      "flos": 17712867081600.0,
      "grad_norm": 1.655076761386125,
      "language_loss": 0.82297587,
      "learning_rate": 2.938453993641938e-06,
      "loss": 0.84483242,
      "num_input_tokens_seen": 65039590,
      "step": 3029,
      "time_per_iteration": 3.347519636154175
    },
    {
      "auxiliary_loss_clip": 0.01161037,
      "auxiliary_loss_mlp": 0.0103147,
      "balance_loss_clip": 1.05648255,
      "balance_loss_mlp": 1.02283382,
      "epoch": 0.36433595863644563,
      "flos": 17639537466240.0,
      "grad_norm": 2.0226674329179355,
      "language_loss": 0.70361483,
      "learning_rate": 2.937766030074973e-06,
      "loss": 0.72553992,
      "num_input_tokens_seen": 65056845,
      "step": 3030,
      "time_per_iteration": 2.472032308578491
    },
    {
      "auxiliary_loss_clip": 0.01151273,
      "auxiliary_loss_mlp": 0.01030014,
      "balance_loss_clip": 1.05215549,
      "balance_loss_mlp": 1.02192557,
      "epoch": 0.3644562015270847,
      "flos": 26833279161600.0,
      "grad_norm": 1.7956553124666847,
      "language_loss": 0.8272683,
      "learning_rate": 2.937077924249646e-06,
      "loss": 0.8490811,
      "num_input_tokens_seen": 65079435,
      "step": 3031,
      "time_per_iteration": 2.582489490509033
    },
    {
      "auxiliary_loss_clip": 0.01165928,
      "auxiliary_loss_mlp": 0.01027589,
      "balance_loss_clip": 1.05304193,
      "balance_loss_mlp": 1.01915455,
      "epoch": 0.3645764444177238,
      "flos": 14282715847680.0,
      "grad_norm": 2.079976720707225,
      "language_loss": 0.76207024,
      "learning_rate": 2.9363896762703443e-06,
      "loss": 0.7840054,
      "num_input_tokens_seen": 65096500,
      "step": 3032,
      "time_per_iteration": 2.459040880203247
    },
    {
      "auxiliary_loss_clip": 0.01189664,
      "auxiliary_loss_mlp": 0.01030266,
      "balance_loss_clip": 1.05624104,
      "balance_loss_mlp": 1.02109265,
      "epoch": 0.3646966873083629,
      "flos": 20667489137280.0,
      "grad_norm": 1.6881177954387239,
      "language_loss": 0.84259558,
      "learning_rate": 2.9357012862414725e-06,
      "loss": 0.86479485,
      "num_input_tokens_seen": 65115860,
      "step": 3033,
      "time_per_iteration": 2.466895818710327
    },
    {
      "auxiliary_loss_clip": 0.01174257,
      "auxiliary_loss_mlp": 0.01030996,
      "balance_loss_clip": 1.05515027,
      "balance_loss_mlp": 1.02252591,
      "epoch": 0.36481693019900197,
      "flos": 27782613665280.0,
      "grad_norm": 1.8057515910878763,
      "language_loss": 0.71681011,
      "learning_rate": 2.9350127542674593e-06,
      "loss": 0.73886263,
      "num_input_tokens_seen": 65138070,
      "step": 3034,
      "time_per_iteration": 2.518522262573242
    },
    {
      "auxiliary_loss_clip": 0.01168921,
      "auxiliary_loss_mlp": 0.01033914,
      "balance_loss_clip": 1.05580187,
      "balance_loss_mlp": 1.02570701,
      "epoch": 0.3649371730896411,
      "flos": 19712588025600.0,
      "grad_norm": 1.7788645567277077,
      "language_loss": 0.76612991,
      "learning_rate": 2.934324080452755e-06,
      "loss": 0.7881583,
      "num_input_tokens_seen": 65155860,
      "step": 3035,
      "time_per_iteration": 3.360319137573242
    },
    {
      "auxiliary_loss_clip": 0.01135732,
      "auxiliary_loss_mlp": 0.00763618,
      "balance_loss_clip": 1.0464834,
      "balance_loss_mlp": 1.00079393,
      "epoch": 0.3650574159802802,
      "flos": 24750496016640.0,
      "grad_norm": 1.4653325612181396,
      "language_loss": 0.78154784,
      "learning_rate": 2.9336352649018307e-06,
      "loss": 0.80054134,
      "num_input_tokens_seen": 65175930,
      "step": 3036,
      "time_per_iteration": 3.4139373302459717
    },
    {
      "auxiliary_loss_clip": 0.01163501,
      "auxiliary_loss_mlp": 0.01033235,
      "balance_loss_clip": 1.05402935,
      "balance_loss_mlp": 1.02508688,
      "epoch": 0.36517765887091924,
      "flos": 32853487363200.0,
      "grad_norm": 2.170677541433056,
      "language_loss": 0.70108736,
      "learning_rate": 2.9329463077191783e-06,
      "loss": 0.72305477,
      "num_input_tokens_seen": 65199305,
      "step": 3037,
      "time_per_iteration": 2.6024792194366455
    },
    {
      "auxiliary_loss_clip": 0.01131867,
      "auxiliary_loss_mlp": 0.01023347,
      "balance_loss_clip": 1.05067217,
      "balance_loss_mlp": 1.01478767,
      "epoch": 0.36529790176155835,
      "flos": 20120318282880.0,
      "grad_norm": 2.21491010429096,
      "language_loss": 0.63917327,
      "learning_rate": 2.9322572090093135e-06,
      "loss": 0.66072547,
      "num_input_tokens_seen": 65218010,
      "step": 3038,
      "time_per_iteration": 2.5622024536132812
    },
    {
      "auxiliary_loss_clip": 0.01132215,
      "auxiliary_loss_mlp": 0.01031505,
      "balance_loss_clip": 1.04761934,
      "balance_loss_mlp": 1.02286851,
      "epoch": 0.36541814465219746,
      "flos": 17639573379840.0,
      "grad_norm": 3.135063262481432,
      "language_loss": 0.76357126,
      "learning_rate": 2.9315679688767713e-06,
      "loss": 0.78520846,
      "num_input_tokens_seen": 65236020,
      "step": 3039,
      "time_per_iteration": 2.531160354614258
    },
    {
      "auxiliary_loss_clip": 0.01155675,
      "auxiliary_loss_mlp": 0.01028757,
      "balance_loss_clip": 1.05148506,
      "balance_loss_mlp": 1.02046919,
      "epoch": 0.3655383875428365,
      "flos": 22674356887680.0,
      "grad_norm": 1.5834482963541567,
      "language_loss": 0.66304111,
      "learning_rate": 2.9308785874261085e-06,
      "loss": 0.68488538,
      "num_input_tokens_seen": 65256210,
      "step": 3040,
      "time_per_iteration": 2.5264980792999268
    },
    {
      "auxiliary_loss_clip": 0.01191947,
      "auxiliary_loss_mlp": 0.01032058,
      "balance_loss_clip": 1.05893612,
      "balance_loss_mlp": 1.02393413,
      "epoch": 0.36565863043347563,
      "flos": 21981173247360.0,
      "grad_norm": 1.7571565599158954,
      "language_loss": 0.81965959,
      "learning_rate": 2.9301890647619045e-06,
      "loss": 0.84189957,
      "num_input_tokens_seen": 65275505,
      "step": 3041,
      "time_per_iteration": 2.4477994441986084
    },
    {
      "auxiliary_loss_clip": 0.01169142,
      "auxiliary_loss_mlp": 0.0103692,
      "balance_loss_clip": 1.05534339,
      "balance_loss_mlp": 1.02774668,
      "epoch": 0.36577887332411474,
      "flos": 24827632473600.0,
      "grad_norm": 1.8183491614990759,
      "language_loss": 0.8045131,
      "learning_rate": 2.929499400988759e-06,
      "loss": 0.82657367,
      "num_input_tokens_seen": 65296665,
      "step": 3042,
      "time_per_iteration": 2.5319924354553223
    },
    {
      "auxiliary_loss_clip": 0.01175415,
      "auxiliary_loss_mlp": 0.01035678,
      "balance_loss_clip": 1.05617571,
      "balance_loss_mlp": 1.02664185,
      "epoch": 0.3658991162147538,
      "flos": 28293191539200.0,
      "grad_norm": 1.957000854596714,
      "language_loss": 0.65246022,
      "learning_rate": 2.9288095962112927e-06,
      "loss": 0.67457116,
      "num_input_tokens_seen": 65317370,
      "step": 3043,
      "time_per_iteration": 2.5277597904205322
    },
    {
      "auxiliary_loss_clip": 0.01190413,
      "auxiliary_loss_mlp": 0.0102892,
      "balance_loss_clip": 1.05742955,
      "balance_loss_mlp": 1.02027154,
      "epoch": 0.3660193591053929,
      "flos": 17785550252160.0,
      "grad_norm": 1.8756394006013428,
      "language_loss": 0.84880692,
      "learning_rate": 2.9281196505341503e-06,
      "loss": 0.87100029,
      "num_input_tokens_seen": 65334540,
      "step": 3044,
      "time_per_iteration": 2.408989429473877
    },
    {
      "auxiliary_loss_clip": 0.01126781,
      "auxiliary_loss_mlp": 0.00763246,
      "balance_loss_clip": 1.05176497,
      "balance_loss_mlp": 1.00071049,
      "epoch": 0.36613960199603196,
      "flos": 10342776839040.0,
      "grad_norm": 2.169158605704409,
      "language_loss": 0.78737581,
      "learning_rate": 2.9274295640619946e-06,
      "loss": 0.80627608,
      "num_input_tokens_seen": 65351670,
      "step": 3045,
      "time_per_iteration": 2.5475940704345703
    },
    {
      "auxiliary_loss_clip": 0.01145906,
      "auxiliary_loss_mlp": 0.01026332,
      "balance_loss_clip": 1.05023766,
      "balance_loss_mlp": 1.01859498,
      "epoch": 0.36625984488667107,
      "flos": 19755609540480.0,
      "grad_norm": 7.06908958868098,
      "language_loss": 0.78218085,
      "learning_rate": 2.9267393368995103e-06,
      "loss": 0.80390322,
      "num_input_tokens_seen": 65370900,
      "step": 3046,
      "time_per_iteration": 2.515636444091797
    },
    {
      "auxiliary_loss_clip": 0.01192896,
      "auxiliary_loss_mlp": 0.01031641,
      "balance_loss_clip": 1.05931115,
      "balance_loss_mlp": 1.02339768,
      "epoch": 0.3663800877773102,
      "flos": 17674262939520.0,
      "grad_norm": 2.106238838173772,
      "language_loss": 0.73751926,
      "learning_rate": 2.926048969151407e-06,
      "loss": 0.75976467,
      "num_input_tokens_seen": 65388185,
      "step": 3047,
      "time_per_iteration": 2.4002580642700195
    },
    {
      "auxiliary_loss_clip": 0.0113082,
      "auxiliary_loss_mlp": 0.01028509,
      "balance_loss_clip": 1.05459344,
      "balance_loss_mlp": 1.01987267,
      "epoch": 0.36650033066794924,
      "flos": 20303606407680.0,
      "grad_norm": 1.9309487664588667,
      "language_loss": 0.68730938,
      "learning_rate": 2.92535846092241e-06,
      "loss": 0.70890272,
      "num_input_tokens_seen": 65407200,
      "step": 3048,
      "time_per_iteration": 2.5554943084716797
    },
    {
      "auxiliary_loss_clip": 0.01165626,
      "auxiliary_loss_mlp": 0.01031549,
      "balance_loss_clip": 1.05683959,
      "balance_loss_mlp": 1.02328801,
      "epoch": 0.36662057355858835,
      "flos": 24716237420160.0,
      "grad_norm": 2.3744138426062364,
      "language_loss": 0.82654011,
      "learning_rate": 2.9246678123172704e-06,
      "loss": 0.84851193,
      "num_input_tokens_seen": 65427290,
      "step": 3049,
      "time_per_iteration": 2.5517947673797607
    },
    {
      "auxiliary_loss_clip": 0.01192093,
      "auxiliary_loss_mlp": 0.0103574,
      "balance_loss_clip": 1.05754066,
      "balance_loss_mlp": 1.02699268,
      "epoch": 0.36674081644922746,
      "flos": 12385267902720.0,
      "grad_norm": 2.156203795964543,
      "language_loss": 0.74395168,
      "learning_rate": 2.9239770234407596e-06,
      "loss": 0.76622999,
      "num_input_tokens_seen": 65445595,
      "step": 3050,
      "time_per_iteration": 2.405888557434082
    },
    {
      "auxiliary_loss_clip": 0.01176614,
      "auxiliary_loss_mlp": 0.01027405,
      "balance_loss_clip": 1.05471969,
      "balance_loss_mlp": 1.01870859,
      "epoch": 0.3668610593398665,
      "flos": 21105922544640.0,
      "grad_norm": 1.5836866752241445,
      "language_loss": 0.68324101,
      "learning_rate": 2.9232860943976686e-06,
      "loss": 0.7052812,
      "num_input_tokens_seen": 65466330,
      "step": 3051,
      "time_per_iteration": 2.472597360610962
    },
    {
      "auxiliary_loss_clip": 0.01159864,
      "auxiliary_loss_mlp": 0.01025299,
      "balance_loss_clip": 1.05535579,
      "balance_loss_mlp": 1.01712132,
      "epoch": 0.3669813022305056,
      "flos": 26758082039040.0,
      "grad_norm": 1.6676428651189341,
      "language_loss": 0.83898771,
      "learning_rate": 2.9225950252928115e-06,
      "loss": 0.86083931,
      "num_input_tokens_seen": 65487180,
      "step": 3052,
      "time_per_iteration": 2.5359652042388916
    },
    {
      "auxiliary_loss_clip": 0.01178571,
      "auxiliary_loss_mlp": 0.00763803,
      "balance_loss_clip": 1.0590893,
      "balance_loss_mlp": 1.0006907,
      "epoch": 0.36710154512114473,
      "flos": 19099521671040.0,
      "grad_norm": 2.181883091419619,
      "language_loss": 0.81489873,
      "learning_rate": 2.9219038162310217e-06,
      "loss": 0.83432251,
      "num_input_tokens_seen": 65505380,
      "step": 3053,
      "time_per_iteration": 2.4488818645477295
    },
    {
      "auxiliary_loss_clip": 0.01109769,
      "auxiliary_loss_mlp": 0.01032799,
      "balance_loss_clip": 1.05098641,
      "balance_loss_mlp": 1.02443075,
      "epoch": 0.3672217880117838,
      "flos": 20812029465600.0,
      "grad_norm": 2.055129007854651,
      "language_loss": 0.8281616,
      "learning_rate": 2.921212467317157e-06,
      "loss": 0.84958726,
      "num_input_tokens_seen": 65524825,
      "step": 3054,
      "time_per_iteration": 2.6163485050201416
    },
    {
      "auxiliary_loss_clip": 0.01149447,
      "auxiliary_loss_mlp": 0.01031691,
      "balance_loss_clip": 1.05147016,
      "balance_loss_mlp": 1.02282739,
      "epoch": 0.3673420309024229,
      "flos": 13590394133760.0,
      "grad_norm": 1.8860546042439028,
      "language_loss": 0.80125493,
      "learning_rate": 2.920520978656093e-06,
      "loss": 0.82306629,
      "num_input_tokens_seen": 65541790,
      "step": 3055,
      "time_per_iteration": 2.4380366802215576
    },
    {
      "auxiliary_loss_clip": 0.01187204,
      "auxiliary_loss_mlp": 0.00763264,
      "balance_loss_clip": 1.05628812,
      "balance_loss_mlp": 1.00071919,
      "epoch": 0.367462273793062,
      "flos": 28986877969920.0,
      "grad_norm": 1.9361724785641345,
      "language_loss": 0.7692529,
      "learning_rate": 2.919829350352729e-06,
      "loss": 0.78875756,
      "num_input_tokens_seen": 65563395,
      "step": 3056,
      "time_per_iteration": 3.316824436187744
    },
    {
      "auxiliary_loss_clip": 0.01096426,
      "auxiliary_loss_mlp": 0.01009773,
      "balance_loss_clip": 1.02844,
      "balance_loss_mlp": 1.00865889,
      "epoch": 0.36758251668370107,
      "flos": 62643148346880.0,
      "grad_norm": 0.7570547829826064,
      "language_loss": 0.60030913,
      "learning_rate": 2.919137582511983e-06,
      "loss": 0.62137109,
      "num_input_tokens_seen": 65619835,
      "step": 3057,
      "time_per_iteration": 2.945138931274414
    },
    {
      "auxiliary_loss_clip": 0.011577,
      "auxiliary_loss_mlp": 0.01029523,
      "balance_loss_clip": 1.05977607,
      "balance_loss_mlp": 1.02150011,
      "epoch": 0.3677027595743402,
      "flos": 12713886455040.0,
      "grad_norm": 1.8878518158817985,
      "language_loss": 0.64157999,
      "learning_rate": 2.918445675238797e-06,
      "loss": 0.66345227,
      "num_input_tokens_seen": 65636760,
      "step": 3058,
      "time_per_iteration": 2.4971792697906494
    },
    {
      "auxiliary_loss_clip": 0.01191634,
      "auxiliary_loss_mlp": 0.0102699,
      "balance_loss_clip": 1.05725479,
      "balance_loss_mlp": 1.01840043,
      "epoch": 0.36782300246497923,
      "flos": 25046579825280.0,
      "grad_norm": 1.7956780496879863,
      "language_loss": 0.69444293,
      "learning_rate": 2.917753628638132e-06,
      "loss": 0.71662921,
      "num_input_tokens_seen": 65657065,
      "step": 3059,
      "time_per_iteration": 2.475940465927124
    },
    {
      "auxiliary_loss_clip": 0.01163311,
      "auxiliary_loss_mlp": 0.01027772,
      "balance_loss_clip": 1.05672216,
      "balance_loss_mlp": 1.01886678,
      "epoch": 0.36794324535561834,
      "flos": 17419512706560.0,
      "grad_norm": 2.0690827775904648,
      "language_loss": 0.70161605,
      "learning_rate": 2.9170614428149716e-06,
      "loss": 0.72352684,
      "num_input_tokens_seen": 65675400,
      "step": 3060,
      "time_per_iteration": 2.4628260135650635
    },
    {
      "auxiliary_loss_clip": 0.01144074,
      "auxiliary_loss_mlp": 0.0103567,
      "balance_loss_clip": 1.05368018,
      "balance_loss_mlp": 1.02645516,
      "epoch": 0.36806348824625745,
      "flos": 24089128848000.0,
      "grad_norm": 3.063828600476903,
      "language_loss": 0.86817425,
      "learning_rate": 2.9163691178743195e-06,
      "loss": 0.88997173,
      "num_input_tokens_seen": 65694050,
      "step": 3061,
      "time_per_iteration": 3.4269320964813232
    },
    {
      "auxiliary_loss_clip": 0.01172165,
      "auxiliary_loss_mlp": 0.01028956,
      "balance_loss_clip": 1.05546153,
      "balance_loss_mlp": 1.0203371,
      "epoch": 0.3681837311368965,
      "flos": 20521871400960.0,
      "grad_norm": 1.7580222715142169,
      "language_loss": 0.7723487,
      "learning_rate": 2.9156766539212006e-06,
      "loss": 0.79435992,
      "num_input_tokens_seen": 65711695,
      "step": 3062,
      "time_per_iteration": 3.234375
    },
    {
      "auxiliary_loss_clip": 0.01178483,
      "auxiliary_loss_mlp": 0.01040463,
      "balance_loss_clip": 1.05431294,
      "balance_loss_mlp": 1.03173709,
      "epoch": 0.3683039740275356,
      "flos": 21466644877440.0,
      "grad_norm": 1.8889714124636634,
      "language_loss": 0.71276712,
      "learning_rate": 2.9149840510606614e-06,
      "loss": 0.73495656,
      "num_input_tokens_seen": 65730350,
      "step": 3063,
      "time_per_iteration": 2.477159261703491
    },
    {
      "auxiliary_loss_clip": 0.01080107,
      "auxiliary_loss_mlp": 0.00754051,
      "balance_loss_clip": 1.02423072,
      "balance_loss_mlp": 1.0013026,
      "epoch": 0.36842421691817473,
      "flos": 70380999987840.0,
      "grad_norm": 1.026092408576825,
      "language_loss": 0.64183807,
      "learning_rate": 2.914291309397769e-06,
      "loss": 0.66017962,
      "num_input_tokens_seen": 65787820,
      "step": 3064,
      "time_per_iteration": 3.151080846786499
    },
    {
      "auxiliary_loss_clip": 0.01107949,
      "auxiliary_loss_mlp": 0.01028576,
      "balance_loss_clip": 1.04762554,
      "balance_loss_mlp": 1.01931953,
      "epoch": 0.3685444598088138,
      "flos": 23331378510720.0,
      "grad_norm": 2.2887398058454718,
      "language_loss": 0.78912234,
      "learning_rate": 2.9135984290376117e-06,
      "loss": 0.81048763,
      "num_input_tokens_seen": 65806685,
      "step": 3065,
      "time_per_iteration": 2.6122238636016846
    },
    {
      "auxiliary_loss_clip": 0.01115805,
      "auxiliary_loss_mlp": 0.0103187,
      "balance_loss_clip": 1.04797053,
      "balance_loss_mlp": 1.02332294,
      "epoch": 0.3686647026994529,
      "flos": 23070271570560.0,
      "grad_norm": 1.683605623804839,
      "language_loss": 0.82786429,
      "learning_rate": 2.9129054100853e-06,
      "loss": 0.84934103,
      "num_input_tokens_seen": 65825525,
      "step": 3066,
      "time_per_iteration": 2.601835012435913
    },
    {
      "auxiliary_loss_clip": 0.01162742,
      "auxiliary_loss_mlp": 0.01028063,
      "balance_loss_clip": 1.05429351,
      "balance_loss_mlp": 1.01949191,
      "epoch": 0.368784945590092,
      "flos": 25119909440640.0,
      "grad_norm": 1.6604839501720707,
      "language_loss": 0.75971091,
      "learning_rate": 2.912212252645963e-06,
      "loss": 0.78161895,
      "num_input_tokens_seen": 65848110,
      "step": 3067,
      "time_per_iteration": 2.551729679107666
    },
    {
      "auxiliary_loss_clip": 0.01180342,
      "auxiliary_loss_mlp": 0.01029661,
      "balance_loss_clip": 1.05453181,
      "balance_loss_mlp": 1.02104223,
      "epoch": 0.36890518848073106,
      "flos": 18442284566400.0,
      "grad_norm": 2.0223062882198777,
      "language_loss": 0.76295191,
      "learning_rate": 2.9115189568247523e-06,
      "loss": 0.78505194,
      "num_input_tokens_seen": 65865670,
      "step": 3068,
      "time_per_iteration": 2.447573184967041
    },
    {
      "auxiliary_loss_clip": 0.0112468,
      "auxiliary_loss_mlp": 0.01032262,
      "balance_loss_clip": 1.0548141,
      "balance_loss_mlp": 1.02363145,
      "epoch": 0.36902543137137017,
      "flos": 16362446336640.0,
      "grad_norm": 1.9020381572551877,
      "language_loss": 0.92165548,
      "learning_rate": 2.910825522726841e-06,
      "loss": 0.94322485,
      "num_input_tokens_seen": 65883195,
      "step": 3069,
      "time_per_iteration": 2.52885103225708
    },
    {
      "auxiliary_loss_clip": 0.01125887,
      "auxiliary_loss_mlp": 0.01032038,
      "balance_loss_clip": 1.04700148,
      "balance_loss_mlp": 1.02358055,
      "epoch": 0.3691456742620093,
      "flos": 12275596702080.0,
      "grad_norm": 1.9053274525265584,
      "language_loss": 0.77213216,
      "learning_rate": 2.9101319504574215e-06,
      "loss": 0.79371142,
      "num_input_tokens_seen": 65899635,
      "step": 3070,
      "time_per_iteration": 2.5252184867858887
    },
    {
      "auxiliary_loss_clip": 0.01164673,
      "auxiliary_loss_mlp": 0.01031266,
      "balance_loss_clip": 1.05078316,
      "balance_loss_mlp": 1.02218807,
      "epoch": 0.36926591715264834,
      "flos": 17786412178560.0,
      "grad_norm": 1.8746489748561939,
      "language_loss": 0.76626921,
      "learning_rate": 2.909438240121709e-06,
      "loss": 0.78822857,
      "num_input_tokens_seen": 65919910,
      "step": 3071,
      "time_per_iteration": 2.4977846145629883
    },
    {
      "auxiliary_loss_clip": 0.01154717,
      "auxiliary_loss_mlp": 0.01025087,
      "balance_loss_clip": 1.05415845,
      "balance_loss_mlp": 1.01650381,
      "epoch": 0.36938616004328745,
      "flos": 28948309741440.0,
      "grad_norm": 1.740603464270166,
      "language_loss": 0.70372581,
      "learning_rate": 2.908744391824939e-06,
      "loss": 0.72552383,
      "num_input_tokens_seen": 65940930,
      "step": 3072,
      "time_per_iteration": 2.576730728149414
    },
    {
      "auxiliary_loss_clip": 0.01119397,
      "auxiliary_loss_mlp": 0.01027762,
      "balance_loss_clip": 1.04758203,
      "balance_loss_mlp": 1.01906002,
      "epoch": 0.36950640293392656,
      "flos": 29205394358400.0,
      "grad_norm": 1.6289692631132964,
      "language_loss": 0.78989983,
      "learning_rate": 2.908050405672367e-06,
      "loss": 0.81137145,
      "num_input_tokens_seen": 65960475,
      "step": 3073,
      "time_per_iteration": 2.6455554962158203
    },
    {
      "auxiliary_loss_clip": 0.01164723,
      "auxiliary_loss_mlp": 0.01030034,
      "balance_loss_clip": 1.05042887,
      "balance_loss_mlp": 1.02133775,
      "epoch": 0.3696266458245656,
      "flos": 24827776128000.0,
      "grad_norm": 1.7226198747933599,
      "language_loss": 0.7927829,
      "learning_rate": 2.9073562817692703e-06,
      "loss": 0.81473047,
      "num_input_tokens_seen": 65979160,
      "step": 3074,
      "time_per_iteration": 2.52937912940979
    },
    {
      "auxiliary_loss_clip": 0.0105101,
      "auxiliary_loss_mlp": 0.01005299,
      "balance_loss_clip": 1.02100444,
      "balance_loss_mlp": 1.00398815,
      "epoch": 0.3697468887152047,
      "flos": 59887257264000.0,
      "grad_norm": 0.7234587820058428,
      "language_loss": 0.56523782,
      "learning_rate": 2.9066620202209468e-06,
      "loss": 0.58580089,
      "num_input_tokens_seen": 66041650,
      "step": 3075,
      "time_per_iteration": 3.073305368423462
    },
    {
      "auxiliary_loss_clip": 0.01137561,
      "auxiliary_loss_mlp": 0.01028035,
      "balance_loss_clip": 1.0513308,
      "balance_loss_mlp": 1.01965511,
      "epoch": 0.3698671316058438,
      "flos": 26137581569280.0,
      "grad_norm": 1.901466344564647,
      "language_loss": 0.77521539,
      "learning_rate": 2.905967621132716e-06,
      "loss": 0.7968713,
      "num_input_tokens_seen": 66059260,
      "step": 3076,
      "time_per_iteration": 2.55488920211792
    },
    {
      "auxiliary_loss_clip": 0.01164719,
      "auxiliary_loss_mlp": 0.01030537,
      "balance_loss_clip": 1.05217421,
      "balance_loss_mlp": 1.02199626,
      "epoch": 0.3699873744964829,
      "flos": 24607464059520.0,
      "grad_norm": 1.7925147239513082,
      "language_loss": 0.75092286,
      "learning_rate": 2.9052730846099172e-06,
      "loss": 0.77287543,
      "num_input_tokens_seen": 66080605,
      "step": 3077,
      "time_per_iteration": 2.532503843307495
    },
    {
      "auxiliary_loss_clip": 0.01067245,
      "auxiliary_loss_mlp": 0.0100115,
      "balance_loss_clip": 1.02181292,
      "balance_loss_mlp": 0.99992198,
      "epoch": 0.370107617387122,
      "flos": 64885340050560.0,
      "grad_norm": 0.8510180624822664,
      "language_loss": 0.60826206,
      "learning_rate": 2.9045784107579123e-06,
      "loss": 0.62894601,
      "num_input_tokens_seen": 66140710,
      "step": 3078,
      "time_per_iteration": 3.093737840652466
    },
    {
      "auxiliary_loss_clip": 0.01188915,
      "auxiliary_loss_mlp": 0.01027928,
      "balance_loss_clip": 1.05695796,
      "balance_loss_mlp": 1.01915979,
      "epoch": 0.37022786027776106,
      "flos": 15961683317760.0,
      "grad_norm": 2.928882061265177,
      "language_loss": 0.66829419,
      "learning_rate": 2.9038835996820807e-06,
      "loss": 0.69046259,
      "num_input_tokens_seen": 66158320,
      "step": 3079,
      "time_per_iteration": 2.416414737701416
    },
    {
      "auxiliary_loss_clip": 0.01149601,
      "auxiliary_loss_mlp": 0.01030944,
      "balance_loss_clip": 1.04854727,
      "balance_loss_mlp": 1.02205062,
      "epoch": 0.37034810316840017,
      "flos": 18546927863040.0,
      "grad_norm": 1.776021176644969,
      "language_loss": 0.79789054,
      "learning_rate": 2.903188651487826e-06,
      "loss": 0.81969607,
      "num_input_tokens_seen": 66176875,
      "step": 3080,
      "time_per_iteration": 2.505650758743286
    },
    {
      "auxiliary_loss_clip": 0.01179263,
      "auxiliary_loss_mlp": 0.01029699,
      "balance_loss_clip": 1.05684805,
      "balance_loss_mlp": 1.02081203,
      "epoch": 0.3704683460590393,
      "flos": 17821927751040.0,
      "grad_norm": 2.434656837212417,
      "language_loss": 0.86916959,
      "learning_rate": 2.902493566280571e-06,
      "loss": 0.89125919,
      "num_input_tokens_seen": 66194980,
      "step": 3081,
      "time_per_iteration": 2.4463279247283936
    },
    {
      "auxiliary_loss_clip": 0.01159309,
      "auxiliary_loss_mlp": 0.01027246,
      "balance_loss_clip": 1.0534687,
      "balance_loss_mlp": 1.01810908,
      "epoch": 0.37058858894967833,
      "flos": 14134081368960.0,
      "grad_norm": 8.847284099012898,
      "language_loss": 0.81579638,
      "learning_rate": 2.9017983441657595e-06,
      "loss": 0.83766186,
      "num_input_tokens_seen": 66212310,
      "step": 3082,
      "time_per_iteration": 2.4657857418060303
    },
    {
      "auxiliary_loss_clip": 0.01131881,
      "auxiliary_loss_mlp": 0.01027357,
      "balance_loss_clip": 1.0462513,
      "balance_loss_mlp": 1.01853573,
      "epoch": 0.37070883184031744,
      "flos": 13954492344960.0,
      "grad_norm": 2.7018465906506113,
      "language_loss": 0.75281215,
      "learning_rate": 2.9011029852488564e-06,
      "loss": 0.77440453,
      "num_input_tokens_seen": 66229545,
      "step": 3083,
      "time_per_iteration": 3.397322654724121
    },
    {
      "auxiliary_loss_clip": 0.01084857,
      "auxiliary_loss_mlp": 0.01000838,
      "balance_loss_clip": 1.01707315,
      "balance_loss_mlp": 0.99968141,
      "epoch": 0.37082907473095655,
      "flos": 52315419306240.0,
      "grad_norm": 0.9860272000605642,
      "language_loss": 0.62533206,
      "learning_rate": 2.9004074896353465e-06,
      "loss": 0.64618897,
      "num_input_tokens_seen": 66283545,
      "step": 3084,
      "time_per_iteration": 2.960568904876709
    },
    {
      "auxiliary_loss_clip": 0.0119141,
      "auxiliary_loss_mlp": 0.01026881,
      "balance_loss_clip": 1.06215191,
      "balance_loss_mlp": 1.01940107,
      "epoch": 0.3709493176215956,
      "flos": 15998096730240.0,
      "grad_norm": 1.7764345363015674,
      "language_loss": 0.81793094,
      "learning_rate": 2.8997118574307362e-06,
      "loss": 0.84011394,
      "num_input_tokens_seen": 66300500,
      "step": 3085,
      "time_per_iteration": 2.4666192531585693
    },
    {
      "auxiliary_loss_clip": 0.01152056,
      "auxiliary_loss_mlp": 0.01030487,
      "balance_loss_clip": 1.05247355,
      "balance_loss_mlp": 1.02201104,
      "epoch": 0.3710695605122347,
      "flos": 20959837931520.0,
      "grad_norm": 2.2091659403696626,
      "language_loss": 0.74374753,
      "learning_rate": 2.899016088740553e-06,
      "loss": 0.76557302,
      "num_input_tokens_seen": 66318610,
      "step": 3086,
      "time_per_iteration": 2.5196192264556885
    },
    {
      "auxiliary_loss_clip": 0.01129173,
      "auxiliary_loss_mlp": 0.01025001,
      "balance_loss_clip": 1.04856503,
      "balance_loss_mlp": 1.0173955,
      "epoch": 0.37118980340287383,
      "flos": 14355578586240.0,
      "grad_norm": 1.7841428804893666,
      "language_loss": 0.78998601,
      "learning_rate": 2.898320183670344e-06,
      "loss": 0.81152773,
      "num_input_tokens_seen": 66336025,
      "step": 3087,
      "time_per_iteration": 3.3778507709503174
    },
    {
      "auxiliary_loss_clip": 0.01132859,
      "auxiliary_loss_mlp": 0.01029608,
      "balance_loss_clip": 1.05426693,
      "balance_loss_mlp": 1.02101898,
      "epoch": 0.3713100462935129,
      "flos": 25885381201920.0,
      "grad_norm": 1.8573648256064648,
      "language_loss": 0.88930058,
      "learning_rate": 2.8976241423256767e-06,
      "loss": 0.91092527,
      "num_input_tokens_seen": 66356120,
      "step": 3088,
      "time_per_iteration": 4.074788331985474
    },
    {
      "auxiliary_loss_clip": 0.01153462,
      "auxiliary_loss_mlp": 0.01031042,
      "balance_loss_clip": 1.05084324,
      "balance_loss_mlp": 1.02316236,
      "epoch": 0.371430289184152,
      "flos": 30518934814080.0,
      "grad_norm": 1.8199088292121155,
      "language_loss": 0.67931402,
      "learning_rate": 2.896927964812142e-06,
      "loss": 0.70115912,
      "num_input_tokens_seen": 66376685,
      "step": 3089,
      "time_per_iteration": 2.5718977451324463
    },
    {
      "auxiliary_loss_clip": 0.01160492,
      "auxiliary_loss_mlp": 0.0102944,
      "balance_loss_clip": 1.0568192,
      "balance_loss_mlp": 1.02039194,
      "epoch": 0.37155053207479105,
      "flos": 15742233175680.0,
      "grad_norm": 2.3323219440267846,
      "language_loss": 0.75253165,
      "learning_rate": 2.8962316512353465e-06,
      "loss": 0.77443093,
      "num_input_tokens_seen": 66394230,
      "step": 3090,
      "time_per_iteration": 2.484325647354126
    },
    {
      "auxiliary_loss_clip": 0.01111884,
      "auxiliary_loss_mlp": 0.01031081,
      "balance_loss_clip": 1.04433692,
      "balance_loss_mlp": 1.02259326,
      "epoch": 0.37167077496543016,
      "flos": 23404061681280.0,
      "grad_norm": 1.5289820764494213,
      "language_loss": 0.74942881,
      "learning_rate": 2.8955352017009233e-06,
      "loss": 0.77085841,
      "num_input_tokens_seen": 66413475,
      "step": 3091,
      "time_per_iteration": 2.6065900325775146
    },
    {
      "auxiliary_loss_clip": 0.01160184,
      "auxiliary_loss_mlp": 0.0103184,
      "balance_loss_clip": 1.0563153,
      "balance_loss_mlp": 1.023054,
      "epoch": 0.3717910178560693,
      "flos": 22088653718400.0,
      "grad_norm": 3.1642929808981046,
      "language_loss": 0.77126801,
      "learning_rate": 2.8948386163145212e-06,
      "loss": 0.79318821,
      "num_input_tokens_seen": 66432685,
      "step": 3092,
      "time_per_iteration": 2.5059359073638916
    },
    {
      "auxiliary_loss_clip": 0.01179229,
      "auxiliary_loss_mlp": 0.01030179,
      "balance_loss_clip": 1.05617952,
      "balance_loss_mlp": 1.02188253,
      "epoch": 0.3719112607467083,
      "flos": 26939969533440.0,
      "grad_norm": 1.8237267847060572,
      "language_loss": 0.79071808,
      "learning_rate": 2.8941418951818135e-06,
      "loss": 0.81281215,
      "num_input_tokens_seen": 66452245,
      "step": 3093,
      "time_per_iteration": 2.4883203506469727
    },
    {
      "auxiliary_loss_clip": 0.01146028,
      "auxiliary_loss_mlp": 0.01032827,
      "balance_loss_clip": 1.05118442,
      "balance_loss_mlp": 1.02492309,
      "epoch": 0.37203150363734744,
      "flos": 12166500119040.0,
      "grad_norm": 2.3930619083808815,
      "language_loss": 0.71113873,
      "learning_rate": 2.8934450384084903e-06,
      "loss": 0.7329272,
      "num_input_tokens_seen": 66469760,
      "step": 3094,
      "time_per_iteration": 2.5112884044647217
    },
    {
      "auxiliary_loss_clip": 0.01155341,
      "auxiliary_loss_mlp": 0.01028875,
      "balance_loss_clip": 1.05298972,
      "balance_loss_mlp": 1.02022672,
      "epoch": 0.37215174652798655,
      "flos": 23697595624320.0,
      "grad_norm": 1.8284074983025054,
      "language_loss": 0.69826859,
      "learning_rate": 2.8927480461002653e-06,
      "loss": 0.72011077,
      "num_input_tokens_seen": 66489730,
      "step": 3095,
      "time_per_iteration": 2.5354344844818115
    },
    {
      "auxiliary_loss_clip": 0.01159411,
      "auxiliary_loss_mlp": 0.01037185,
      "balance_loss_clip": 1.05086005,
      "balance_loss_mlp": 1.02761877,
      "epoch": 0.3722719894186256,
      "flos": 17887751424000.0,
      "grad_norm": 3.192339297479285,
      "language_loss": 0.86417294,
      "learning_rate": 2.892050918362872e-06,
      "loss": 0.88613892,
      "num_input_tokens_seen": 66504785,
      "step": 3096,
      "time_per_iteration": 2.472733974456787
    },
    {
      "auxiliary_loss_clip": 0.010186,
      "auxiliary_loss_mlp": 0.01003817,
      "balance_loss_clip": 1.01639569,
      "balance_loss_mlp": 1.00242794,
      "epoch": 0.3723922323092647,
      "flos": 62419891363200.0,
      "grad_norm": 0.8496562944322783,
      "language_loss": 0.55903089,
      "learning_rate": 2.8913536553020626e-06,
      "loss": 0.5792551,
      "num_input_tokens_seen": 66558840,
      "step": 3097,
      "time_per_iteration": 3.5324528217315674
    },
    {
      "auxiliary_loss_clip": 0.01123517,
      "auxiliary_loss_mlp": 0.01027855,
      "balance_loss_clip": 1.04819775,
      "balance_loss_mlp": 1.01969504,
      "epoch": 0.3725124751999038,
      "flos": 23039747988480.0,
      "grad_norm": 2.1081206833218884,
      "language_loss": 0.84853786,
      "learning_rate": 2.8906562570236137e-06,
      "loss": 0.87005162,
      "num_input_tokens_seen": 66576750,
      "step": 3098,
      "time_per_iteration": 2.938476324081421
    },
    {
      "auxiliary_loss_clip": 0.01112249,
      "auxiliary_loss_mlp": 0.01034002,
      "balance_loss_clip": 1.04490685,
      "balance_loss_mlp": 1.02617621,
      "epoch": 0.3726327180905429,
      "flos": 20920551431040.0,
      "grad_norm": 1.4888644190420457,
      "language_loss": 0.76502621,
      "learning_rate": 2.889958723633318e-06,
      "loss": 0.78648865,
      "num_input_tokens_seen": 66595690,
      "step": 3099,
      "time_per_iteration": 2.6059322357177734
    },
    {
      "auxiliary_loss_clip": 0.01145617,
      "auxiliary_loss_mlp": 0.01028392,
      "balance_loss_clip": 1.05128145,
      "balance_loss_mlp": 1.01993418,
      "epoch": 0.372752960981182,
      "flos": 30592156688640.0,
      "grad_norm": 1.852895213674984,
      "language_loss": 0.73783535,
      "learning_rate": 2.889261055236992e-06,
      "loss": 0.75957549,
      "num_input_tokens_seen": 66617905,
      "step": 3100,
      "time_per_iteration": 2.6023433208465576
    },
    {
      "auxiliary_loss_clip": 0.01157703,
      "auxiliary_loss_mlp": 0.01026852,
      "balance_loss_clip": 1.0545969,
      "balance_loss_mlp": 1.01884675,
      "epoch": 0.3728732038718211,
      "flos": 25116749043840.0,
      "grad_norm": 1.9898948264863654,
      "language_loss": 0.82810473,
      "learning_rate": 2.8885632519404704e-06,
      "loss": 0.84995031,
      "num_input_tokens_seen": 66638175,
      "step": 3101,
      "time_per_iteration": 2.5411581993103027
    },
    {
      "auxiliary_loss_clip": 0.01161053,
      "auxiliary_loss_mlp": 0.01028527,
      "balance_loss_clip": 1.0559783,
      "balance_loss_mlp": 1.02005136,
      "epoch": 0.37299344676246016,
      "flos": 25302048330240.0,
      "grad_norm": 1.996871536090944,
      "language_loss": 0.75514901,
      "learning_rate": 2.8878653138496107e-06,
      "loss": 0.77704489,
      "num_input_tokens_seen": 66658670,
      "step": 3102,
      "time_per_iteration": 2.539137601852417
    },
    {
      "auxiliary_loss_clip": 0.01112825,
      "auxiliary_loss_mlp": 0.0103097,
      "balance_loss_clip": 1.04293156,
      "balance_loss_mlp": 1.02227378,
      "epoch": 0.37311368965309927,
      "flos": 23842531002240.0,
      "grad_norm": 2.252413582131106,
      "language_loss": 0.76544374,
      "learning_rate": 2.8871672410702878e-06,
      "loss": 0.78688169,
      "num_input_tokens_seen": 66676030,
      "step": 3103,
      "time_per_iteration": 2.6596555709838867
    },
    {
      "auxiliary_loss_clip": 0.01154628,
      "auxiliary_loss_mlp": 0.01029672,
      "balance_loss_clip": 1.05094266,
      "balance_loss_mlp": 1.02044511,
      "epoch": 0.3732339325437384,
      "flos": 25811943845760.0,
      "grad_norm": 1.8148055266422836,
      "language_loss": 0.82128298,
      "learning_rate": 2.8864690337084008e-06,
      "loss": 0.84312606,
      "num_input_tokens_seen": 66695305,
      "step": 3104,
      "time_per_iteration": 2.5743672847747803
    },
    {
      "auxiliary_loss_clip": 0.01169001,
      "auxiliary_loss_mlp": 0.01031418,
      "balance_loss_clip": 1.05298913,
      "balance_loss_mlp": 1.02213144,
      "epoch": 0.37335417543437743,
      "flos": 26208433146240.0,
      "grad_norm": 1.6933106138856056,
      "language_loss": 0.78106463,
      "learning_rate": 2.885770691869866e-06,
      "loss": 0.80306888,
      "num_input_tokens_seen": 66716185,
      "step": 3105,
      "time_per_iteration": 2.5092620849609375
    },
    {
      "auxiliary_loss_clip": 0.01170532,
      "auxiliary_loss_mlp": 0.0103245,
      "balance_loss_clip": 1.05424166,
      "balance_loss_mlp": 1.02455795,
      "epoch": 0.37347441832501654,
      "flos": 24023879792640.0,
      "grad_norm": 2.4903035712098958,
      "language_loss": 0.7450949,
      "learning_rate": 2.8850722156606207e-06,
      "loss": 0.76712465,
      "num_input_tokens_seen": 66734575,
      "step": 3106,
      "time_per_iteration": 2.496767044067383
    },
    {
      "auxiliary_loss_clip": 0.01167309,
      "auxiliary_loss_mlp": 0.01035184,
      "balance_loss_clip": 1.05325627,
      "balance_loss_mlp": 1.02698195,
      "epoch": 0.3735946612156556,
      "flos": 19714922409600.0,
      "grad_norm": 1.5434421292237188,
      "language_loss": 0.66832906,
      "learning_rate": 2.8843736051866252e-06,
      "loss": 0.69035399,
      "num_input_tokens_seen": 66753500,
      "step": 3107,
      "time_per_iteration": 2.4811577796936035
    },
    {
      "auxiliary_loss_clip": 0.01127216,
      "auxiliary_loss_mlp": 0.00763304,
      "balance_loss_clip": 1.04810548,
      "balance_loss_mlp": 1.00075698,
      "epoch": 0.3737149041062947,
      "flos": 23039604334080.0,
      "grad_norm": 1.698243829242408,
      "language_loss": 0.69489014,
      "learning_rate": 2.8836748605538557e-06,
      "loss": 0.7137953,
      "num_input_tokens_seen": 66775140,
      "step": 3108,
      "time_per_iteration": 2.6121256351470947
    },
    {
      "auxiliary_loss_clip": 0.01164737,
      "auxiliary_loss_mlp": 0.01025373,
      "balance_loss_clip": 1.05248666,
      "balance_loss_mlp": 1.01650369,
      "epoch": 0.3738351469969338,
      "flos": 34678108483200.0,
      "grad_norm": 3.0445117398238133,
      "language_loss": 0.63462484,
      "learning_rate": 2.882975981868313e-06,
      "loss": 0.65652591,
      "num_input_tokens_seen": 66795525,
      "step": 3109,
      "time_per_iteration": 3.6112146377563477
    },
    {
      "auxiliary_loss_clip": 0.01173385,
      "auxiliary_loss_mlp": 0.01027485,
      "balance_loss_clip": 1.05592632,
      "balance_loss_mlp": 1.01895547,
      "epoch": 0.3739553898875729,
      "flos": 43507967448960.0,
      "grad_norm": 2.392857825358932,
      "language_loss": 0.68686962,
      "learning_rate": 2.882276969236016e-06,
      "loss": 0.70887834,
      "num_input_tokens_seen": 66816885,
      "step": 3110,
      "time_per_iteration": 2.663795232772827
    },
    {
      "auxiliary_loss_clip": 0.01156132,
      "auxiliary_loss_mlp": 0.01029268,
      "balance_loss_clip": 1.05128145,
      "balance_loss_mlp": 1.0205245,
      "epoch": 0.374075632778212,
      "flos": 12856487448960.0,
      "grad_norm": 2.046303823997311,
      "language_loss": 0.76305211,
      "learning_rate": 2.881577822763005e-06,
      "loss": 0.78490615,
      "num_input_tokens_seen": 66834835,
      "step": 3111,
      "time_per_iteration": 2.464536190032959
    },
    {
      "auxiliary_loss_clip": 0.01171917,
      "auxiliary_loss_mlp": 0.01024302,
      "balance_loss_clip": 1.05332112,
      "balance_loss_mlp": 1.01644015,
      "epoch": 0.3741958756688511,
      "flos": 26024031699840.0,
      "grad_norm": 1.7865842072161702,
      "language_loss": 0.87259078,
      "learning_rate": 2.880878542555338e-06,
      "loss": 0.89455301,
      "num_input_tokens_seen": 66852600,
      "step": 3112,
      "time_per_iteration": 2.4953858852386475
    },
    {
      "auxiliary_loss_clip": 0.01189556,
      "auxiliary_loss_mlp": 0.01029965,
      "balance_loss_clip": 1.05648804,
      "balance_loss_mlp": 1.02138758,
      "epoch": 0.37431611855949015,
      "flos": 21433894652160.0,
      "grad_norm": 2.258291429189346,
      "language_loss": 0.80836713,
      "learning_rate": 2.8801791287190976e-06,
      "loss": 0.83056241,
      "num_input_tokens_seen": 66870595,
      "step": 3113,
      "time_per_iteration": 2.4246292114257812
    },
    {
      "auxiliary_loss_clip": 0.01173967,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.05243027,
      "balance_loss_mlp": 1.01953852,
      "epoch": 0.37443636145012926,
      "flos": 24207096090240.0,
      "grad_norm": 2.8971806978876233,
      "language_loss": 0.8596946,
      "learning_rate": 2.8794795813603817e-06,
      "loss": 0.88171387,
      "num_input_tokens_seen": 66886060,
      "step": 3114,
      "time_per_iteration": 3.3249690532684326
    },
    {
      "auxiliary_loss_clip": 0.01178232,
      "auxiliary_loss_mlp": 0.01031788,
      "balance_loss_clip": 1.0536319,
      "balance_loss_mlp": 1.02303791,
      "epoch": 0.3745566043407684,
      "flos": 15378601841280.0,
      "grad_norm": 1.8865316440154212,
      "language_loss": 0.81706363,
      "learning_rate": 2.878779900585314e-06,
      "loss": 0.83916384,
      "num_input_tokens_seen": 66903900,
      "step": 3115,
      "time_per_iteration": 3.2313897609710693
    },
    {
      "auxiliary_loss_clip": 0.01163875,
      "auxiliary_loss_mlp": 0.01029151,
      "balance_loss_clip": 1.0530107,
      "balance_loss_mlp": 1.02072275,
      "epoch": 0.37467684723140743,
      "flos": 24608218245120.0,
      "grad_norm": 1.5205094399668995,
      "language_loss": 0.75291908,
      "learning_rate": 2.8780800865000336e-06,
      "loss": 0.7748493,
      "num_input_tokens_seen": 66925210,
      "step": 3116,
      "time_per_iteration": 2.549765110015869
    },
    {
      "auxiliary_loss_clip": 0.01076843,
      "auxiliary_loss_mlp": 0.01006435,
      "balance_loss_clip": 1.01985288,
      "balance_loss_mlp": 1.00526702,
      "epoch": 0.37479709012204654,
      "flos": 64377491610240.0,
      "grad_norm": 0.975965565243997,
      "language_loss": 0.59196055,
      "learning_rate": 2.877380139210702e-06,
      "loss": 0.61279333,
      "num_input_tokens_seen": 66983880,
      "step": 3117,
      "time_per_iteration": 3.0146188735961914
    },
    {
      "auxiliary_loss_clip": 0.01146186,
      "auxiliary_loss_mlp": 0.01033,
      "balance_loss_clip": 1.05080748,
      "balance_loss_mlp": 1.02374387,
      "epoch": 0.37491733301268565,
      "flos": 23803962773760.0,
      "grad_norm": 2.1955290934198333,
      "language_loss": 0.76614767,
      "learning_rate": 2.876680058823501e-06,
      "loss": 0.78793955,
      "num_input_tokens_seen": 67004280,
      "step": 3118,
      "time_per_iteration": 2.557805299758911
    },
    {
      "auxiliary_loss_clip": 0.01148362,
      "auxiliary_loss_mlp": 0.01030191,
      "balance_loss_clip": 1.04963684,
      "balance_loss_mlp": 1.02151227,
      "epoch": 0.3750375759033247,
      "flos": 32160950167680.0,
      "grad_norm": 1.6945757924703555,
      "language_loss": 0.6601609,
      "learning_rate": 2.8759798454446314e-06,
      "loss": 0.68194646,
      "num_input_tokens_seen": 67027445,
      "step": 3119,
      "time_per_iteration": 2.588473081588745
    },
    {
      "auxiliary_loss_clip": 0.0117607,
      "auxiliary_loss_mlp": 0.01038566,
      "balance_loss_clip": 1.0547415,
      "balance_loss_mlp": 1.0303998,
      "epoch": 0.3751578187939638,
      "flos": 23367791923200.0,
      "grad_norm": 2.007764817957837,
      "language_loss": 0.81639218,
      "learning_rate": 2.8752794991803173e-06,
      "loss": 0.83853859,
      "num_input_tokens_seen": 67045130,
      "step": 3120,
      "time_per_iteration": 2.4711804389953613
    },
    {
      "auxiliary_loss_clip": 0.01156865,
      "auxiliary_loss_mlp": 0.01032258,
      "balance_loss_clip": 1.05206013,
      "balance_loss_mlp": 1.02383029,
      "epoch": 0.37527806168460287,
      "flos": 14605731878400.0,
      "grad_norm": 2.614544852039984,
      "language_loss": 0.75237489,
      "learning_rate": 2.8745790201367976e-06,
      "loss": 0.77426612,
      "num_input_tokens_seen": 67060885,
      "step": 3121,
      "time_per_iteration": 2.4698715209960938
    },
    {
      "auxiliary_loss_clip": 0.01190986,
      "auxiliary_loss_mlp": 0.01036863,
      "balance_loss_clip": 1.05686259,
      "balance_loss_mlp": 1.02804184,
      "epoch": 0.375398304575242,
      "flos": 26390823431040.0,
      "grad_norm": 3.286901787291444,
      "language_loss": 0.84436589,
      "learning_rate": 2.8738784084203373e-06,
      "loss": 0.86664438,
      "num_input_tokens_seen": 67080960,
      "step": 3122,
      "time_per_iteration": 2.469667673110962
    },
    {
      "auxiliary_loss_clip": 0.01149316,
      "auxiliary_loss_mlp": 0.01026991,
      "balance_loss_clip": 1.04714763,
      "balance_loss_mlp": 1.01893294,
      "epoch": 0.3755185474658811,
      "flos": 22236605838720.0,
      "grad_norm": 1.6258532256173992,
      "language_loss": 0.78889155,
      "learning_rate": 2.873177664137216e-06,
      "loss": 0.81065464,
      "num_input_tokens_seen": 67101890,
      "step": 3123,
      "time_per_iteration": 2.5259437561035156
    },
    {
      "auxiliary_loss_clip": 0.01138532,
      "auxiliary_loss_mlp": 0.01024502,
      "balance_loss_clip": 1.05095363,
      "balance_loss_mlp": 1.01613986,
      "epoch": 0.37563879035652015,
      "flos": 30812935633920.0,
      "grad_norm": 1.700286020426204,
      "language_loss": 0.69339776,
      "learning_rate": 2.8724767873937384e-06,
      "loss": 0.71502805,
      "num_input_tokens_seen": 67126010,
      "step": 3124,
      "time_per_iteration": 2.6263017654418945
    },
    {
      "auxiliary_loss_clip": 0.01158443,
      "auxiliary_loss_mlp": 0.01034542,
      "balance_loss_clip": 1.0518086,
      "balance_loss_mlp": 1.02633464,
      "epoch": 0.37575903324715926,
      "flos": 20773533064320.0,
      "grad_norm": 2.5622354056585364,
      "language_loss": 0.87532628,
      "learning_rate": 2.871775778296225e-06,
      "loss": 0.89725614,
      "num_input_tokens_seen": 67143100,
      "step": 3125,
      "time_per_iteration": 2.4712038040161133
    },
    {
      "auxiliary_loss_clip": 0.01176876,
      "auxiliary_loss_mlp": 0.0103428,
      "balance_loss_clip": 1.05744195,
      "balance_loss_mlp": 1.02479053,
      "epoch": 0.37587927613779837,
      "flos": 18697681244160.0,
      "grad_norm": 1.9549841962074688,
      "language_loss": 0.78357667,
      "learning_rate": 2.8710746369510196e-06,
      "loss": 0.8056882,
      "num_input_tokens_seen": 67161085,
      "step": 3126,
      "time_per_iteration": 2.443674325942993
    },
    {
      "auxiliary_loss_clip": 0.01153252,
      "auxiliary_loss_mlp": 0.01029526,
      "balance_loss_clip": 1.05298221,
      "balance_loss_mlp": 1.02138352,
      "epoch": 0.3759995190284374,
      "flos": 13624796384640.0,
      "grad_norm": 2.392028777529309,
      "language_loss": 0.83229095,
      "learning_rate": 2.8703733634644846e-06,
      "loss": 0.8541187,
      "num_input_tokens_seen": 67175840,
      "step": 3127,
      "time_per_iteration": 2.449815273284912
    },
    {
      "auxiliary_loss_clip": 0.01184592,
      "auxiliary_loss_mlp": 0.01028502,
      "balance_loss_clip": 1.05540895,
      "balance_loss_mlp": 1.02022266,
      "epoch": 0.37611976191907653,
      "flos": 20484847457280.0,
      "grad_norm": 1.765232265266723,
      "language_loss": 0.79220593,
      "learning_rate": 2.869671957943002e-06,
      "loss": 0.8143369,
      "num_input_tokens_seen": 67194995,
      "step": 3128,
      "time_per_iteration": 2.4282548427581787
    },
    {
      "auxiliary_loss_clip": 0.0115593,
      "auxiliary_loss_mlp": 0.01028357,
      "balance_loss_clip": 1.05754912,
      "balance_loss_mlp": 1.01995301,
      "epoch": 0.37624000480971564,
      "flos": 21141797253120.0,
      "grad_norm": 1.7464556906682398,
      "language_loss": 0.73718297,
      "learning_rate": 2.8689704204929747e-06,
      "loss": 0.75902581,
      "num_input_tokens_seen": 67214175,
      "step": 3129,
      "time_per_iteration": 2.496959924697876
    },
    {
      "auxiliary_loss_clip": 0.01186897,
      "auxiliary_loss_mlp": 0.01030624,
      "balance_loss_clip": 1.05572617,
      "balance_loss_mlp": 1.02212489,
      "epoch": 0.3763602477003547,
      "flos": 22564470205440.0,
      "grad_norm": 1.723356042838573,
      "language_loss": 0.81053418,
      "learning_rate": 2.8682687512208253e-06,
      "loss": 0.83270943,
      "num_input_tokens_seen": 67233185,
      "step": 3130,
      "time_per_iteration": 2.434842348098755
    },
    {
      "auxiliary_loss_clip": 0.01179706,
      "auxiliary_loss_mlp": 0.01033594,
      "balance_loss_clip": 1.05539799,
      "balance_loss_mlp": 1.02483177,
      "epoch": 0.3764804905909938,
      "flos": 27526857851520.0,
      "grad_norm": 2.198284008165177,
      "language_loss": 0.80208343,
      "learning_rate": 2.8675669502329972e-06,
      "loss": 0.82421649,
      "num_input_tokens_seen": 67254715,
      "step": 3131,
      "time_per_iteration": 2.5020923614501953
    },
    {
      "auxiliary_loss_clip": 0.01175319,
      "auxiliary_loss_mlp": 0.00763399,
      "balance_loss_clip": 1.0556109,
      "balance_loss_mlp": 1.00060058,
      "epoch": 0.3766007334816329,
      "flos": 22528092706560.0,
      "grad_norm": 2.4549663549152756,
      "language_loss": 0.85911304,
      "learning_rate": 2.866865017635952e-06,
      "loss": 0.87850022,
      "num_input_tokens_seen": 67272535,
      "step": 3132,
      "time_per_iteration": 2.4791293144226074
    },
    {
      "auxiliary_loss_clip": 0.01144208,
      "auxiliary_loss_mlp": 0.01027583,
      "balance_loss_clip": 1.05422175,
      "balance_loss_mlp": 1.01894581,
      "epoch": 0.376720976372272,
      "flos": 25957166532480.0,
      "grad_norm": 1.5446627872810743,
      "language_loss": 0.79567385,
      "learning_rate": 2.866162953536174e-06,
      "loss": 0.81739175,
      "num_input_tokens_seen": 67293505,
      "step": 3133,
      "time_per_iteration": 2.5564582347869873
    },
    {
      "auxiliary_loss_clip": 0.01156655,
      "auxiliary_loss_mlp": 0.00763008,
      "balance_loss_clip": 1.05179453,
      "balance_loss_mlp": 1.00053525,
      "epoch": 0.3768412192629111,
      "flos": 18041162411520.0,
      "grad_norm": 1.6167623823912223,
      "language_loss": 0.75061142,
      "learning_rate": 2.8654607580401634e-06,
      "loss": 0.76980805,
      "num_input_tokens_seen": 67313240,
      "step": 3134,
      "time_per_iteration": 2.5095746517181396
    },
    {
      "auxiliary_loss_clip": 0.01074758,
      "auxiliary_loss_mlp": 0.01001733,
      "balance_loss_clip": 1.01861119,
      "balance_loss_mlp": 1.00059497,
      "epoch": 0.3769614621535502,
      "flos": 62989472304000.0,
      "grad_norm": 0.8811534456221024,
      "language_loss": 0.65200293,
      "learning_rate": 2.8647584312544446e-06,
      "loss": 0.67276788,
      "num_input_tokens_seen": 67378445,
      "step": 3135,
      "time_per_iteration": 3.070974588394165
    },
    {
      "auxiliary_loss_clip": 0.01138074,
      "auxiliary_loss_mlp": 0.00763079,
      "balance_loss_clip": 1.04827738,
      "balance_loss_mlp": 1.00057137,
      "epoch": 0.37708170504418925,
      "flos": 23661685002240.0,
      "grad_norm": 1.3930429172460972,
      "language_loss": 0.85039067,
      "learning_rate": 2.864055973285559e-06,
      "loss": 0.86940217,
      "num_input_tokens_seen": 67400445,
      "step": 3136,
      "time_per_iteration": 3.4620230197906494
    },
    {
      "auxiliary_loss_clip": 0.01148162,
      "auxiliary_loss_mlp": 0.01033728,
      "balance_loss_clip": 1.05047369,
      "balance_loss_mlp": 1.02497792,
      "epoch": 0.37720194793482836,
      "flos": 24423170353920.0,
      "grad_norm": 1.7110220162382623,
      "language_loss": 0.8608647,
      "learning_rate": 2.8633533842400698e-06,
      "loss": 0.88268363,
      "num_input_tokens_seen": 67420645,
      "step": 3137,
      "time_per_iteration": 2.5413384437561035
    },
    {
      "auxiliary_loss_clip": 0.01172878,
      "auxiliary_loss_mlp": 0.00763479,
      "balance_loss_clip": 1.05405211,
      "balance_loss_mlp": 1.00065172,
      "epoch": 0.3773221908254674,
      "flos": 20996502739200.0,
      "grad_norm": 1.8067888065358066,
      "language_loss": 0.77202499,
      "learning_rate": 2.862650664224558e-06,
      "loss": 0.79138851,
      "num_input_tokens_seen": 67439495,
      "step": 3138,
      "time_per_iteration": 2.474343776702881
    },
    {
      "auxiliary_loss_clip": 0.01171471,
      "auxiliary_loss_mlp": 0.01025909,
      "balance_loss_clip": 1.05762529,
      "balance_loss_mlp": 1.01851761,
      "epoch": 0.37744243371610653,
      "flos": 37631724958080.0,
      "grad_norm": 1.400180293354287,
      "language_loss": 0.6971063,
      "learning_rate": 2.861947813345627e-06,
      "loss": 0.71908009,
      "num_input_tokens_seen": 67462195,
      "step": 3139,
      "time_per_iteration": 2.6568715572357178
    },
    {
      "auxiliary_loss_clip": 0.0119042,
      "auxiliary_loss_mlp": 0.00763213,
      "balance_loss_clip": 1.05854702,
      "balance_loss_mlp": 1.000525,
      "epoch": 0.37756267660674564,
      "flos": 26140526484480.0,
      "grad_norm": 2.9436515015122033,
      "language_loss": 0.72500849,
      "learning_rate": 2.8612448317098974e-06,
      "loss": 0.74454486,
      "num_input_tokens_seen": 67482530,
      "step": 3140,
      "time_per_iteration": 3.3453752994537354
    },
    {
      "auxiliary_loss_clip": 0.01147114,
      "auxiliary_loss_mlp": 0.00763002,
      "balance_loss_clip": 1.05067492,
      "balance_loss_mlp": 1.00060177,
      "epoch": 0.3776829194973847,
      "flos": 19427888828160.0,
      "grad_norm": 2.1740787755318682,
      "language_loss": 0.83223599,
      "learning_rate": 2.8605417194240114e-06,
      "loss": 0.85133713,
      "num_input_tokens_seen": 67500890,
      "step": 3141,
      "time_per_iteration": 2.5429534912109375
    },
    {
      "auxiliary_loss_clip": 0.01164911,
      "auxiliary_loss_mlp": 0.01027254,
      "balance_loss_clip": 1.05115652,
      "balance_loss_mlp": 1.01951098,
      "epoch": 0.3778031623880238,
      "flos": 17382309194880.0,
      "grad_norm": 2.9788640345191495,
      "language_loss": 0.78936994,
      "learning_rate": 2.8598384765946315e-06,
      "loss": 0.81129164,
      "num_input_tokens_seen": 67519545,
      "step": 3142,
      "time_per_iteration": 3.2364134788513184
    },
    {
      "auxiliary_loss_clip": 0.01185699,
      "auxiliary_loss_mlp": 0.0102849,
      "balance_loss_clip": 1.05346358,
      "balance_loss_mlp": 1.02037144,
      "epoch": 0.3779234052786629,
      "flos": 27125843437440.0,
      "grad_norm": 1.8769201854494677,
      "language_loss": 0.7180388,
      "learning_rate": 2.8591351033284377e-06,
      "loss": 0.74018073,
      "num_input_tokens_seen": 67539275,
      "step": 3143,
      "time_per_iteration": 3.2351582050323486
    },
    {
      "auxiliary_loss_clip": 0.011739,
      "auxiliary_loss_mlp": 0.01026651,
      "balance_loss_clip": 1.05205965,
      "balance_loss_mlp": 1.01873887,
      "epoch": 0.37804364816930197,
      "flos": 19682639061120.0,
      "grad_norm": 2.015355793944523,
      "language_loss": 0.83983552,
      "learning_rate": 2.8584315997321325e-06,
      "loss": 0.86184108,
      "num_input_tokens_seen": 67558280,
      "step": 3144,
      "time_per_iteration": 2.50832462310791
    },
    {
      "auxiliary_loss_clip": 0.01186059,
      "auxiliary_loss_mlp": 0.0102712,
      "balance_loss_clip": 1.05478227,
      "balance_loss_mlp": 1.01867449,
      "epoch": 0.3781638910599411,
      "flos": 22702905221760.0,
      "grad_norm": 4.6456889317445595,
      "language_loss": 0.780828,
      "learning_rate": 2.8577279659124356e-06,
      "loss": 0.8029598,
      "num_input_tokens_seen": 67575955,
      "step": 3145,
      "time_per_iteration": 2.43192458152771
    },
    {
      "auxiliary_loss_clip": 0.0116686,
      "auxiliary_loss_mlp": 0.01026431,
      "balance_loss_clip": 1.0515486,
      "balance_loss_mlp": 1.01945472,
      "epoch": 0.3782841339505802,
      "flos": 14647604158080.0,
      "grad_norm": 1.746516935686437,
      "language_loss": 0.83336186,
      "learning_rate": 2.857024201976089e-06,
      "loss": 0.85529482,
      "num_input_tokens_seen": 67593515,
      "step": 3146,
      "time_per_iteration": 2.4630327224731445
    },
    {
      "auxiliary_loss_clip": 0.01155457,
      "auxiliary_loss_mlp": 0.01026075,
      "balance_loss_clip": 1.05408263,
      "balance_loss_mlp": 1.01752472,
      "epoch": 0.37840437684121925,
      "flos": 32818223185920.0,
      "grad_norm": 1.8728539770349364,
      "language_loss": 0.73618233,
      "learning_rate": 2.8563203080298516e-06,
      "loss": 0.75799763,
      "num_input_tokens_seen": 67614290,
      "step": 3147,
      "time_per_iteration": 2.5908761024475098
    },
    {
      "auxiliary_loss_clip": 0.01157094,
      "auxiliary_loss_mlp": 0.00763172,
      "balance_loss_clip": 1.05285096,
      "balance_loss_mlp": 1.00055099,
      "epoch": 0.37852461973185836,
      "flos": 18369206346240.0,
      "grad_norm": 2.079168619004708,
      "language_loss": 0.89210075,
      "learning_rate": 2.855616284180505e-06,
      "loss": 0.9113034,
      "num_input_tokens_seen": 67631340,
      "step": 3148,
      "time_per_iteration": 2.492553472518921
    },
    {
      "auxiliary_loss_clip": 0.01077776,
      "auxiliary_loss_mlp": 0.01002141,
      "balance_loss_clip": 1.01902878,
      "balance_loss_mlp": 1.0008595,
      "epoch": 0.37864486262249747,
      "flos": 59500680117120.0,
      "grad_norm": 0.8727031367201991,
      "language_loss": 0.66163015,
      "learning_rate": 2.8549121305348477e-06,
      "loss": 0.68242937,
      "num_input_tokens_seen": 67691125,
      "step": 3149,
      "time_per_iteration": 3.014672040939331
    },
    {
      "auxiliary_loss_clip": 0.01171632,
      "auxiliary_loss_mlp": 0.01030233,
      "balance_loss_clip": 1.05333805,
      "balance_loss_mlp": 1.02304149,
      "epoch": 0.3787651055131365,
      "flos": 23363015414400.0,
      "grad_norm": 2.0741758902264285,
      "language_loss": 0.83218241,
      "learning_rate": 2.8542078471997006e-06,
      "loss": 0.85420108,
      "num_input_tokens_seen": 67708740,
      "step": 3150,
      "time_per_iteration": 2.4627511501312256
    },
    {
      "auxiliary_loss_clip": 0.01170095,
      "auxiliary_loss_mlp": 0.01024641,
      "balance_loss_clip": 1.05196786,
      "balance_loss_mlp": 1.01736963,
      "epoch": 0.37888534840377563,
      "flos": 24601394661120.0,
      "grad_norm": 1.6798829687995607,
      "language_loss": 0.75739157,
      "learning_rate": 2.8535034342819013e-06,
      "loss": 0.77933896,
      "num_input_tokens_seen": 67726150,
      "step": 3151,
      "time_per_iteration": 2.4999353885650635
    },
    {
      "auxiliary_loss_clip": 0.01180723,
      "auxiliary_loss_mlp": 0.01031003,
      "balance_loss_clip": 1.0526576,
      "balance_loss_mlp": 1.0228374,
      "epoch": 0.37900559129441475,
      "flos": 23986891762560.0,
      "grad_norm": 1.8985002836540235,
      "language_loss": 0.72623253,
      "learning_rate": 2.85279889188831e-06,
      "loss": 0.74834985,
      "num_input_tokens_seen": 67746525,
      "step": 3152,
      "time_per_iteration": 2.4679253101348877
    },
    {
      "auxiliary_loss_clip": 0.01140687,
      "auxiliary_loss_mlp": 0.01025549,
      "balance_loss_clip": 1.04550624,
      "balance_loss_mlp": 1.01670969,
      "epoch": 0.3791258341850538,
      "flos": 24644667571200.0,
      "grad_norm": 1.8830747005249437,
      "language_loss": 0.81117892,
      "learning_rate": 2.852094220125805e-06,
      "loss": 0.83284134,
      "num_input_tokens_seen": 67766035,
      "step": 3153,
      "time_per_iteration": 2.553452968597412
    },
    {
      "auxiliary_loss_clip": 0.01172363,
      "auxiliary_loss_mlp": 0.01036381,
      "balance_loss_clip": 1.05350137,
      "balance_loss_mlp": 1.02798831,
      "epoch": 0.3792460770756929,
      "flos": 17420841509760.0,
      "grad_norm": 2.011030628955631,
      "language_loss": 0.71175748,
      "learning_rate": 2.8513894191012846e-06,
      "loss": 0.73384488,
      "num_input_tokens_seen": 67785015,
      "step": 3154,
      "time_per_iteration": 2.4432120323181152
    },
    {
      "auxiliary_loss_clip": 0.01185751,
      "auxiliary_loss_mlp": 0.01029871,
      "balance_loss_clip": 1.05495906,
      "balance_loss_mlp": 1.02147841,
      "epoch": 0.37936631996633197,
      "flos": 24206557386240.0,
      "grad_norm": 1.4748569945703571,
      "language_loss": 0.78986973,
      "learning_rate": 2.8506844889216664e-06,
      "loss": 0.8120259,
      "num_input_tokens_seen": 67804400,
      "step": 3155,
      "time_per_iteration": 2.451220989227295
    },
    {
      "auxiliary_loss_clip": 0.0107217,
      "auxiliary_loss_mlp": 0.01003007,
      "balance_loss_clip": 1.01990426,
      "balance_loss_mlp": 1.00193441,
      "epoch": 0.3794865628569711,
      "flos": 70297114752000.0,
      "grad_norm": 0.8607819496203047,
      "language_loss": 0.62838602,
      "learning_rate": 2.849979429693887e-06,
      "loss": 0.64913774,
      "num_input_tokens_seen": 67865385,
      "step": 3156,
      "time_per_iteration": 3.119034767150879
    },
    {
      "auxiliary_loss_clip": 0.01181803,
      "auxiliary_loss_mlp": 0.01027025,
      "balance_loss_clip": 1.05283451,
      "balance_loss_mlp": 1.01924086,
      "epoch": 0.3796068057476102,
      "flos": 15779364860160.0,
      "grad_norm": 2.0084870282797436,
      "language_loss": 0.74210215,
      "learning_rate": 2.8492742415249042e-06,
      "loss": 0.76419044,
      "num_input_tokens_seen": 67883030,
      "step": 3157,
      "time_per_iteration": 2.5124590396881104
    },
    {
      "auxiliary_loss_clip": 0.01181307,
      "auxiliary_loss_mlp": 0.01024898,
      "balance_loss_clip": 1.05115533,
      "balance_loss_mlp": 1.01722717,
      "epoch": 0.37972704863824924,
      "flos": 25191694771200.0,
      "grad_norm": 1.6691102650817446,
      "language_loss": 0.76180249,
      "learning_rate": 2.848568924521694e-06,
      "loss": 0.78386456,
      "num_input_tokens_seen": 67903810,
      "step": 3158,
      "time_per_iteration": 2.460641860961914
    },
    {
      "auxiliary_loss_clip": 0.01161962,
      "auxiliary_loss_mlp": 0.01024617,
      "balance_loss_clip": 1.04823327,
      "balance_loss_mlp": 1.01620042,
      "epoch": 0.37984729152888835,
      "flos": 26210372480640.0,
      "grad_norm": 1.7492059876551591,
      "language_loss": 0.73334646,
      "learning_rate": 2.8478634787912526e-06,
      "loss": 0.75521225,
      "num_input_tokens_seen": 67921865,
      "step": 3159,
      "time_per_iteration": 2.494974136352539
    },
    {
      "auxiliary_loss_clip": 0.01170244,
      "auxiliary_loss_mlp": 0.0102926,
      "balance_loss_clip": 1.05121291,
      "balance_loss_mlp": 1.02118349,
      "epoch": 0.37996753441952746,
      "flos": 25629302165760.0,
      "grad_norm": 1.9521277796201606,
      "language_loss": 0.76368785,
      "learning_rate": 2.847157904440596e-06,
      "loss": 0.78568292,
      "num_input_tokens_seen": 67941595,
      "step": 3160,
      "time_per_iteration": 2.488924503326416
    },
    {
      "auxiliary_loss_clip": 0.01168733,
      "auxiliary_loss_mlp": 0.01028002,
      "balance_loss_clip": 1.05106163,
      "balance_loss_mlp": 1.0202713,
      "epoch": 0.3800877773101665,
      "flos": 20118414862080.0,
      "grad_norm": 1.5681746370415068,
      "language_loss": 0.73709714,
      "learning_rate": 2.846452201576759e-06,
      "loss": 0.75906456,
      "num_input_tokens_seen": 67960970,
      "step": 3161,
      "time_per_iteration": 2.504160165786743
    },
    {
      "auxiliary_loss_clip": 0.01066212,
      "auxiliary_loss_mlp": 0.01001792,
      "balance_loss_clip": 1.01723588,
      "balance_loss_mlp": 1.00053406,
      "epoch": 0.38020802020080563,
      "flos": 63053608037760.0,
      "grad_norm": 0.8522443023093426,
      "language_loss": 0.62794399,
      "learning_rate": 2.845746370306795e-06,
      "loss": 0.64862406,
      "num_input_tokens_seen": 68026160,
      "step": 3162,
      "time_per_iteration": 3.1608963012695312
    },
    {
      "auxiliary_loss_clip": 0.01170603,
      "auxiliary_loss_mlp": 0.01028638,
      "balance_loss_clip": 1.05271387,
      "balance_loss_mlp": 1.02056789,
      "epoch": 0.38032826309144474,
      "flos": 21288420570240.0,
      "grad_norm": 2.0289624462301235,
      "language_loss": 0.78352225,
      "learning_rate": 2.84504041073778e-06,
      "loss": 0.80551469,
      "num_input_tokens_seen": 68044575,
      "step": 3163,
      "time_per_iteration": 3.2978944778442383
    },
    {
      "auxiliary_loss_clip": 0.01149744,
      "auxiliary_loss_mlp": 0.01035938,
      "balance_loss_clip": 1.05061769,
      "balance_loss_mlp": 1.02728987,
      "epoch": 0.3804485059820838,
      "flos": 18954119416320.0,
      "grad_norm": 1.68518794615801,
      "language_loss": 0.78961897,
      "learning_rate": 2.844334322976806e-06,
      "loss": 0.81147581,
      "num_input_tokens_seen": 68064790,
      "step": 3164,
      "time_per_iteration": 2.4990808963775635
    },
    {
      "auxiliary_loss_clip": 0.01129429,
      "auxiliary_loss_mlp": 0.01035414,
      "balance_loss_clip": 1.04753578,
      "balance_loss_mlp": 1.02714658,
      "epoch": 0.3805687488727229,
      "flos": 21833759831040.0,
      "grad_norm": 1.7590650241902899,
      "language_loss": 0.8353042,
      "learning_rate": 2.8436281071309866e-06,
      "loss": 0.85695261,
      "num_input_tokens_seen": 68083330,
      "step": 3165,
      "time_per_iteration": 2.578946352005005
    },
    {
      "auxiliary_loss_clip": 0.01043514,
      "auxiliary_loss_mlp": 0.01003189,
      "balance_loss_clip": 1.01465905,
      "balance_loss_mlp": 1.00191975,
      "epoch": 0.380688991763362,
      "flos": 58546209968640.0,
      "grad_norm": 0.7280922055584209,
      "language_loss": 0.5307098,
      "learning_rate": 2.842921763307455e-06,
      "loss": 0.55117679,
      "num_input_tokens_seen": 68146140,
      "step": 3166,
      "time_per_iteration": 3.1237878799438477
    },
    {
      "auxiliary_loss_clip": 0.01147546,
      "auxiliary_loss_mlp": 0.01026673,
      "balance_loss_clip": 1.048563,
      "balance_loss_mlp": 1.01918101,
      "epoch": 0.38080923465400107,
      "flos": 23799509487360.0,
      "grad_norm": 1.76003533843277,
      "language_loss": 0.82244015,
      "learning_rate": 2.842215291613361e-06,
      "loss": 0.84418237,
      "num_input_tokens_seen": 68164520,
      "step": 3167,
      "time_per_iteration": 3.350708484649658
    },
    {
      "auxiliary_loss_clip": 0.01008367,
      "auxiliary_loss_mlp": 0.01000715,
      "balance_loss_clip": 1.01497889,
      "balance_loss_mlp": 0.99937433,
      "epoch": 0.3809294775446402,
      "flos": 54969866380800.0,
      "grad_norm": 0.8370077518291396,
      "language_loss": 0.59312236,
      "learning_rate": 2.8415086921558774e-06,
      "loss": 0.61321318,
      "num_input_tokens_seen": 68227945,
      "step": 3168,
      "time_per_iteration": 3.525718927383423
    },
    {
      "auxiliary_loss_clip": 0.01137842,
      "auxiliary_loss_mlp": 0.01025148,
      "balance_loss_clip": 1.04266143,
      "balance_loss_mlp": 1.01776886,
      "epoch": 0.38104972043527924,
      "flos": 24643697904000.0,
      "grad_norm": 1.6110889269009732,
      "language_loss": 0.78616571,
      "learning_rate": 2.840801965042194e-06,
      "loss": 0.80779564,
      "num_input_tokens_seen": 68247405,
      "step": 3169,
      "time_per_iteration": 4.051271438598633
    },
    {
      "auxiliary_loss_clip": 0.01145332,
      "auxiliary_loss_mlp": 0.01025198,
      "balance_loss_clip": 1.04595494,
      "balance_loss_mlp": 1.01638222,
      "epoch": 0.38116996332591835,
      "flos": 22856783086080.0,
      "grad_norm": 1.791710454339517,
      "language_loss": 0.83780932,
      "learning_rate": 2.840095110379521e-06,
      "loss": 0.85951465,
      "num_input_tokens_seen": 68266925,
      "step": 3170,
      "time_per_iteration": 2.502591848373413
    },
    {
      "auxiliary_loss_clip": 0.01039021,
      "auxiliary_loss_mlp": 0.01001923,
      "balance_loss_clip": 1.01768005,
      "balance_loss_mlp": 1.00064707,
      "epoch": 0.38129020621655746,
      "flos": 60836160804480.0,
      "grad_norm": 0.7370650773994835,
      "language_loss": 0.53917062,
      "learning_rate": 2.8393881282750884e-06,
      "loss": 0.55958009,
      "num_input_tokens_seen": 68329755,
      "step": 3171,
      "time_per_iteration": 3.2006514072418213
    },
    {
      "auxiliary_loss_clip": 0.01152919,
      "auxiliary_loss_mlp": 0.01026307,
      "balance_loss_clip": 1.05122185,
      "balance_loss_mlp": 1.01796198,
      "epoch": 0.3814104491071965,
      "flos": 21648101408640.0,
      "grad_norm": 2.1167779888947877,
      "language_loss": 0.78833842,
      "learning_rate": 2.838681018836144e-06,
      "loss": 0.8101306,
      "num_input_tokens_seen": 68347075,
      "step": 3172,
      "time_per_iteration": 2.709188938140869
    },
    {
      "auxiliary_loss_clip": 0.01140165,
      "auxiliary_loss_mlp": 0.00762271,
      "balance_loss_clip": 1.04590583,
      "balance_loss_mlp": 1.00048661,
      "epoch": 0.3815306919978356,
      "flos": 19099090707840.0,
      "grad_norm": 3.1572789620936867,
      "language_loss": 0.78345037,
      "learning_rate": 2.837973782169955e-06,
      "loss": 0.80247474,
      "num_input_tokens_seen": 68365450,
      "step": 3173,
      "time_per_iteration": 2.5054147243499756
    },
    {
      "auxiliary_loss_clip": 0.01083746,
      "auxiliary_loss_mlp": 0.01003965,
      "balance_loss_clip": 1.01695681,
      "balance_loss_mlp": 1.00276124,
      "epoch": 0.38165093488847474,
      "flos": 67067918156160.0,
      "grad_norm": 0.8089844439337767,
      "language_loss": 0.59189057,
      "learning_rate": 2.8372664183838096e-06,
      "loss": 0.6127677,
      "num_input_tokens_seen": 68428470,
      "step": 3174,
      "time_per_iteration": 3.059325695037842
    },
    {
      "auxiliary_loss_clip": 0.01181734,
      "auxiliary_loss_mlp": 0.01028834,
      "balance_loss_clip": 1.05307579,
      "balance_loss_mlp": 1.02084374,
      "epoch": 0.3817711777791138,
      "flos": 22341105480960.0,
      "grad_norm": 2.2264495596923406,
      "language_loss": 0.68303359,
      "learning_rate": 2.836558927585015e-06,
      "loss": 0.70513928,
      "num_input_tokens_seen": 68445440,
      "step": 3175,
      "time_per_iteration": 2.4270942211151123
    },
    {
      "auxiliary_loss_clip": 0.01171557,
      "auxiliary_loss_mlp": 0.01032745,
      "balance_loss_clip": 1.05329132,
      "balance_loss_mlp": 1.02558327,
      "epoch": 0.3818914206697529,
      "flos": 22820621068800.0,
      "grad_norm": 1.8640719254165987,
      "language_loss": 0.82636273,
      "learning_rate": 2.8358513098808957e-06,
      "loss": 0.84840578,
      "num_input_tokens_seen": 68465755,
      "step": 3176,
      "time_per_iteration": 2.477747917175293
    },
    {
      "auxiliary_loss_clip": 0.01117358,
      "auxiliary_loss_mlp": 0.010266,
      "balance_loss_clip": 1.04605341,
      "balance_loss_mlp": 1.01851141,
      "epoch": 0.382011663560392,
      "flos": 24386074583040.0,
      "grad_norm": 1.7021520832148063,
      "language_loss": 0.76432312,
      "learning_rate": 2.835143565378798e-06,
      "loss": 0.78576267,
      "num_input_tokens_seen": 68486220,
      "step": 3177,
      "time_per_iteration": 2.5898873805999756
    },
    {
      "auxiliary_loss_clip": 0.01109025,
      "auxiliary_loss_mlp": 0.01021945,
      "balance_loss_clip": 1.04546857,
      "balance_loss_mlp": 1.01432776,
      "epoch": 0.38213190645103107,
      "flos": 21981568296960.0,
      "grad_norm": 2.1506874675296968,
      "language_loss": 0.78206319,
      "learning_rate": 2.8344356941860847e-06,
      "loss": 0.80337286,
      "num_input_tokens_seen": 68505850,
      "step": 3178,
      "time_per_iteration": 2.603647232055664
    },
    {
      "auxiliary_loss_clip": 0.01137521,
      "auxiliary_loss_mlp": 0.01028246,
      "balance_loss_clip": 1.05007863,
      "balance_loss_mlp": 1.02045584,
      "epoch": 0.3822521493416702,
      "flos": 35516945773440.0,
      "grad_norm": 2.206334388246954,
      "language_loss": 0.66234505,
      "learning_rate": 2.8337276964101403e-06,
      "loss": 0.6840027,
      "num_input_tokens_seen": 68526290,
      "step": 3179,
      "time_per_iteration": 2.696027994155884
    },
    {
      "auxiliary_loss_clip": 0.01169485,
      "auxiliary_loss_mlp": 0.01027547,
      "balance_loss_clip": 1.05123329,
      "balance_loss_mlp": 1.02000737,
      "epoch": 0.3823723922323093,
      "flos": 21069904181760.0,
      "grad_norm": 4.254980769317907,
      "language_loss": 0.7682572,
      "learning_rate": 2.833019572158367e-06,
      "loss": 0.79022747,
      "num_input_tokens_seen": 68544725,
      "step": 3180,
      "time_per_iteration": 2.488067388534546
    },
    {
      "auxiliary_loss_clip": 0.0115579,
      "auxiliary_loss_mlp": 0.01027538,
      "balance_loss_clip": 1.05235767,
      "balance_loss_mlp": 1.0198189,
      "epoch": 0.38249263512294834,
      "flos": 19789149864960.0,
      "grad_norm": 1.9127208185256865,
      "language_loss": 0.79979289,
      "learning_rate": 2.8323113215381872e-06,
      "loss": 0.82162619,
      "num_input_tokens_seen": 68563070,
      "step": 3181,
      "time_per_iteration": 2.5239553451538086
    },
    {
      "auxiliary_loss_clip": 0.01135962,
      "auxiliary_loss_mlp": 0.01031835,
      "balance_loss_clip": 1.04739428,
      "balance_loss_mlp": 1.02378821,
      "epoch": 0.38261287801358745,
      "flos": 21433930565760.0,
      "grad_norm": 3.3638053866364523,
      "language_loss": 0.76438922,
      "learning_rate": 2.831602944657042e-06,
      "loss": 0.78606719,
      "num_input_tokens_seen": 68581150,
      "step": 3182,
      "time_per_iteration": 2.53416109085083
    },
    {
      "auxiliary_loss_clip": 0.01161672,
      "auxiliary_loss_mlp": 0.01022724,
      "balance_loss_clip": 1.0515033,
      "balance_loss_mlp": 1.01535058,
      "epoch": 0.38273312090422656,
      "flos": 21981568296960.0,
      "grad_norm": 2.4542562068155216,
      "language_loss": 0.7424916,
      "learning_rate": 2.830894441622391e-06,
      "loss": 0.76433551,
      "num_input_tokens_seen": 68597800,
      "step": 3183,
      "time_per_iteration": 2.496521472930908
    },
    {
      "auxiliary_loss_clip": 0.01137753,
      "auxiliary_loss_mlp": 0.00762416,
      "balance_loss_clip": 1.04529953,
      "balance_loss_mlp": 1.00051236,
      "epoch": 0.3828533637948656,
      "flos": 24790895838720.0,
      "grad_norm": 2.063231703597195,
      "language_loss": 0.80283439,
      "learning_rate": 2.8301858125417134e-06,
      "loss": 0.82183611,
      "num_input_tokens_seen": 68617640,
      "step": 3184,
      "time_per_iteration": 2.565103769302368
    },
    {
      "auxiliary_loss_clip": 0.01155518,
      "auxiliary_loss_mlp": 0.01025726,
      "balance_loss_clip": 1.05390048,
      "balance_loss_mlp": 1.01862717,
      "epoch": 0.38297360668550473,
      "flos": 22455445449600.0,
      "grad_norm": 1.7793085337085333,
      "language_loss": 0.73788238,
      "learning_rate": 2.8294770575225082e-06,
      "loss": 0.75969481,
      "num_input_tokens_seen": 68637770,
      "step": 3185,
      "time_per_iteration": 2.5026817321777344
    },
    {
      "auxiliary_loss_clip": 0.01171711,
      "auxiliary_loss_mlp": 0.01030731,
      "balance_loss_clip": 1.05589783,
      "balance_loss_mlp": 1.02291048,
      "epoch": 0.3830938495761438,
      "flos": 24896903852160.0,
      "grad_norm": 1.8995471365824692,
      "language_loss": 0.84415257,
      "learning_rate": 2.828768176672293e-06,
      "loss": 0.86617702,
      "num_input_tokens_seen": 68656885,
      "step": 3186,
      "time_per_iteration": 2.499512195587158
    },
    {
      "auxiliary_loss_clip": 0.01137167,
      "auxiliary_loss_mlp": 0.01027959,
      "balance_loss_clip": 1.0474906,
      "balance_loss_mlp": 1.01999009,
      "epoch": 0.3832140924667829,
      "flos": 33036236784000.0,
      "grad_norm": 1.7281969888694617,
      "language_loss": 0.71628308,
      "learning_rate": 2.8280591700986044e-06,
      "loss": 0.73793435,
      "num_input_tokens_seen": 68678750,
      "step": 3187,
      "time_per_iteration": 2.6123199462890625
    },
    {
      "auxiliary_loss_clip": 0.01156838,
      "auxiliary_loss_mlp": 0.01030437,
      "balance_loss_clip": 1.04954779,
      "balance_loss_mlp": 1.02254581,
      "epoch": 0.383334335357422,
      "flos": 31903721896320.0,
      "grad_norm": 2.1807006866721266,
      "language_loss": 0.75257981,
      "learning_rate": 2.827350037908999e-06,
      "loss": 0.77445263,
      "num_input_tokens_seen": 68698190,
      "step": 3188,
      "time_per_iteration": 2.575436592102051
    },
    {
      "auxiliary_loss_clip": 0.01144218,
      "auxiliary_loss_mlp": 0.01031584,
      "balance_loss_clip": 1.04833341,
      "balance_loss_mlp": 1.02313757,
      "epoch": 0.38345457824806106,
      "flos": 19791915212160.0,
      "grad_norm": 2.4878051564816346,
      "language_loss": 0.7945323,
      "learning_rate": 2.8266407802110496e-06,
      "loss": 0.81629032,
      "num_input_tokens_seen": 68716445,
      "step": 3189,
      "time_per_iteration": 3.3202626705169678
    },
    {
      "auxiliary_loss_clip": 0.01102949,
      "auxiliary_loss_mlp": 0.01027768,
      "balance_loss_clip": 1.04342628,
      "balance_loss_mlp": 1.01929879,
      "epoch": 0.3835748211387002,
      "flos": 22419391173120.0,
      "grad_norm": 1.6348695623562643,
      "language_loss": 0.75552332,
      "learning_rate": 2.8259313971123515e-06,
      "loss": 0.77683049,
      "num_input_tokens_seen": 68737565,
      "step": 3190,
      "time_per_iteration": 2.631824493408203
    },
    {
      "auxiliary_loss_clip": 0.01165465,
      "auxiliary_loss_mlp": 0.01026559,
      "balance_loss_clip": 1.05310309,
      "balance_loss_mlp": 1.01922727,
      "epoch": 0.3836950640293393,
      "flos": 25118436983040.0,
      "grad_norm": 1.4647485065380128,
      "language_loss": 0.78014088,
      "learning_rate": 2.8252218887205166e-06,
      "loss": 0.80206108,
      "num_input_tokens_seen": 68758255,
      "step": 3191,
      "time_per_iteration": 2.5090203285217285
    },
    {
      "auxiliary_loss_clip": 0.01113985,
      "auxiliary_loss_mlp": 0.01027931,
      "balance_loss_clip": 1.04734313,
      "balance_loss_mlp": 1.02055228,
      "epoch": 0.38381530691997834,
      "flos": 21799213925760.0,
      "grad_norm": 1.7546562748941634,
      "language_loss": 0.8057124,
      "learning_rate": 2.824512255143178e-06,
      "loss": 0.82713151,
      "num_input_tokens_seen": 68777490,
      "step": 3192,
      "time_per_iteration": 2.5692050457000732
    },
    {
      "auxiliary_loss_clip": 0.011409,
      "auxiliary_loss_mlp": 0.01022626,
      "balance_loss_clip": 1.04771256,
      "balance_loss_mlp": 1.01530337,
      "epoch": 0.38393554981061745,
      "flos": 21252689516160.0,
      "grad_norm": 1.5942724656274099,
      "language_loss": 0.79068297,
      "learning_rate": 2.8238024964879855e-06,
      "loss": 0.81231821,
      "num_input_tokens_seen": 68798385,
      "step": 3193,
      "time_per_iteration": 3.372929811477661
    },
    {
      "auxiliary_loss_clip": 0.01184641,
      "auxiliary_loss_mlp": 0.01028911,
      "balance_loss_clip": 1.05431449,
      "balance_loss_mlp": 1.02031624,
      "epoch": 0.38405579270125656,
      "flos": 17019360218880.0,
      "grad_norm": 2.2270292550574275,
      "language_loss": 0.76967895,
      "learning_rate": 2.8230926128626095e-06,
      "loss": 0.79181457,
      "num_input_tokens_seen": 68816880,
      "step": 3194,
      "time_per_iteration": 2.4135050773620605
    },
    {
      "auxiliary_loss_clip": 0.01150216,
      "auxiliary_loss_mlp": 0.01030705,
      "balance_loss_clip": 1.04955673,
      "balance_loss_mlp": 1.02208662,
      "epoch": 0.3841760355918956,
      "flos": 21835375943040.0,
      "grad_norm": 1.9727650672589567,
      "language_loss": 0.79329073,
      "learning_rate": 2.822382604374738e-06,
      "loss": 0.8150999,
      "num_input_tokens_seen": 68835805,
      "step": 3195,
      "time_per_iteration": 2.4910168647766113
    },
    {
      "auxiliary_loss_clip": 0.01155221,
      "auxiliary_loss_mlp": 0.01033727,
      "balance_loss_clip": 1.05323637,
      "balance_loss_mlp": 1.02565098,
      "epoch": 0.3842962784825347,
      "flos": 25915114684800.0,
      "grad_norm": 2.0532166436378527,
      "language_loss": 0.65856123,
      "learning_rate": 2.8216724711320793e-06,
      "loss": 0.68045068,
      "num_input_tokens_seen": 68854930,
      "step": 3196,
      "time_per_iteration": 3.3022334575653076
    },
    {
      "auxiliary_loss_clip": 0.01181011,
      "auxiliary_loss_mlp": 0.00762096,
      "balance_loss_clip": 1.05351889,
      "balance_loss_mlp": 1.0004859,
      "epoch": 0.38441652137317384,
      "flos": 25337492075520.0,
      "grad_norm": 1.47877781050593,
      "language_loss": 0.79610753,
      "learning_rate": 2.820962213242361e-06,
      "loss": 0.81553864,
      "num_input_tokens_seen": 68874260,
      "step": 3197,
      "time_per_iteration": 2.4673094749450684
    },
    {
      "auxiliary_loss_clip": 0.01165806,
      "auxiliary_loss_mlp": 0.01030596,
      "balance_loss_clip": 1.05499542,
      "balance_loss_mlp": 1.02281749,
      "epoch": 0.3845367642638129,
      "flos": 18113486446080.0,
      "grad_norm": 2.0839808971588023,
      "language_loss": 0.84196889,
      "learning_rate": 2.8202518308133264e-06,
      "loss": 0.86393297,
      "num_input_tokens_seen": 68891535,
      "step": 3198,
      "time_per_iteration": 2.4500091075897217
    },
    {
      "auxiliary_loss_clip": 0.01183405,
      "auxiliary_loss_mlp": 0.010283,
      "balance_loss_clip": 1.05307496,
      "balance_loss_mlp": 1.01988339,
      "epoch": 0.384657007154452,
      "flos": 25228395492480.0,
      "grad_norm": 1.7723069316482476,
      "language_loss": 0.73268163,
      "learning_rate": 2.8195413239527426e-06,
      "loss": 0.75479865,
      "num_input_tokens_seen": 68911275,
      "step": 3199,
      "time_per_iteration": 2.4608614444732666
    },
    {
      "auxiliary_loss_clip": 0.01165938,
      "auxiliary_loss_mlp": 0.01025012,
      "balance_loss_clip": 1.05086684,
      "balance_loss_mlp": 1.01728714,
      "epoch": 0.38477725004509106,
      "flos": 19865855358720.0,
      "grad_norm": 2.984715420168236,
      "language_loss": 0.80649495,
      "learning_rate": 2.8188306927683906e-06,
      "loss": 0.82840443,
      "num_input_tokens_seen": 68930745,
      "step": 3200,
      "time_per_iteration": 2.479733467102051
    },
    {
      "auxiliary_loss_clip": 0.01157421,
      "auxiliary_loss_mlp": 0.01026386,
      "balance_loss_clip": 1.05219245,
      "balance_loss_mlp": 1.01882255,
      "epoch": 0.38489749293573017,
      "flos": 18259391491200.0,
      "grad_norm": 2.0538072666640517,
      "language_loss": 0.74708247,
      "learning_rate": 2.818119937368074e-06,
      "loss": 0.76892054,
      "num_input_tokens_seen": 68949380,
      "step": 3201,
      "time_per_iteration": 2.48122501373291
    },
    {
      "auxiliary_loss_clip": 0.01174571,
      "auxiliary_loss_mlp": 0.01026306,
      "balance_loss_clip": 1.05240822,
      "balance_loss_mlp": 1.01789594,
      "epoch": 0.3850177358263693,
      "flos": 24389163152640.0,
      "grad_norm": 1.7977164773743752,
      "language_loss": 0.6552316,
      "learning_rate": 2.817409057859613e-06,
      "loss": 0.67724037,
      "num_input_tokens_seen": 68968370,
      "step": 3202,
      "time_per_iteration": 2.495089530944824
    },
    {
      "auxiliary_loss_clip": 0.01119938,
      "auxiliary_loss_mlp": 0.01029223,
      "balance_loss_clip": 1.04628098,
      "balance_loss_mlp": 1.02043748,
      "epoch": 0.38513797871700833,
      "flos": 17671533505920.0,
      "grad_norm": 1.8553361074795838,
      "language_loss": 0.79098183,
      "learning_rate": 2.8166980543508482e-06,
      "loss": 0.81247342,
      "num_input_tokens_seen": 68984260,
      "step": 3203,
      "time_per_iteration": 2.5306320190429688
    },
    {
      "auxiliary_loss_clip": 0.01184912,
      "auxiliary_loss_mlp": 0.01027669,
      "balance_loss_clip": 1.05515623,
      "balance_loss_mlp": 1.0195688,
      "epoch": 0.38525822160764744,
      "flos": 25739583897600.0,
      "grad_norm": 2.2440143030083197,
      "language_loss": 0.79745102,
      "learning_rate": 2.815986926949638e-06,
      "loss": 0.81957686,
      "num_input_tokens_seen": 69002760,
      "step": 3204,
      "time_per_iteration": 2.4764270782470703
    },
    {
      "auxiliary_loss_clip": 0.01169411,
      "auxiliary_loss_mlp": 0.0102993,
      "balance_loss_clip": 1.05345321,
      "balance_loss_mlp": 1.02224088,
      "epoch": 0.38537846449828655,
      "flos": 20193647898240.0,
      "grad_norm": 1.9366310159940756,
      "language_loss": 0.80485642,
      "learning_rate": 2.8152756757638597e-06,
      "loss": 0.82684982,
      "num_input_tokens_seen": 69021260,
      "step": 3205,
      "time_per_iteration": 2.525256633758545
    },
    {
      "auxiliary_loss_clip": 0.01168349,
      "auxiliary_loss_mlp": 0.01025131,
      "balance_loss_clip": 1.05362892,
      "balance_loss_mlp": 1.01730502,
      "epoch": 0.3854987073889256,
      "flos": 23039352938880.0,
      "grad_norm": 1.957318819849449,
      "language_loss": 0.84805369,
      "learning_rate": 2.8145643009014093e-06,
      "loss": 0.86998856,
      "num_input_tokens_seen": 69039755,
      "step": 3206,
      "time_per_iteration": 2.469200611114502
    },
    {
      "auxiliary_loss_clip": 0.01169757,
      "auxiliary_loss_mlp": 0.01027836,
      "balance_loss_clip": 1.05223334,
      "balance_loss_mlp": 1.02062714,
      "epoch": 0.3856189502795647,
      "flos": 20190631155840.0,
      "grad_norm": 1.8727171650986616,
      "language_loss": 0.79482269,
      "learning_rate": 2.813852802470202e-06,
      "loss": 0.81679857,
      "num_input_tokens_seen": 69057650,
      "step": 3207,
      "time_per_iteration": 2.4493305683135986
    },
    {
      "auxiliary_loss_clip": 0.01149093,
      "auxiliary_loss_mlp": 0.01028307,
      "balance_loss_clip": 1.04925859,
      "balance_loss_mlp": 1.01988482,
      "epoch": 0.38573919317020383,
      "flos": 25702631781120.0,
      "grad_norm": 1.8564654416961834,
      "language_loss": 0.72477341,
      "learning_rate": 2.8131411805781717e-06,
      "loss": 0.74654734,
      "num_input_tokens_seen": 69077775,
      "step": 3208,
      "time_per_iteration": 2.52998685836792
    },
    {
      "auxiliary_loss_clip": 0.0115728,
      "auxiliary_loss_mlp": 0.0103279,
      "balance_loss_clip": 1.05320108,
      "balance_loss_mlp": 1.02411771,
      "epoch": 0.3858594360608429,
      "flos": 29821405628160.0,
      "grad_norm": 2.4154988743530725,
      "language_loss": 0.64251673,
      "learning_rate": 2.8124294353332707e-06,
      "loss": 0.66441751,
      "num_input_tokens_seen": 69096450,
      "step": 3209,
      "time_per_iteration": 2.5405120849609375
    },
    {
      "auxiliary_loss_clip": 0.01147448,
      "auxiliary_loss_mlp": 0.01029025,
      "balance_loss_clip": 1.05024803,
      "balance_loss_mlp": 1.02114272,
      "epoch": 0.385979678951482,
      "flos": 24790428961920.0,
      "grad_norm": 2.0734802543577624,
      "language_loss": 0.77002323,
      "learning_rate": 2.8117175668434713e-06,
      "loss": 0.79178798,
      "num_input_tokens_seen": 69116110,
      "step": 3210,
      "time_per_iteration": 2.569413185119629
    },
    {
      "auxiliary_loss_clip": 0.0118342,
      "auxiliary_loss_mlp": 0.01025331,
      "balance_loss_clip": 1.05286443,
      "balance_loss_mlp": 1.01746917,
      "epoch": 0.3860999218421211,
      "flos": 21287881866240.0,
      "grad_norm": 2.156595512260456,
      "language_loss": 0.70295525,
      "learning_rate": 2.811005575216762e-06,
      "loss": 0.72504276,
      "num_input_tokens_seen": 69134825,
      "step": 3211,
      "time_per_iteration": 2.426785469055176
    },
    {
      "auxiliary_loss_clip": 0.01136072,
      "auxiliary_loss_mlp": 0.01029479,
      "balance_loss_clip": 1.04940605,
      "balance_loss_mlp": 1.02164114,
      "epoch": 0.38622016473276016,
      "flos": 24536720223360.0,
      "grad_norm": 1.4879811042686577,
      "language_loss": 0.78962409,
      "learning_rate": 2.8102934605611513e-06,
      "loss": 0.81127954,
      "num_input_tokens_seen": 69156460,
      "step": 3212,
      "time_per_iteration": 2.630581855773926
    },
    {
      "auxiliary_loss_clip": 0.01162519,
      "auxiliary_loss_mlp": 0.01027802,
      "balance_loss_clip": 1.05406499,
      "balance_loss_mlp": 1.02010679,
      "epoch": 0.3863404076233993,
      "flos": 20558212986240.0,
      "grad_norm": 1.9334712343041998,
      "language_loss": 0.67580163,
      "learning_rate": 2.8095812229846665e-06,
      "loss": 0.69770491,
      "num_input_tokens_seen": 69176420,
      "step": 3213,
      "time_per_iteration": 2.4756698608398438
    },
    {
      "auxiliary_loss_clip": 0.01155649,
      "auxiliary_loss_mlp": 0.01027104,
      "balance_loss_clip": 1.05022776,
      "balance_loss_mlp": 1.01885474,
      "epoch": 0.3864606505140384,
      "flos": 22346277039360.0,
      "grad_norm": 2.2897727708591593,
      "language_loss": 0.69092166,
      "learning_rate": 2.808868862595355e-06,
      "loss": 0.71274924,
      "num_input_tokens_seen": 69196665,
      "step": 3214,
      "time_per_iteration": 2.5003111362457275
    },
    {
      "auxiliary_loss_clip": 0.01173008,
      "auxiliary_loss_mlp": 0.01029896,
      "balance_loss_clip": 1.05276656,
      "balance_loss_mlp": 1.02217674,
      "epoch": 0.38658089340467744,
      "flos": 25703601448320.0,
      "grad_norm": 1.794278604247436,
      "language_loss": 0.79630697,
      "learning_rate": 2.8081563795012795e-06,
      "loss": 0.81833601,
      "num_input_tokens_seen": 69216290,
      "step": 3215,
      "time_per_iteration": 2.505335569381714
    },
    {
      "auxiliary_loss_clip": 0.01162649,
      "auxiliary_loss_mlp": 0.01025337,
      "balance_loss_clip": 1.0503298,
      "balance_loss_mlp": 1.01706409,
      "epoch": 0.38670113629531655,
      "flos": 33802534558080.0,
      "grad_norm": 1.7089105332158925,
      "language_loss": 0.73840106,
      "learning_rate": 2.807443773810524e-06,
      "loss": 0.76028097,
      "num_input_tokens_seen": 69237550,
      "step": 3216,
      "time_per_iteration": 3.442344903945923
    },
    {
      "auxiliary_loss_clip": 0.01140815,
      "auxiliary_loss_mlp": 0.01029235,
      "balance_loss_clip": 1.0520426,
      "balance_loss_mlp": 1.02167737,
      "epoch": 0.3868213791859556,
      "flos": 23331522165120.0,
      "grad_norm": 1.7494564552128187,
      "language_loss": 0.89491987,
      "learning_rate": 2.80673104563119e-06,
      "loss": 0.91662037,
      "num_input_tokens_seen": 69258175,
      "step": 3217,
      "time_per_iteration": 2.5437746047973633
    },
    {
      "auxiliary_loss_clip": 0.01167748,
      "auxiliary_loss_mlp": 0.01026458,
      "balance_loss_clip": 1.05380428,
      "balance_loss_mlp": 1.01898932,
      "epoch": 0.3869416220765947,
      "flos": 18441530380800.0,
      "grad_norm": 1.8935414531063413,
      "language_loss": 0.78779399,
      "learning_rate": 2.8060181950713976e-06,
      "loss": 0.80973607,
      "num_input_tokens_seen": 69274965,
      "step": 3218,
      "time_per_iteration": 2.445941925048828
    },
    {
      "auxiliary_loss_clip": 0.01140586,
      "auxiliary_loss_mlp": 0.01030357,
      "balance_loss_clip": 1.04957235,
      "balance_loss_mlp": 1.02179217,
      "epoch": 0.3870618649672338,
      "flos": 15632992938240.0,
      "grad_norm": 2.0349587359910974,
      "language_loss": 0.80779105,
      "learning_rate": 2.805305222239286e-06,
      "loss": 0.82950044,
      "num_input_tokens_seen": 69292220,
      "step": 3219,
      "time_per_iteration": 2.5065672397613525
    },
    {
      "auxiliary_loss_clip": 0.01152443,
      "auxiliary_loss_mlp": 0.01032107,
      "balance_loss_clip": 1.05142903,
      "balance_loss_mlp": 1.02383995,
      "epoch": 0.3871821078578729,
      "flos": 23513804709120.0,
      "grad_norm": 1.8508080525203485,
      "language_loss": 0.74011767,
      "learning_rate": 2.8045921272430118e-06,
      "loss": 0.76196313,
      "num_input_tokens_seen": 69311900,
      "step": 3220,
      "time_per_iteration": 3.388962984085083
    },
    {
      "auxiliary_loss_clip": 0.01177741,
      "auxiliary_loss_mlp": 0.01029053,
      "balance_loss_clip": 1.05307615,
      "balance_loss_mlp": 1.02063656,
      "epoch": 0.387302350748512,
      "flos": 17778259791360.0,
      "grad_norm": 2.8917845533421893,
      "language_loss": 0.7632134,
      "learning_rate": 2.803878910190753e-06,
      "loss": 0.7852813,
      "num_input_tokens_seen": 69328820,
      "step": 3221,
      "time_per_iteration": 2.4334919452667236
    },
    {
      "auxiliary_loss_clip": 0.01173599,
      "auxiliary_loss_mlp": 0.01028342,
      "balance_loss_clip": 1.0524317,
      "balance_loss_mlp": 1.02012825,
      "epoch": 0.3874225936391511,
      "flos": 11503409097600.0,
      "grad_norm": 2.2293242821823527,
      "language_loss": 0.82540745,
      "learning_rate": 2.8031655711907017e-06,
      "loss": 0.84742689,
      "num_input_tokens_seen": 69342525,
      "step": 3222,
      "time_per_iteration": 3.234394073486328
    },
    {
      "auxiliary_loss_clip": 0.01176365,
      "auxiliary_loss_mlp": 0.01028772,
      "balance_loss_clip": 1.05663884,
      "balance_loss_mlp": 1.02091599,
      "epoch": 0.38754283652979016,
      "flos": 21945154884480.0,
      "grad_norm": 2.0200332788840996,
      "language_loss": 0.80665362,
      "learning_rate": 2.8024521103510723e-06,
      "loss": 0.82870495,
      "num_input_tokens_seen": 69359295,
      "step": 3223,
      "time_per_iteration": 3.2388243675231934
    },
    {
      "auxiliary_loss_clip": 0.01168977,
      "auxiliary_loss_mlp": 0.01026865,
      "balance_loss_clip": 1.05062079,
      "balance_loss_mlp": 1.01933742,
      "epoch": 0.38766307942042927,
      "flos": 21175984022400.0,
      "grad_norm": 1.6623459373167608,
      "language_loss": 0.74894947,
      "learning_rate": 2.8017385277800952e-06,
      "loss": 0.77090788,
      "num_input_tokens_seen": 69377650,
      "step": 3224,
      "time_per_iteration": 2.4779410362243652
    },
    {
      "auxiliary_loss_clip": 0.01147191,
      "auxiliary_loss_mlp": 0.01030838,
      "balance_loss_clip": 1.05246294,
      "balance_loss_mlp": 1.02270842,
      "epoch": 0.3877833223110684,
      "flos": 27417294391680.0,
      "grad_norm": 2.1022691550027806,
      "language_loss": 0.74999893,
      "learning_rate": 2.8010248235860213e-06,
      "loss": 0.77177918,
      "num_input_tokens_seen": 69397765,
      "step": 3225,
      "time_per_iteration": 2.5907249450683594
    },
    {
      "auxiliary_loss_clip": 0.01064901,
      "auxiliary_loss_mlp": 0.00753871,
      "balance_loss_clip": 1.01622009,
      "balance_loss_mlp": 1.00104749,
      "epoch": 0.38790356520170743,
      "flos": 64500019879680.0,
      "grad_norm": 0.8316356263695989,
      "language_loss": 0.62737465,
      "learning_rate": 2.8003109978771192e-06,
      "loss": 0.64556241,
      "num_input_tokens_seen": 69458930,
      "step": 3226,
      "time_per_iteration": 3.1397085189819336
    },
    {
      "auxiliary_loss_clip": 0.01132178,
      "auxiliary_loss_mlp": 0.01028599,
      "balance_loss_clip": 1.04446375,
      "balance_loss_mlp": 1.02046585,
      "epoch": 0.38802380809234654,
      "flos": 22345415112960.0,
      "grad_norm": 1.9025751700006426,
      "language_loss": 0.78832626,
      "learning_rate": 2.799597050761674e-06,
      "loss": 0.80993408,
      "num_input_tokens_seen": 69475135,
      "step": 3227,
      "time_per_iteration": 2.5487070083618164
    },
    {
      "auxiliary_loss_clip": 0.01185842,
      "auxiliary_loss_mlp": 0.0102982,
      "balance_loss_clip": 1.0546875,
      "balance_loss_mlp": 1.02154112,
      "epoch": 0.38814405098298566,
      "flos": 25261361199360.0,
      "grad_norm": 2.2978969370769353,
      "language_loss": 0.78845787,
      "learning_rate": 2.7988829823479924e-06,
      "loss": 0.81061447,
      "num_input_tokens_seen": 69493525,
      "step": 3228,
      "time_per_iteration": 2.4963016510009766
    },
    {
      "auxiliary_loss_clip": 0.01151832,
      "auxiliary_loss_mlp": 0.01034917,
      "balance_loss_clip": 1.04949355,
      "balance_loss_mlp": 1.02595854,
      "epoch": 0.3882642938736247,
      "flos": 18841180078080.0,
      "grad_norm": 1.8006423740272943,
      "language_loss": 0.64153314,
      "learning_rate": 2.7981687927443976e-06,
      "loss": 0.66340065,
      "num_input_tokens_seen": 69510325,
      "step": 3229,
      "time_per_iteration": 2.518388509750366
    },
    {
      "auxiliary_loss_clip": 0.01170466,
      "auxiliary_loss_mlp": 0.01024373,
      "balance_loss_clip": 1.05142999,
      "balance_loss_mlp": 1.01699448,
      "epoch": 0.3883845367642638,
      "flos": 21652806090240.0,
      "grad_norm": 2.1126366535360583,
      "language_loss": 0.85848439,
      "learning_rate": 2.797454482059231e-06,
      "loss": 0.88043278,
      "num_input_tokens_seen": 69530480,
      "step": 3230,
      "time_per_iteration": 2.4896163940429688
    },
    {
      "auxiliary_loss_clip": 0.01188312,
      "auxiliary_loss_mlp": 0.01022747,
      "balance_loss_clip": 1.05571604,
      "balance_loss_mlp": 1.0150522,
      "epoch": 0.3885047796549029,
      "flos": 20557530627840.0,
      "grad_norm": 1.6989963090614713,
      "language_loss": 0.84420633,
      "learning_rate": 2.7967400504008537e-06,
      "loss": 0.86631697,
      "num_input_tokens_seen": 69549780,
      "step": 3231,
      "time_per_iteration": 2.447850227355957
    },
    {
      "auxiliary_loss_clip": 0.01037035,
      "auxiliary_loss_mlp": 0.01001894,
      "balance_loss_clip": 1.01485324,
      "balance_loss_mlp": 1.00045156,
      "epoch": 0.388625022545542,
      "flos": 64325491695360.0,
      "grad_norm": 0.8022526018366798,
      "language_loss": 0.57465744,
      "learning_rate": 2.7960254978776456e-06,
      "loss": 0.59504676,
      "num_input_tokens_seen": 69611870,
      "step": 3232,
      "time_per_iteration": 3.1105988025665283
    },
    {
      "auxiliary_loss_clip": 0.01192168,
      "auxiliary_loss_mlp": 0.01036197,
      "balance_loss_clip": 1.05890322,
      "balance_loss_mlp": 1.02764356,
      "epoch": 0.3887452654361811,
      "flos": 18113881495680.0,
      "grad_norm": 2.205559368003149,
      "language_loss": 0.81482327,
      "learning_rate": 2.7953108245980006e-06,
      "loss": 0.83710694,
      "num_input_tokens_seen": 69630385,
      "step": 3233,
      "time_per_iteration": 2.427133321762085
    },
    {
      "auxiliary_loss_clip": 0.01150324,
      "auxiliary_loss_mlp": 0.01028908,
      "balance_loss_clip": 1.05208707,
      "balance_loss_mlp": 1.02127862,
      "epoch": 0.38886550832682015,
      "flos": 24975261371520.0,
      "grad_norm": 1.732846160751345,
      "language_loss": 0.73512185,
      "learning_rate": 2.7945960306703365e-06,
      "loss": 0.75691414,
      "num_input_tokens_seen": 69653370,
      "step": 3234,
      "time_per_iteration": 2.526604652404785
    },
    {
      "auxiliary_loss_clip": 0.0117497,
      "auxiliary_loss_mlp": 0.01028342,
      "balance_loss_clip": 1.05309224,
      "balance_loss_mlp": 1.02005076,
      "epoch": 0.38898575121745926,
      "flos": 27199496275200.0,
      "grad_norm": 1.588899129368604,
      "language_loss": 0.65811324,
      "learning_rate": 2.7938811162030865e-06,
      "loss": 0.6801464,
      "num_input_tokens_seen": 69673635,
      "step": 3235,
      "time_per_iteration": 2.509488105773926
    },
    {
      "auxiliary_loss_clip": 0.01170715,
      "auxiliary_loss_mlp": 0.01032393,
      "balance_loss_clip": 1.05484247,
      "balance_loss_mlp": 1.02483487,
      "epoch": 0.3891059941080984,
      "flos": 28763728727040.0,
      "grad_norm": 1.7277976198209395,
      "language_loss": 0.82470143,
      "learning_rate": 2.793166081304702e-06,
      "loss": 0.84673244,
      "num_input_tokens_seen": 69694130,
      "step": 3236,
      "time_per_iteration": 2.5227341651916504
    },
    {
      "auxiliary_loss_clip": 0.01147911,
      "auxiliary_loss_mlp": 0.01031027,
      "balance_loss_clip": 1.04874611,
      "balance_loss_mlp": 1.02245021,
      "epoch": 0.38922623699873743,
      "flos": 22893447893760.0,
      "grad_norm": 2.0276868238012424,
      "language_loss": 0.8219837,
      "learning_rate": 2.7924509260836543e-06,
      "loss": 0.84377313,
      "num_input_tokens_seen": 69713255,
      "step": 3237,
      "time_per_iteration": 2.543348550796509
    },
    {
      "auxiliary_loss_clip": 0.0114166,
      "auxiliary_loss_mlp": 0.01025844,
      "balance_loss_clip": 1.04836845,
      "balance_loss_mlp": 1.01775575,
      "epoch": 0.38934647988937654,
      "flos": 19792418002560.0,
      "grad_norm": 2.050684782806791,
      "language_loss": 0.68545651,
      "learning_rate": 2.791735650648431e-06,
      "loss": 0.70713162,
      "num_input_tokens_seen": 69732375,
      "step": 3238,
      "time_per_iteration": 2.5303473472595215
    },
    {
      "auxiliary_loss_clip": 0.01155465,
      "auxiliary_loss_mlp": 0.01027622,
      "balance_loss_clip": 1.05193543,
      "balance_loss_mlp": 1.01992679,
      "epoch": 0.38946672278001565,
      "flos": 19202081978880.0,
      "grad_norm": 1.950539965555811,
      "language_loss": 0.74656129,
      "learning_rate": 2.791020255107538e-06,
      "loss": 0.76839209,
      "num_input_tokens_seen": 69749745,
      "step": 3239,
      "time_per_iteration": 2.480377674102783
    },
    {
      "auxiliary_loss_clip": 0.01137207,
      "auxiliary_loss_mlp": 0.01023849,
      "balance_loss_clip": 1.04646337,
      "balance_loss_mlp": 1.01620817,
      "epoch": 0.3895869656706547,
      "flos": 24936477661440.0,
      "grad_norm": 1.5719149804481403,
      "language_loss": 0.80506647,
      "learning_rate": 2.7903047395695023e-06,
      "loss": 0.82667708,
      "num_input_tokens_seen": 69769645,
      "step": 3240,
      "time_per_iteration": 2.578160285949707
    },
    {
      "auxiliary_loss_clip": 0.01171727,
      "auxiliary_loss_mlp": 0.00763544,
      "balance_loss_clip": 1.05551887,
      "balance_loss_mlp": 1.00050211,
      "epoch": 0.3897072085612938,
      "flos": 24133622820480.0,
      "grad_norm": 2.2504149998124725,
      "language_loss": 0.90364242,
      "learning_rate": 2.789589104142865e-06,
      "loss": 0.92299509,
      "num_input_tokens_seen": 69787270,
      "step": 3241,
      "time_per_iteration": 2.501742362976074
    },
    {
      "auxiliary_loss_clip": 0.01147688,
      "auxiliary_loss_mlp": 0.01030536,
      "balance_loss_clip": 1.05282593,
      "balance_loss_mlp": 1.02234602,
      "epoch": 0.3898274514519329,
      "flos": 17166342672000.0,
      "grad_norm": 1.636412462729543,
      "language_loss": 0.76452243,
      "learning_rate": 2.7888733489361895e-06,
      "loss": 0.78630471,
      "num_input_tokens_seen": 69805685,
      "step": 3242,
      "time_per_iteration": 3.352848768234253
    },
    {
      "auxiliary_loss_clip": 0.01082728,
      "auxiliary_loss_mlp": 0.0100187,
      "balance_loss_clip": 1.01639056,
      "balance_loss_mlp": 1.0006541,
      "epoch": 0.389947694342572,
      "flos": 66074807952000.0,
      "grad_norm": 0.7385728826113218,
      "language_loss": 0.58735919,
      "learning_rate": 2.788157474058054e-06,
      "loss": 0.6082052,
      "num_input_tokens_seen": 69867960,
      "step": 3243,
      "time_per_iteration": 3.1290664672851562
    },
    {
      "auxiliary_loss_clip": 0.01183456,
      "auxiliary_loss_mlp": 0.01025089,
      "balance_loss_clip": 1.05488944,
      "balance_loss_mlp": 1.01723337,
      "epoch": 0.3900679372332111,
      "flos": 25740912700800.0,
      "grad_norm": 2.033501168117817,
      "language_loss": 0.69751221,
      "learning_rate": 2.7874414796170555e-06,
      "loss": 0.7195977,
      "num_input_tokens_seen": 69889450,
      "step": 3244,
      "time_per_iteration": 2.499852180480957
    },
    {
      "auxiliary_loss_clip": 0.0116801,
      "auxiliary_loss_mlp": 0.01030488,
      "balance_loss_clip": 1.0513339,
      "balance_loss_mlp": 1.02149379,
      "epoch": 0.3901881801238502,
      "flos": 11801611808640.0,
      "grad_norm": 2.7595706859646376,
      "language_loss": 0.83891404,
      "learning_rate": 2.7867253657218113e-06,
      "loss": 0.86089897,
      "num_input_tokens_seen": 69903340,
      "step": 3245,
      "time_per_iteration": 2.443608045578003
    },
    {
      "auxiliary_loss_clip": 0.01153751,
      "auxiliary_loss_mlp": 0.0076292,
      "balance_loss_clip": 1.04865181,
      "balance_loss_mlp": 1.00037992,
      "epoch": 0.39030842301448926,
      "flos": 27308951994240.0,
      "grad_norm": 1.6194609997213272,
      "language_loss": 0.73074603,
      "learning_rate": 2.7860091324809544e-06,
      "loss": 0.74991274,
      "num_input_tokens_seen": 69924400,
      "step": 3246,
      "time_per_iteration": 3.4300479888916016
    },
    {
      "auxiliary_loss_clip": 0.01169921,
      "auxiliary_loss_mlp": 0.01024381,
      "balance_loss_clip": 1.05628026,
      "balance_loss_mlp": 1.01642418,
      "epoch": 0.39042866590512837,
      "flos": 27163334257920.0,
      "grad_norm": 1.6558130652737997,
      "language_loss": 0.81046462,
      "learning_rate": 2.7852927800031377e-06,
      "loss": 0.83240759,
      "num_input_tokens_seen": 69944565,
      "step": 3247,
      "time_per_iteration": 2.5320394039154053
    },
    {
      "auxiliary_loss_clip": 0.01160907,
      "auxiliary_loss_mlp": 0.01026287,
      "balance_loss_clip": 1.05256021,
      "balance_loss_mlp": 1.0188725,
      "epoch": 0.3905489087957674,
      "flos": 29716115886720.0,
      "grad_norm": 1.6210112064440096,
      "language_loss": 0.82835734,
      "learning_rate": 2.7845763083970298e-06,
      "loss": 0.85022926,
      "num_input_tokens_seen": 69964965,
      "step": 3248,
      "time_per_iteration": 2.5642549991607666
    },
    {
      "auxiliary_loss_clip": 0.01162749,
      "auxiliary_loss_mlp": 0.01026318,
      "balance_loss_clip": 1.05101204,
      "balance_loss_mlp": 1.01803303,
      "epoch": 0.39066915168640653,
      "flos": 24498618871680.0,
      "grad_norm": 1.7764446345116056,
      "language_loss": 0.81763041,
      "learning_rate": 2.7838597177713205e-06,
      "loss": 0.83952105,
      "num_input_tokens_seen": 69986055,
      "step": 3249,
      "time_per_iteration": 3.9981529712677
    },
    {
      "auxiliary_loss_clip": 0.01103956,
      "auxiliary_loss_mlp": 0.01033713,
      "balance_loss_clip": 1.04911184,
      "balance_loss_mlp": 1.02456367,
      "epoch": 0.39078939457704565,
      "flos": 20558572122240.0,
      "grad_norm": 1.6756429597870666,
      "language_loss": 0.73753262,
      "learning_rate": 2.7831430082347143e-06,
      "loss": 0.75890934,
      "num_input_tokens_seen": 70005260,
      "step": 3250,
      "time_per_iteration": 2.58373761177063
    },
    {
      "auxiliary_loss_clip": 0.0117313,
      "auxiliary_loss_mlp": 0.0076195,
      "balance_loss_clip": 1.05479407,
      "balance_loss_mlp": 1.0004319,
      "epoch": 0.3909096374676847,
      "flos": 22783417557120.0,
      "grad_norm": 2.2673661345969656,
      "language_loss": 0.82241976,
      "learning_rate": 2.7824261798959373e-06,
      "loss": 0.84177053,
      "num_input_tokens_seen": 70023440,
      "step": 3251,
      "time_per_iteration": 2.4760334491729736
    },
    {
      "auxiliary_loss_clip": 0.01159385,
      "auxiliary_loss_mlp": 0.01031265,
      "balance_loss_clip": 1.04956758,
      "balance_loss_mlp": 1.02312326,
      "epoch": 0.3910298803583238,
      "flos": 23003119094400.0,
      "grad_norm": 3.302233109838475,
      "language_loss": 0.79711246,
      "learning_rate": 2.78170923286373e-06,
      "loss": 0.81901896,
      "num_input_tokens_seen": 70043040,
      "step": 3252,
      "time_per_iteration": 2.503570795059204
    },
    {
      "auxiliary_loss_clip": 0.01095594,
      "auxiliary_loss_mlp": 0.01029322,
      "balance_loss_clip": 1.04935503,
      "balance_loss_mlp": 1.02100134,
      "epoch": 0.3911501232489629,
      "flos": 24316264500480.0,
      "grad_norm": 2.524476181282204,
      "language_loss": 0.84054238,
      "learning_rate": 2.780992167246854e-06,
      "loss": 0.86179155,
      "num_input_tokens_seen": 70060565,
      "step": 3253,
      "time_per_iteration": 2.6338977813720703
    },
    {
      "auxiliary_loss_clip": 0.01063074,
      "auxiliary_loss_mlp": 0.01000957,
      "balance_loss_clip": 1.01545167,
      "balance_loss_mlp": 0.99963409,
      "epoch": 0.391270366139602,
      "flos": 60869054684160.0,
      "grad_norm": 0.9770640126092774,
      "language_loss": 0.7217567,
      "learning_rate": 2.7802749831540883e-06,
      "loss": 0.74239695,
      "num_input_tokens_seen": 70119465,
      "step": 3254,
      "time_per_iteration": 3.1400845050811768
    },
    {
      "auxiliary_loss_clip": 0.01135714,
      "auxiliary_loss_mlp": 0.01024722,
      "balance_loss_clip": 1.05231929,
      "balance_loss_mlp": 1.01810563,
      "epoch": 0.3913906090302411,
      "flos": 21543494025600.0,
      "grad_norm": 1.8329094666211643,
      "language_loss": 0.81702727,
      "learning_rate": 2.7795576806942268e-06,
      "loss": 0.83863163,
      "num_input_tokens_seen": 70138270,
      "step": 3255,
      "time_per_iteration": 2.5657639503479004
    },
    {
      "auxiliary_loss_clip": 0.01065919,
      "auxiliary_loss_mlp": 0.01015118,
      "balance_loss_clip": 1.0305686,
      "balance_loss_mlp": 1.01323438,
      "epoch": 0.3915108519208802,
      "flos": 49839953702400.0,
      "grad_norm": 0.7636547175130955,
      "language_loss": 0.54911411,
      "learning_rate": 2.778840259976085e-06,
      "loss": 0.56992459,
      "num_input_tokens_seen": 70193500,
      "step": 3256,
      "time_per_iteration": 3.0667998790740967
    },
    {
      "auxiliary_loss_clip": 0.01172861,
      "auxiliary_loss_mlp": 0.01029573,
      "balance_loss_clip": 1.05391896,
      "balance_loss_mlp": 1.02144861,
      "epoch": 0.39163109481151925,
      "flos": 16506447960960.0,
      "grad_norm": 2.222199025902046,
      "language_loss": 0.76880467,
      "learning_rate": 2.778122721108495e-06,
      "loss": 0.79082906,
      "num_input_tokens_seen": 70211730,
      "step": 3257,
      "time_per_iteration": 2.4855129718780518
    },
    {
      "auxiliary_loss_clip": 0.01169594,
      "auxiliary_loss_mlp": 0.01025829,
      "balance_loss_clip": 1.05525732,
      "balance_loss_mlp": 1.01809812,
      "epoch": 0.39175133770215836,
      "flos": 26067484177920.0,
      "grad_norm": 1.864806882877144,
      "language_loss": 0.88906801,
      "learning_rate": 2.7774050642003076e-06,
      "loss": 0.91102219,
      "num_input_tokens_seen": 70232540,
      "step": 3258,
      "time_per_iteration": 2.519649028778076
    },
    {
      "auxiliary_loss_clip": 0.01189462,
      "auxiliary_loss_mlp": 0.01031847,
      "balance_loss_clip": 1.05702865,
      "balance_loss_mlp": 1.02312708,
      "epoch": 0.3918715805927975,
      "flos": 21872076664320.0,
      "grad_norm": 2.824304961261037,
      "language_loss": 0.93794262,
      "learning_rate": 2.7766872893603896e-06,
      "loss": 0.96015573,
      "num_input_tokens_seen": 70252515,
      "step": 3259,
      "time_per_iteration": 2.465919017791748
    },
    {
      "auxiliary_loss_clip": 0.01170958,
      "auxiliary_loss_mlp": 0.01034205,
      "balance_loss_clip": 1.05340433,
      "balance_loss_mlp": 1.02673042,
      "epoch": 0.39199182348343653,
      "flos": 20376181837440.0,
      "grad_norm": 1.8652807224471422,
      "language_loss": 0.72857916,
      "learning_rate": 2.7759693966976275e-06,
      "loss": 0.7506308,
      "num_input_tokens_seen": 70271020,
      "step": 3260,
      "time_per_iteration": 2.4618334770202637
    },
    {
      "auxiliary_loss_clip": 0.01139869,
      "auxiliary_loss_mlp": 0.01029917,
      "balance_loss_clip": 1.04971039,
      "balance_loss_mlp": 1.02138758,
      "epoch": 0.39211206637407564,
      "flos": 21683545153920.0,
      "grad_norm": 1.8205055729924071,
      "language_loss": 0.85115933,
      "learning_rate": 2.7752513863209242e-06,
      "loss": 0.87285715,
      "num_input_tokens_seen": 70289600,
      "step": 3261,
      "time_per_iteration": 2.5293543338775635
    },
    {
      "auxiliary_loss_clip": 0.01151067,
      "auxiliary_loss_mlp": 0.00762549,
      "balance_loss_clip": 1.05329037,
      "balance_loss_mlp": 1.00041938,
      "epoch": 0.39223230926471475,
      "flos": 21066276908160.0,
      "grad_norm": 1.8292048440569295,
      "language_loss": 0.84580135,
      "learning_rate": 2.774533258339203e-06,
      "loss": 0.86493748,
      "num_input_tokens_seen": 70307060,
      "step": 3262,
      "time_per_iteration": 2.505244493484497
    },
    {
      "auxiliary_loss_clip": 0.01127517,
      "auxiliary_loss_mlp": 0.01032921,
      "balance_loss_clip": 1.04422212,
      "balance_loss_mlp": 1.02389729,
      "epoch": 0.3923525521553538,
      "flos": 17603016312960.0,
      "grad_norm": 2.1113699509031942,
      "language_loss": 0.7951684,
      "learning_rate": 2.7738150128614014e-06,
      "loss": 0.81677282,
      "num_input_tokens_seen": 70324465,
      "step": 3263,
      "time_per_iteration": 2.55241322517395
    },
    {
      "auxiliary_loss_clip": 0.01136683,
      "auxiliary_loss_mlp": 0.01032606,
      "balance_loss_clip": 1.05254436,
      "balance_loss_mlp": 1.02444577,
      "epoch": 0.3924727950459929,
      "flos": 20558284813440.0,
      "grad_norm": 1.6557518437488754,
      "language_loss": 0.89292258,
      "learning_rate": 2.7730966499964777e-06,
      "loss": 0.91461545,
      "num_input_tokens_seen": 70341415,
      "step": 3264,
      "time_per_iteration": 2.530471086502075
    },
    {
      "auxiliary_loss_clip": 0.01189101,
      "auxiliary_loss_mlp": 0.01032901,
      "balance_loss_clip": 1.05687809,
      "balance_loss_mlp": 1.02475333,
      "epoch": 0.39259303793663197,
      "flos": 16216110328320.0,
      "grad_norm": 2.5280735980282953,
      "language_loss": 0.80623281,
      "learning_rate": 2.772378169853408e-06,
      "loss": 0.82845283,
      "num_input_tokens_seen": 70358985,
      "step": 3265,
      "time_per_iteration": 2.406672716140747
    },
    {
      "auxiliary_loss_clip": 0.01143242,
      "auxiliary_loss_mlp": 0.01029589,
      "balance_loss_clip": 1.05250764,
      "balance_loss_mlp": 1.02130437,
      "epoch": 0.3927132808272711,
      "flos": 16797001075200.0,
      "grad_norm": 1.6516614693086602,
      "language_loss": 0.74094671,
      "learning_rate": 2.771659572541183e-06,
      "loss": 0.76267505,
      "num_input_tokens_seen": 70376915,
      "step": 3266,
      "time_per_iteration": 2.5126864910125732
    },
    {
      "auxiliary_loss_clip": 0.01176372,
      "auxiliary_loss_mlp": 0.01026713,
      "balance_loss_clip": 1.05828738,
      "balance_loss_mlp": 1.01848209,
      "epoch": 0.3928335237179102,
      "flos": 20267228908800.0,
      "grad_norm": 3.2506458650837264,
      "language_loss": 0.86854154,
      "learning_rate": 2.7709408581688143e-06,
      "loss": 0.89057237,
      "num_input_tokens_seen": 70396900,
      "step": 3267,
      "time_per_iteration": 2.467857599258423
    },
    {
      "auxiliary_loss_clip": 0.01153162,
      "auxiliary_loss_mlp": 0.01032149,
      "balance_loss_clip": 1.05427825,
      "balance_loss_mlp": 1.02444792,
      "epoch": 0.39295376660854925,
      "flos": 24973250209920.0,
      "grad_norm": 2.6326070050291,
      "language_loss": 0.8742345,
      "learning_rate": 2.7702220268453307e-06,
      "loss": 0.89608765,
      "num_input_tokens_seen": 70417260,
      "step": 3268,
      "time_per_iteration": 2.5738706588745117
    },
    {
      "auxiliary_loss_clip": 0.01158392,
      "auxiliary_loss_mlp": 0.01030681,
      "balance_loss_clip": 1.05179596,
      "balance_loss_mlp": 1.02229166,
      "epoch": 0.39307400949918836,
      "flos": 18697788984960.0,
      "grad_norm": 1.9635488571225173,
      "language_loss": 0.85025442,
      "learning_rate": 2.7695030786797785e-06,
      "loss": 0.87214512,
      "num_input_tokens_seen": 70433155,
      "step": 3269,
      "time_per_iteration": 3.3012287616729736
    },
    {
      "auxiliary_loss_clip": 0.01123626,
      "auxiliary_loss_mlp": 0.0102671,
      "balance_loss_clip": 1.04863238,
      "balance_loss_mlp": 1.01891422,
      "epoch": 0.39319425238982747,
      "flos": 22415476590720.0,
      "grad_norm": 2.439163504877315,
      "language_loss": 0.74610567,
      "learning_rate": 2.7687840137812206e-06,
      "loss": 0.76760912,
      "num_input_tokens_seen": 70451240,
      "step": 3270,
      "time_per_iteration": 2.560053825378418
    },
    {
      "auxiliary_loss_clip": 0.01066677,
      "auxiliary_loss_mlp": 0.01009795,
      "balance_loss_clip": 1.01438951,
      "balance_loss_mlp": 1.00873423,
      "epoch": 0.3933144952804665,
      "flos": 66192954762240.0,
      "grad_norm": 0.8055337839960424,
      "language_loss": 0.62112021,
      "learning_rate": 2.7680648322587395e-06,
      "loss": 0.64188492,
      "num_input_tokens_seen": 70516115,
      "step": 3271,
      "time_per_iteration": 3.0788683891296387
    },
    {
      "auxiliary_loss_clip": 0.01184828,
      "auxiliary_loss_mlp": 0.01025247,
      "balance_loss_clip": 1.05506277,
      "balance_loss_mlp": 1.01798701,
      "epoch": 0.39343473817110564,
      "flos": 15487159720320.0,
      "grad_norm": 2.014715414844573,
      "language_loss": 0.81232679,
      "learning_rate": 2.7673455342214334e-06,
      "loss": 0.83442754,
      "num_input_tokens_seen": 70533105,
      "step": 3272,
      "time_per_iteration": 3.252774238586426
    },
    {
      "auxiliary_loss_clip": 0.01171341,
      "auxiliary_loss_mlp": 0.01029245,
      "balance_loss_clip": 1.05465829,
      "balance_loss_mlp": 1.021842,
      "epoch": 0.39355498106174475,
      "flos": 21324905809920.0,
      "grad_norm": 1.7149714659231003,
      "language_loss": 0.76088572,
      "learning_rate": 2.7666261197784198e-06,
      "loss": 0.78289151,
      "num_input_tokens_seen": 70551920,
      "step": 3273,
      "time_per_iteration": 2.4684314727783203
    },
    {
      "auxiliary_loss_clip": 0.01154392,
      "auxiliary_loss_mlp": 0.01027443,
      "balance_loss_clip": 1.05688882,
      "balance_loss_mlp": 1.01951253,
      "epoch": 0.3936752239523838,
      "flos": 13296357400320.0,
      "grad_norm": 1.8535559958076178,
      "language_loss": 0.76550823,
      "learning_rate": 2.7659065890388336e-06,
      "loss": 0.78732657,
      "num_input_tokens_seen": 70567920,
      "step": 3274,
      "time_per_iteration": 2.4735236167907715
    },
    {
      "auxiliary_loss_clip": 0.0116069,
      "auxiliary_loss_mlp": 0.01032118,
      "balance_loss_clip": 1.05290723,
      "balance_loss_mlp": 1.02391624,
      "epoch": 0.3937954668430229,
      "flos": 16800161472000.0,
      "grad_norm": 1.8541142995945012,
      "language_loss": 0.84696734,
      "learning_rate": 2.7651869421118266e-06,
      "loss": 0.86889541,
      "num_input_tokens_seen": 70584530,
      "step": 3275,
      "time_per_iteration": 2.4858474731445312
    },
    {
      "auxiliary_loss_clip": 0.01177781,
      "auxiliary_loss_mlp": 0.01030098,
      "balance_loss_clip": 1.05752325,
      "balance_loss_mlp": 1.02223027,
      "epoch": 0.393915709733662,
      "flos": 21064229832960.0,
      "grad_norm": 1.8418147506381601,
      "language_loss": 0.82987547,
      "learning_rate": 2.76446717910657e-06,
      "loss": 0.85195422,
      "num_input_tokens_seen": 70605235,
      "step": 3276,
      "time_per_iteration": 4.063091516494751
    },
    {
      "auxiliary_loss_clip": 0.01169298,
      "auxiliary_loss_mlp": 0.01028292,
      "balance_loss_clip": 1.05397987,
      "balance_loss_mlp": 1.0208652,
      "epoch": 0.3940359526243011,
      "flos": 17165265264000.0,
      "grad_norm": 2.278352460426256,
      "language_loss": 0.76940465,
      "learning_rate": 2.763747300132249e-06,
      "loss": 0.79138058,
      "num_input_tokens_seen": 70622675,
      "step": 3277,
      "time_per_iteration": 2.530925750732422
    },
    {
      "auxiliary_loss_clip": 0.01186039,
      "auxiliary_loss_mlp": 0.01027329,
      "balance_loss_clip": 1.05651355,
      "balance_loss_mlp": 1.01966929,
      "epoch": 0.3941561955149402,
      "flos": 20995856294400.0,
      "grad_norm": 2.4553229721494967,
      "language_loss": 0.86797357,
      "learning_rate": 2.7630273052980704e-06,
      "loss": 0.89010721,
      "num_input_tokens_seen": 70643265,
      "step": 3278,
      "time_per_iteration": 2.4568777084350586
    },
    {
      "auxiliary_loss_clip": 0.01147155,
      "auxiliary_loss_mlp": 0.01028119,
      "balance_loss_clip": 1.05091381,
      "balance_loss_mlp": 1.02043939,
      "epoch": 0.39427643840557924,
      "flos": 18843406721280.0,
      "grad_norm": 2.1139731394065517,
      "language_loss": 0.67189413,
      "learning_rate": 2.7623071947132554e-06,
      "loss": 0.69364685,
      "num_input_tokens_seen": 70660295,
      "step": 3279,
      "time_per_iteration": 2.4969375133514404
    },
    {
      "auxiliary_loss_clip": 0.0116263,
      "auxiliary_loss_mlp": 0.01029876,
      "balance_loss_clip": 1.0519774,
      "balance_loss_mlp": 1.02210319,
      "epoch": 0.39439668129621835,
      "flos": 23258659426560.0,
      "grad_norm": 2.0309250766719122,
      "language_loss": 0.78960013,
      "learning_rate": 2.7615869684870458e-06,
      "loss": 0.81152523,
      "num_input_tokens_seen": 70679605,
      "step": 3280,
      "time_per_iteration": 2.5203399658203125
    },
    {
      "auxiliary_loss_clip": 0.0116898,
      "auxiliary_loss_mlp": 0.01025797,
      "balance_loss_clip": 1.05459476,
      "balance_loss_mlp": 1.01813805,
      "epoch": 0.39451692418685746,
      "flos": 26652289507200.0,
      "grad_norm": 1.6496987635799056,
      "language_loss": 0.84804529,
      "learning_rate": 2.7608666267286986e-06,
      "loss": 0.86999303,
      "num_input_tokens_seen": 70699835,
      "step": 3281,
      "time_per_iteration": 2.522454023361206
    },
    {
      "auxiliary_loss_clip": 0.01110992,
      "auxiliary_loss_mlp": 0.01028882,
      "balance_loss_clip": 1.04392552,
      "balance_loss_mlp": 1.02028084,
      "epoch": 0.3946371670774965,
      "flos": 18258709132800.0,
      "grad_norm": 2.1428145301837414,
      "language_loss": 0.8665821,
      "learning_rate": 2.760146169547489e-06,
      "loss": 0.88798082,
      "num_input_tokens_seen": 70716600,
      "step": 3282,
      "time_per_iteration": 2.5824480056762695
    },
    {
      "auxiliary_loss_clip": 0.01159153,
      "auxiliary_loss_mlp": 0.01029247,
      "balance_loss_clip": 1.056229,
      "balance_loss_mlp": 1.0214746,
      "epoch": 0.39475740996813563,
      "flos": 24206126423040.0,
      "grad_norm": 1.4804869744378997,
      "language_loss": 0.76509339,
      "learning_rate": 2.75942559705271e-06,
      "loss": 0.78697741,
      "num_input_tokens_seen": 70736335,
      "step": 3283,
      "time_per_iteration": 2.5750155448913574
    },
    {
      "auxiliary_loss_clip": 0.01168846,
      "auxiliary_loss_mlp": 0.01031608,
      "balance_loss_clip": 1.05418956,
      "balance_loss_mlp": 1.0238657,
      "epoch": 0.39487765285877474,
      "flos": 19317858491520.0,
      "grad_norm": 1.885595811139525,
      "language_loss": 0.89396888,
      "learning_rate": 2.7587049093536713e-06,
      "loss": 0.91597342,
      "num_input_tokens_seen": 70752665,
      "step": 3284,
      "time_per_iteration": 2.457979679107666
    },
    {
      "auxiliary_loss_clip": 0.01175037,
      "auxiliary_loss_mlp": 0.01033707,
      "balance_loss_clip": 1.05383873,
      "balance_loss_mlp": 1.02561891,
      "epoch": 0.3949978957494138,
      "flos": 17311744926720.0,
      "grad_norm": 1.828702407432217,
      "language_loss": 0.80783391,
      "learning_rate": 2.757984106559701e-06,
      "loss": 0.82992136,
      "num_input_tokens_seen": 70771650,
      "step": 3285,
      "time_per_iteration": 2.451009750366211
    },
    {
      "auxiliary_loss_clip": 0.01150458,
      "auxiliary_loss_mlp": 0.01029947,
      "balance_loss_clip": 1.05323124,
      "balance_loss_mlp": 1.02134013,
      "epoch": 0.3951181386400529,
      "flos": 36317861280000.0,
      "grad_norm": 2.131549892746281,
      "language_loss": 0.71350396,
      "learning_rate": 2.7572631887801446e-06,
      "loss": 0.73530793,
      "num_input_tokens_seen": 70793275,
      "step": 3286,
      "time_per_iteration": 2.6279146671295166
    },
    {
      "auxiliary_loss_clip": 0.01173297,
      "auxiliary_loss_mlp": 0.0103289,
      "balance_loss_clip": 1.05439973,
      "balance_loss_mlp": 1.02444983,
      "epoch": 0.395238381530692,
      "flos": 23110348170240.0,
      "grad_norm": 1.736145692358794,
      "language_loss": 0.76463598,
      "learning_rate": 2.7565421561243654e-06,
      "loss": 0.78669786,
      "num_input_tokens_seen": 70811440,
      "step": 3287,
      "time_per_iteration": 2.483351230621338
    },
    {
      "auxiliary_loss_clip": 0.01135227,
      "auxiliary_loss_mlp": 0.01027277,
      "balance_loss_clip": 1.04888797,
      "balance_loss_mlp": 1.01925969,
      "epoch": 0.3953586244213311,
      "flos": 24347614095360.0,
      "grad_norm": 2.104480105140653,
      "language_loss": 0.82103282,
      "learning_rate": 2.7558210087017413e-06,
      "loss": 0.8426578,
      "num_input_tokens_seen": 70831375,
      "step": 3288,
      "time_per_iteration": 2.595792531967163
    },
    {
      "auxiliary_loss_clip": 0.01140659,
      "auxiliary_loss_mlp": 0.01029322,
      "balance_loss_clip": 1.05560422,
      "balance_loss_mlp": 1.02059579,
      "epoch": 0.3954788673119702,
      "flos": 23440080044160.0,
      "grad_norm": 1.7170317189933366,
      "language_loss": 0.73595458,
      "learning_rate": 2.7550997466216724e-06,
      "loss": 0.75765437,
      "num_input_tokens_seen": 70849170,
      "step": 3289,
      "time_per_iteration": 2.534391403198242
    },
    {
      "auxiliary_loss_clip": 0.01156417,
      "auxiliary_loss_mlp": 0.01030762,
      "balance_loss_clip": 1.05793762,
      "balance_loss_mlp": 1.02278113,
      "epoch": 0.3955991102026093,
      "flos": 17494063384320.0,
      "grad_norm": 1.795017574319635,
      "language_loss": 0.81494516,
      "learning_rate": 2.7543783699935714e-06,
      "loss": 0.83681697,
      "num_input_tokens_seen": 70867200,
      "step": 3290,
      "time_per_iteration": 2.4821085929870605
    },
    {
      "auxiliary_loss_clip": 0.0117174,
      "auxiliary_loss_mlp": 0.01029309,
      "balance_loss_clip": 1.05730796,
      "balance_loss_mlp": 1.02088094,
      "epoch": 0.39571935309324835,
      "flos": 18221326053120.0,
      "grad_norm": 3.3685546131809874,
      "language_loss": 0.86355239,
      "learning_rate": 2.753656878926872e-06,
      "loss": 0.8855629,
      "num_input_tokens_seen": 70883080,
      "step": 3291,
      "time_per_iteration": 2.444225549697876
    },
    {
      "auxiliary_loss_clip": 0.01145082,
      "auxiliary_loss_mlp": 0.01028411,
      "balance_loss_clip": 1.04885352,
      "balance_loss_mlp": 1.01986361,
      "epoch": 0.39583959598388746,
      "flos": 17748813617280.0,
      "grad_norm": 1.757528766332802,
      "language_loss": 0.74279284,
      "learning_rate": 2.752935273531023e-06,
      "loss": 0.76452774,
      "num_input_tokens_seen": 70901230,
      "step": 3292,
      "time_per_iteration": 2.4704136848449707
    },
    {
      "auxiliary_loss_clip": 0.01172147,
      "auxiliary_loss_mlp": 0.01026466,
      "balance_loss_clip": 1.05391765,
      "balance_loss_mlp": 1.01784158,
      "epoch": 0.39595983887452657,
      "flos": 19352368483200.0,
      "grad_norm": 1.9902570651557974,
      "language_loss": 0.78126264,
      "learning_rate": 2.752213553915492e-06,
      "loss": 0.80324876,
      "num_input_tokens_seen": 70919585,
      "step": 3293,
      "time_per_iteration": 2.4594690799713135
    },
    {
      "auxiliary_loss_clip": 0.01060207,
      "auxiliary_loss_mlp": 0.01005951,
      "balance_loss_clip": 1.01696682,
      "balance_loss_mlp": 1.00484204,
      "epoch": 0.3960800817651656,
      "flos": 60682282940160.0,
      "grad_norm": 0.8258775305275187,
      "language_loss": 0.66093391,
      "learning_rate": 2.751491720189762e-06,
      "loss": 0.6815955,
      "num_input_tokens_seen": 70977695,
      "step": 3294,
      "time_per_iteration": 3.0303730964660645
    },
    {
      "auxiliary_loss_clip": 0.01155984,
      "auxiliary_loss_mlp": 0.00762895,
      "balance_loss_clip": 1.05299711,
      "balance_loss_mlp": 1.00045896,
      "epoch": 0.39620032465580474,
      "flos": 16836718538880.0,
      "grad_norm": 2.099570115804247,
      "language_loss": 0.91771901,
      "learning_rate": 2.7507697724633364e-06,
      "loss": 0.93690777,
      "num_input_tokens_seen": 70994455,
      "step": 3295,
      "time_per_iteration": 3.3124825954437256
    },
    {
      "auxiliary_loss_clip": 0.01050161,
      "auxiliary_loss_mlp": 0.01019746,
      "balance_loss_clip": 1.02417922,
      "balance_loss_mlp": 1.01763618,
      "epoch": 0.3963205675464438,
      "flos": 69071445941760.0,
      "grad_norm": 0.779361968435171,
      "language_loss": 0.54674315,
      "learning_rate": 2.7500477108457327e-06,
      "loss": 0.56744224,
      "num_input_tokens_seen": 71046465,
      "step": 3296,
      "time_per_iteration": 2.927067279815674
    },
    {
      "auxiliary_loss_clip": 0.0112561,
      "auxiliary_loss_mlp": 0.01029293,
      "balance_loss_clip": 1.04574406,
      "balance_loss_mlp": 1.02032816,
      "epoch": 0.3964408104370829,
      "flos": 25667439431040.0,
      "grad_norm": 1.8923014506886564,
      "language_loss": 0.80510932,
      "learning_rate": 2.7493255354464877e-06,
      "loss": 0.82665831,
      "num_input_tokens_seen": 71064275,
      "step": 3297,
      "time_per_iteration": 2.58343243598938
    },
    {
      "auxiliary_loss_clip": 0.01099265,
      "auxiliary_loss_mlp": 0.0102735,
      "balance_loss_clip": 1.04591966,
      "balance_loss_mlp": 1.01944637,
      "epoch": 0.396561053327722,
      "flos": 24277480790400.0,
      "grad_norm": 2.2940253043828736,
      "language_loss": 0.76256281,
      "learning_rate": 2.748603246375156e-06,
      "loss": 0.78382897,
      "num_input_tokens_seen": 71082290,
      "step": 3298,
      "time_per_iteration": 2.6529412269592285
    },
    {
      "auxiliary_loss_clip": 0.01185318,
      "auxiliary_loss_mlp": 0.01034995,
      "balance_loss_clip": 1.05652261,
      "balance_loss_mlp": 1.02679586,
      "epoch": 0.39668129621836107,
      "flos": 20522302364160.0,
      "grad_norm": 2.2860200728620304,
      "language_loss": 0.70224375,
      "learning_rate": 2.7478808437413055e-06,
      "loss": 0.72444689,
      "num_input_tokens_seen": 71101700,
      "step": 3299,
      "time_per_iteration": 3.29171085357666
    },
    {
      "auxiliary_loss_clip": 0.01131616,
      "auxiliary_loss_mlp": 0.01027461,
      "balance_loss_clip": 1.05585361,
      "balance_loss_mlp": 1.01921213,
      "epoch": 0.3968015391090002,
      "flos": 27052585649280.0,
      "grad_norm": 1.8374207464210466,
      "language_loss": 0.66066754,
      "learning_rate": 2.7471583276545263e-06,
      "loss": 0.68225831,
      "num_input_tokens_seen": 71122360,
      "step": 3300,
      "time_per_iteration": 2.6148300170898438
    },
    {
      "auxiliary_loss_clip": 0.0115726,
      "auxiliary_loss_mlp": 0.01030419,
      "balance_loss_clip": 1.0520978,
      "balance_loss_mlp": 1.0227536,
      "epoch": 0.3969217819996393,
      "flos": 12531819392640.0,
      "grad_norm": 2.314079434148456,
      "language_loss": 0.70782506,
      "learning_rate": 2.7464356982244224e-06,
      "loss": 0.72970188,
      "num_input_tokens_seen": 71140360,
      "step": 3301,
      "time_per_iteration": 2.4854118824005127
    },
    {
      "auxiliary_loss_clip": 0.01074202,
      "auxiliary_loss_mlp": 0.01004612,
      "balance_loss_clip": 1.02616191,
      "balance_loss_mlp": 1.0026803,
      "epoch": 0.39704202489027834,
      "flos": 66241399230720.0,
      "grad_norm": 0.7736484151504005,
      "language_loss": 0.61720312,
      "learning_rate": 2.745712955560617e-06,
      "loss": 0.63799131,
      "num_input_tokens_seen": 71196565,
      "step": 3302,
      "time_per_iteration": 3.7873544692993164
    },
    {
      "auxiliary_loss_clip": 0.01113514,
      "auxiliary_loss_mlp": 0.01029657,
      "balance_loss_clip": 1.04912424,
      "balance_loss_mlp": 1.02062714,
      "epoch": 0.39716226778091746,
      "flos": 16982982720000.0,
      "grad_norm": 3.027195481383852,
      "language_loss": 0.77458537,
      "learning_rate": 2.7449900997727496e-06,
      "loss": 0.79601711,
      "num_input_tokens_seen": 71214675,
      "step": 3303,
      "time_per_iteration": 3.357720375061035
    },
    {
      "auxiliary_loss_clip": 0.01156318,
      "auxiliary_loss_mlp": 0.01031392,
      "balance_loss_clip": 1.05588877,
      "balance_loss_mlp": 1.02373898,
      "epoch": 0.39728251067155657,
      "flos": 23477139901440.0,
      "grad_norm": 2.061190945910885,
      "language_loss": 0.84094739,
      "learning_rate": 2.744267130970476e-06,
      "loss": 0.86282444,
      "num_input_tokens_seen": 71234400,
      "step": 3304,
      "time_per_iteration": 2.509260892868042
    },
    {
      "auxiliary_loss_clip": 0.01153412,
      "auxiliary_loss_mlp": 0.01030891,
      "balance_loss_clip": 1.05415094,
      "balance_loss_mlp": 1.02217698,
      "epoch": 0.3974027535621956,
      "flos": 20704441253760.0,
      "grad_norm": 1.738133313676558,
      "language_loss": 0.77289069,
      "learning_rate": 2.7435440492634697e-06,
      "loss": 0.7947337,
      "num_input_tokens_seen": 71253725,
      "step": 3305,
      "time_per_iteration": 2.5091795921325684
    },
    {
      "auxiliary_loss_clip": 0.01155957,
      "auxiliary_loss_mlp": 0.01032191,
      "balance_loss_clip": 1.05122185,
      "balance_loss_mlp": 1.02337551,
      "epoch": 0.39752299645283473,
      "flos": 21543278544000.0,
      "grad_norm": 2.6393113719047503,
      "language_loss": 0.67367053,
      "learning_rate": 2.7428208547614228e-06,
      "loss": 0.69555199,
      "num_input_tokens_seen": 71273220,
      "step": 3306,
      "time_per_iteration": 2.504953384399414
    },
    {
      "auxiliary_loss_clip": 0.01173746,
      "auxiliary_loss_mlp": 0.01032724,
      "balance_loss_clip": 1.05654716,
      "balance_loss_mlp": 1.02468312,
      "epoch": 0.39764323934347384,
      "flos": 19208295031680.0,
      "grad_norm": 1.9824610939563567,
      "language_loss": 0.77812493,
      "learning_rate": 2.742097547574043e-06,
      "loss": 0.80018967,
      "num_input_tokens_seen": 71291445,
      "step": 3307,
      "time_per_iteration": 2.4473907947540283
    },
    {
      "auxiliary_loss_clip": 0.01162014,
      "auxiliary_loss_mlp": 0.00762786,
      "balance_loss_clip": 1.05259252,
      "balance_loss_mlp": 1.00036943,
      "epoch": 0.3977634822341129,
      "flos": 20850202644480.0,
      "grad_norm": 1.9633182423986277,
      "language_loss": 0.77651799,
      "learning_rate": 2.7413741278110544e-06,
      "loss": 0.795766,
      "num_input_tokens_seen": 71310135,
      "step": 3308,
      "time_per_iteration": 2.5101242065429688
    },
    {
      "auxiliary_loss_clip": 0.01162292,
      "auxiliary_loss_mlp": 0.01031036,
      "balance_loss_clip": 1.05502284,
      "balance_loss_mlp": 1.02219105,
      "epoch": 0.397883725124752,
      "flos": 39786042038400.0,
      "grad_norm": 2.3258358048190773,
      "language_loss": 0.68837166,
      "learning_rate": 2.7406505955822016e-06,
      "loss": 0.71030498,
      "num_input_tokens_seen": 71331160,
      "step": 3309,
      "time_per_iteration": 2.662421941757202
    },
    {
      "auxiliary_loss_clip": 0.01157974,
      "auxiliary_loss_mlp": 0.01030105,
      "balance_loss_clip": 1.05229855,
      "balance_loss_mlp": 1.02196276,
      "epoch": 0.39800396801539106,
      "flos": 17379507934080.0,
      "grad_norm": 4.099665470047776,
      "language_loss": 0.66162884,
      "learning_rate": 2.7399269509972415e-06,
      "loss": 0.68350959,
      "num_input_tokens_seen": 71345315,
      "step": 3310,
      "time_per_iteration": 2.463991403579712
    },
    {
      "auxiliary_loss_clip": 0.01148981,
      "auxiliary_loss_mlp": 0.01029731,
      "balance_loss_clip": 1.047032,
      "balance_loss_mlp": 1.02070713,
      "epoch": 0.3981242109060302,
      "flos": 19202764337280.0,
      "grad_norm": 2.4049500361340743,
      "language_loss": 0.85456431,
      "learning_rate": 2.7392031941659514e-06,
      "loss": 0.87635148,
      "num_input_tokens_seen": 71363160,
      "step": 3311,
      "time_per_iteration": 2.4707274436950684
    },
    {
      "auxiliary_loss_clip": 0.01158328,
      "auxiliary_loss_mlp": 0.01038866,
      "balance_loss_clip": 1.05653465,
      "balance_loss_mlp": 1.03034306,
      "epoch": 0.3982444537966693,
      "flos": 24565124903040.0,
      "grad_norm": 1.7721481224473024,
      "language_loss": 0.85882705,
      "learning_rate": 2.7384793251981244e-06,
      "loss": 0.88079906,
      "num_input_tokens_seen": 71382145,
      "step": 3312,
      "time_per_iteration": 2.520378589630127
    },
    {
      "auxiliary_loss_clip": 0.01177656,
      "auxiliary_loss_mlp": 0.01030183,
      "balance_loss_clip": 1.05459666,
      "balance_loss_mlp": 1.02215445,
      "epoch": 0.39836469668730834,
      "flos": 26213856099840.0,
      "grad_norm": 1.8308992717947343,
      "language_loss": 0.81013483,
      "learning_rate": 2.737755344203571e-06,
      "loss": 0.83221328,
      "num_input_tokens_seen": 71402095,
      "step": 3313,
      "time_per_iteration": 2.517637252807617
    },
    {
      "auxiliary_loss_clip": 0.01177489,
      "auxiliary_loss_mlp": 0.01031432,
      "balance_loss_clip": 1.05829382,
      "balance_loss_mlp": 1.02346861,
      "epoch": 0.39848493957794745,
      "flos": 27636134002560.0,
      "grad_norm": 1.7078794558468287,
      "language_loss": 0.79774928,
      "learning_rate": 2.7370312512921186e-06,
      "loss": 0.81983846,
      "num_input_tokens_seen": 71423875,
      "step": 3314,
      "time_per_iteration": 2.5207886695861816
    },
    {
      "auxiliary_loss_clip": 0.01160218,
      "auxiliary_loss_mlp": 0.01036248,
      "balance_loss_clip": 1.05208588,
      "balance_loss_mlp": 1.02687252,
      "epoch": 0.39860518246858656,
      "flos": 12239326944000.0,
      "grad_norm": 2.4720947779905376,
      "language_loss": 0.76794827,
      "learning_rate": 2.736307046573611e-06,
      "loss": 0.78991288,
      "num_input_tokens_seen": 71439745,
      "step": 3315,
      "time_per_iteration": 2.466719150543213
    },
    {
      "auxiliary_loss_clip": 0.01184243,
      "auxiliary_loss_mlp": 0.01025953,
      "balance_loss_clip": 1.05461872,
      "balance_loss_mlp": 1.01816869,
      "epoch": 0.3987254253592256,
      "flos": 22379135005440.0,
      "grad_norm": 1.6471847826049513,
      "language_loss": 0.82011878,
      "learning_rate": 2.73558273015791e-06,
      "loss": 0.84222078,
      "num_input_tokens_seen": 71459575,
      "step": 3316,
      "time_per_iteration": 2.444368362426758
    },
    {
      "auxiliary_loss_clip": 0.01189895,
      "auxiliary_loss_mlp": 0.01032375,
      "balance_loss_clip": 1.05685544,
      "balance_loss_mlp": 1.02311206,
      "epoch": 0.3988456682498647,
      "flos": 23514020190720.0,
      "grad_norm": 2.686505729113726,
      "language_loss": 0.70428991,
      "learning_rate": 2.734858302154894e-06,
      "loss": 0.72651255,
      "num_input_tokens_seen": 71481075,
      "step": 3317,
      "time_per_iteration": 2.4510557651519775
    },
    {
      "auxiliary_loss_clip": 0.01152969,
      "auxiliary_loss_mlp": 0.01033165,
      "balance_loss_clip": 1.05282629,
      "balance_loss_mlp": 1.02468944,
      "epoch": 0.39896591114050384,
      "flos": 19208761908480.0,
      "grad_norm": 2.3979298983238664,
      "language_loss": 0.76490843,
      "learning_rate": 2.734133762674457e-06,
      "loss": 0.78676975,
      "num_input_tokens_seen": 71500665,
      "step": 3318,
      "time_per_iteration": 2.485837697982788
    },
    {
      "auxiliary_loss_clip": 0.01159016,
      "auxiliary_loss_mlp": 0.01031311,
      "balance_loss_clip": 1.05365241,
      "balance_loss_mlp": 1.02275157,
      "epoch": 0.3990861540311429,
      "flos": 28401031146240.0,
      "grad_norm": 2.025857031041854,
      "language_loss": 0.70574152,
      "learning_rate": 2.7334091118265124e-06,
      "loss": 0.7276448,
      "num_input_tokens_seen": 71522560,
      "step": 3319,
      "time_per_iteration": 2.5475356578826904
    },
    {
      "auxiliary_loss_clip": 0.01073373,
      "auxiliary_loss_mlp": 0.01011207,
      "balance_loss_clip": 1.01629364,
      "balance_loss_mlp": 1.01012802,
      "epoch": 0.399206396921782,
      "flos": 61758563086080.0,
      "grad_norm": 0.6793530427591613,
      "language_loss": 0.57815641,
      "learning_rate": 2.732684349720989e-06,
      "loss": 0.59900218,
      "num_input_tokens_seen": 71590520,
      "step": 3320,
      "time_per_iteration": 3.0509047508239746
    },
    {
      "auxiliary_loss_clip": 0.01148158,
      "auxiliary_loss_mlp": 0.01028459,
      "balance_loss_clip": 1.05172324,
      "balance_loss_mlp": 1.02026963,
      "epoch": 0.3993266398124211,
      "flos": 28074567409920.0,
      "grad_norm": 1.7190728990591573,
      "language_loss": 0.75219119,
      "learning_rate": 2.7319594764678318e-06,
      "loss": 0.77395737,
      "num_input_tokens_seen": 71612620,
      "step": 3321,
      "time_per_iteration": 2.583740711212158
    },
    {
      "auxiliary_loss_clip": 0.01134205,
      "auxiliary_loss_mlp": 0.01035025,
      "balance_loss_clip": 1.0516715,
      "balance_loss_mlp": 1.02612579,
      "epoch": 0.39944688270306017,
      "flos": 23225083188480.0,
      "grad_norm": 1.7419892730984303,
      "language_loss": 0.83222747,
      "learning_rate": 2.7312344921770044e-06,
      "loss": 0.85391974,
      "num_input_tokens_seen": 71634320,
      "step": 3322,
      "time_per_iteration": 3.4414803981781006
    },
    {
      "auxiliary_loss_clip": 0.01156907,
      "auxiliary_loss_mlp": 0.01032003,
      "balance_loss_clip": 1.05087423,
      "balance_loss_mlp": 1.02412939,
      "epoch": 0.3995671255936993,
      "flos": 19390433921280.0,
      "grad_norm": 2.1001649743543584,
      "language_loss": 0.78482485,
      "learning_rate": 2.7305093969584857e-06,
      "loss": 0.80671394,
      "num_input_tokens_seen": 71653145,
      "step": 3323,
      "time_per_iteration": 2.519064426422119
    },
    {
      "auxiliary_loss_clip": 0.01165869,
      "auxiliary_loss_mlp": 0.01031495,
      "balance_loss_clip": 1.05294657,
      "balance_loss_mlp": 1.02350187,
      "epoch": 0.3996873684843384,
      "flos": 23842638743040.0,
      "grad_norm": 2.1214910284084754,
      "language_loss": 0.80035973,
      "learning_rate": 2.729784190922272e-06,
      "loss": 0.8223334,
      "num_input_tokens_seen": 71674580,
      "step": 3324,
      "time_per_iteration": 2.483886957168579
    },
    {
      "auxiliary_loss_clip": 0.01062868,
      "auxiliary_loss_mlp": 0.0100181,
      "balance_loss_clip": 1.01761556,
      "balance_loss_mlp": 1.00079083,
      "epoch": 0.39980761137497745,
      "flos": 66576877280640.0,
      "grad_norm": 0.93636568524594,
      "language_loss": 0.57192075,
      "learning_rate": 2.729058874178378e-06,
      "loss": 0.59256756,
      "num_input_tokens_seen": 71745260,
      "step": 3325,
      "time_per_iteration": 3.1218621730804443
    },
    {
      "auxiliary_loss_clip": 0.01148908,
      "auxiliary_loss_mlp": 0.01032106,
      "balance_loss_clip": 1.05112278,
      "balance_loss_mlp": 1.02364182,
      "epoch": 0.39992785426561656,
      "flos": 28549162834560.0,
      "grad_norm": 2.227029100420481,
      "language_loss": 0.69408619,
      "learning_rate": 2.7283334468368315e-06,
      "loss": 0.71589637,
      "num_input_tokens_seen": 71766540,
      "step": 3326,
      "time_per_iteration": 3.4614078998565674
    },
    {
      "auxiliary_loss_clip": 0.01094289,
      "auxiliary_loss_mlp": 0.01028907,
      "balance_loss_clip": 1.04281664,
      "balance_loss_mlp": 1.02024627,
      "epoch": 0.4000480971562556,
      "flos": 15049408671360.0,
      "grad_norm": 1.794609623628467,
      "language_loss": 0.73329663,
      "learning_rate": 2.72760790900768e-06,
      "loss": 0.75452864,
      "num_input_tokens_seen": 71783125,
      "step": 3327,
      "time_per_iteration": 2.5859737396240234
    },
    {
      "auxiliary_loss_clip": 0.01187686,
      "auxiliary_loss_mlp": 0.01028176,
      "balance_loss_clip": 1.05732679,
      "balance_loss_mlp": 1.02002192,
      "epoch": 0.4001683400468947,
      "flos": 23915609222400.0,
      "grad_norm": 3.023837708122538,
      "language_loss": 0.78608185,
      "learning_rate": 2.7268822608009875e-06,
      "loss": 0.80824047,
      "num_input_tokens_seen": 71802500,
      "step": 3328,
      "time_per_iteration": 2.461385726928711
    },
    {
      "auxiliary_loss_clip": 0.01148319,
      "auxiliary_loss_mlp": 0.01030315,
      "balance_loss_clip": 1.0515132,
      "balance_loss_mlp": 1.02224481,
      "epoch": 0.40028858293753383,
      "flos": 24352677912960.0,
      "grad_norm": 2.002312701284234,
      "language_loss": 0.78241062,
      "learning_rate": 2.726156502326834e-06,
      "loss": 0.80419701,
      "num_input_tokens_seen": 71823800,
      "step": 3329,
      "time_per_iteration": 3.4065518379211426
    },
    {
      "auxiliary_loss_clip": 0.01037392,
      "auxiliary_loss_mlp": 0.01006301,
      "balance_loss_clip": 1.02475178,
      "balance_loss_mlp": 1.00453663,
      "epoch": 0.4004088258281729,
      "flos": 66787025800320.0,
      "grad_norm": 0.6954955816553068,
      "language_loss": 0.60248381,
      "learning_rate": 2.725430633695316e-06,
      "loss": 0.62292069,
      "num_input_tokens_seen": 71886880,
      "step": 3330,
      "time_per_iteration": 3.9855246543884277
    },
    {
      "auxiliary_loss_clip": 0.01081976,
      "auxiliary_loss_mlp": 0.01001849,
      "balance_loss_clip": 1.01486754,
      "balance_loss_mlp": 1.00086582,
      "epoch": 0.400529068718812,
      "flos": 58598386473600.0,
      "grad_norm": 0.8859229209750509,
      "language_loss": 0.57941341,
      "learning_rate": 2.7247046550165485e-06,
      "loss": 0.60025156,
      "num_input_tokens_seen": 71939005,
      "step": 3331,
      "time_per_iteration": 3.096722364425659
    },
    {
      "auxiliary_loss_clip": 0.0118897,
      "auxiliary_loss_mlp": 0.01029622,
      "balance_loss_clip": 1.05753112,
      "balance_loss_mlp": 1.02147448,
      "epoch": 0.4006493116094511,
      "flos": 25377460934400.0,
      "grad_norm": 1.405170299905092,
      "language_loss": 0.76002222,
      "learning_rate": 2.7239785664006606e-06,
      "loss": 0.78220809,
      "num_input_tokens_seen": 71962545,
      "step": 3332,
      "time_per_iteration": 2.495569944381714
    },
    {
      "auxiliary_loss_clip": 0.01070862,
      "auxiliary_loss_mlp": 0.01002334,
      "balance_loss_clip": 1.01329231,
      "balance_loss_mlp": 1.00127268,
      "epoch": 0.40076955450009016,
      "flos": 60280729822080.0,
      "grad_norm": 0.8899218323031897,
      "language_loss": 0.6179955,
      "learning_rate": 2.7232523679578002e-06,
      "loss": 0.63872755,
      "num_input_tokens_seen": 72025625,
      "step": 3333,
      "time_per_iteration": 3.0788426399230957
    },
    {
      "auxiliary_loss_clip": 0.01168915,
      "auxiliary_loss_mlp": 0.01026054,
      "balance_loss_clip": 1.05561137,
      "balance_loss_mlp": 1.01842427,
      "epoch": 0.4008897973907293,
      "flos": 16617268396800.0,
      "grad_norm": 2.2695694963287236,
      "language_loss": 0.79483086,
      "learning_rate": 2.7225260597981295e-06,
      "loss": 0.81678057,
      "num_input_tokens_seen": 72043330,
      "step": 3334,
      "time_per_iteration": 2.439711332321167
    },
    {
      "auxiliary_loss_clip": 0.01144179,
      "auxiliary_loss_mlp": 0.00763604,
      "balance_loss_clip": 1.05355275,
      "balance_loss_mlp": 1.0002563,
      "epoch": 0.4010100402813684,
      "flos": 15377344865280.0,
      "grad_norm": 2.8254115072612023,
      "language_loss": 0.78385192,
      "learning_rate": 2.721799642031831e-06,
      "loss": 0.80292976,
      "num_input_tokens_seen": 72059500,
      "step": 3335,
      "time_per_iteration": 2.494485378265381
    },
    {
      "auxiliary_loss_clip": 0.01160482,
      "auxiliary_loss_mlp": 0.01032835,
      "balance_loss_clip": 1.04908133,
      "balance_loss_mlp": 1.02450252,
      "epoch": 0.40113028317200744,
      "flos": 13298835438720.0,
      "grad_norm": 1.926169863973025,
      "language_loss": 0.78056395,
      "learning_rate": 2.721073114769101e-06,
      "loss": 0.80249709,
      "num_input_tokens_seen": 72077175,
      "step": 3336,
      "time_per_iteration": 2.4852471351623535
    },
    {
      "auxiliary_loss_clip": 0.01138502,
      "auxiliary_loss_mlp": 0.01032638,
      "balance_loss_clip": 1.05059814,
      "balance_loss_mlp": 1.02478826,
      "epoch": 0.40125052606264655,
      "flos": 20668027841280.0,
      "grad_norm": 1.7518925359226891,
      "language_loss": 0.75076759,
      "learning_rate": 2.7203464781201523e-06,
      "loss": 0.772479,
      "num_input_tokens_seen": 72096490,
      "step": 3337,
      "time_per_iteration": 2.537952423095703
    },
    {
      "auxiliary_loss_clip": 0.01187696,
      "auxiliary_loss_mlp": 0.01032416,
      "balance_loss_clip": 1.05787134,
      "balance_loss_mlp": 1.02460778,
      "epoch": 0.40137076895328566,
      "flos": 24607679541120.0,
      "grad_norm": 2.6937475281623735,
      "language_loss": 0.78147501,
      "learning_rate": 2.719619732195215e-06,
      "loss": 0.80367613,
      "num_input_tokens_seen": 72118130,
      "step": 3338,
      "time_per_iteration": 2.4856388568878174
    },
    {
      "auxiliary_loss_clip": 0.01144691,
      "auxiliary_loss_mlp": 0.01026656,
      "balance_loss_clip": 1.05097723,
      "balance_loss_mlp": 1.01882362,
      "epoch": 0.4014910118439247,
      "flos": 24206593299840.0,
      "grad_norm": 1.5229698028392253,
      "language_loss": 0.72668022,
      "learning_rate": 2.7188928771045377e-06,
      "loss": 0.74839365,
      "num_input_tokens_seen": 72139450,
      "step": 3339,
      "time_per_iteration": 2.5688064098358154
    },
    {
      "auxiliary_loss_clip": 0.01138505,
      "auxiliary_loss_mlp": 0.01028556,
      "balance_loss_clip": 1.04984426,
      "balance_loss_mlp": 1.02093291,
      "epoch": 0.4016112547345638,
      "flos": 26725080418560.0,
      "grad_norm": 1.8002474030386153,
      "language_loss": 0.79796803,
      "learning_rate": 2.7181659129583815e-06,
      "loss": 0.81963861,
      "num_input_tokens_seen": 72159040,
      "step": 3340,
      "time_per_iteration": 2.5726680755615234
    },
    {
      "auxiliary_loss_clip": 0.0114701,
      "auxiliary_loss_mlp": 0.0102788,
      "balance_loss_clip": 1.0453589,
      "balance_loss_mlp": 1.0197382,
      "epoch": 0.4017314976252029,
      "flos": 21288025520640.0,
      "grad_norm": 1.7388359306240395,
      "language_loss": 0.75771153,
      "learning_rate": 2.7174388398670276e-06,
      "loss": 0.77946043,
      "num_input_tokens_seen": 72178220,
      "step": 3341,
      "time_per_iteration": 2.530214548110962
    },
    {
      "auxiliary_loss_clip": 0.01186283,
      "auxiliary_loss_mlp": 0.01034798,
      "balance_loss_clip": 1.05222499,
      "balance_loss_mlp": 1.02652526,
      "epoch": 0.401851740515842,
      "flos": 25484690010240.0,
      "grad_norm": 2.952699611839034,
      "language_loss": 0.92222655,
      "learning_rate": 2.716711657940773e-06,
      "loss": 0.94443727,
      "num_input_tokens_seen": 72199230,
      "step": 3342,
      "time_per_iteration": 2.481170415878296
    },
    {
      "auxiliary_loss_clip": 0.01052325,
      "auxiliary_loss_mlp": 0.0100261,
      "balance_loss_clip": 1.01494741,
      "balance_loss_mlp": 1.0016917,
      "epoch": 0.4019719834064811,
      "flos": 55395334978560.0,
      "grad_norm": 0.8103047138159378,
      "language_loss": 0.56454337,
      "learning_rate": 2.7159843672899284e-06,
      "loss": 0.58509266,
      "num_input_tokens_seen": 72263430,
      "step": 3343,
      "time_per_iteration": 3.2316973209381104
    },
    {
      "auxiliary_loss_clip": 0.01173996,
      "auxiliary_loss_mlp": 0.01031473,
      "balance_loss_clip": 1.056916,
      "balance_loss_mlp": 1.02306914,
      "epoch": 0.40209222629712016,
      "flos": 18180100218240.0,
      "grad_norm": 1.9041991478084044,
      "language_loss": 0.81355417,
      "learning_rate": 2.715256968024825e-06,
      "loss": 0.83560884,
      "num_input_tokens_seen": 72280505,
      "step": 3344,
      "time_per_iteration": 2.44474720954895
    },
    {
      "auxiliary_loss_clip": 0.0116394,
      "auxiliary_loss_mlp": 0.01031575,
      "balance_loss_clip": 1.05348635,
      "balance_loss_mlp": 1.02326047,
      "epoch": 0.40221246918775927,
      "flos": 25961009287680.0,
      "grad_norm": 1.488175491406286,
      "language_loss": 0.82278991,
      "learning_rate": 2.7145294602558083e-06,
      "loss": 0.84474498,
      "num_input_tokens_seen": 72301215,
      "step": 3345,
      "time_per_iteration": 2.579402208328247
    },
    {
      "auxiliary_loss_clip": 0.01172224,
      "auxiliary_loss_mlp": 0.01026417,
      "balance_loss_clip": 1.05363774,
      "balance_loss_mlp": 1.01757193,
      "epoch": 0.4023327120783984,
      "flos": 33838912056960.0,
      "grad_norm": 1.6961621906706548,
      "language_loss": 0.70874989,
      "learning_rate": 2.713801844093241e-06,
      "loss": 0.73073626,
      "num_input_tokens_seen": 72322365,
      "step": 3346,
      "time_per_iteration": 2.5805366039276123
    },
    {
      "auxiliary_loss_clip": 0.01172387,
      "auxiliary_loss_mlp": 0.0103336,
      "balance_loss_clip": 1.05469322,
      "balance_loss_mlp": 1.02582586,
      "epoch": 0.40245295496903744,
      "flos": 26900252069760.0,
      "grad_norm": 4.329955084016142,
      "language_loss": 0.88497341,
      "learning_rate": 2.7130741196475014e-06,
      "loss": 0.90703082,
      "num_input_tokens_seen": 72340495,
      "step": 3347,
      "time_per_iteration": 2.5114622116088867
    },
    {
      "auxiliary_loss_clip": 0.01163658,
      "auxiliary_loss_mlp": 0.01031185,
      "balance_loss_clip": 1.0569824,
      "balance_loss_mlp": 1.02243531,
      "epoch": 0.40257319785967655,
      "flos": 36902738436480.0,
      "grad_norm": 1.7347704773764612,
      "language_loss": 0.79523528,
      "learning_rate": 2.7123462870289848e-06,
      "loss": 0.81718373,
      "num_input_tokens_seen": 72360545,
      "step": 3348,
      "time_per_iteration": 3.445404052734375
    },
    {
      "auxiliary_loss_clip": 0.01158163,
      "auxiliary_loss_mlp": 0.01027335,
      "balance_loss_clip": 1.04989457,
      "balance_loss_mlp": 1.0188446,
      "epoch": 0.40269344075031566,
      "flos": 24353180703360.0,
      "grad_norm": 1.5570079333617244,
      "language_loss": 0.81313539,
      "learning_rate": 2.711618346348102e-06,
      "loss": 0.83499038,
      "num_input_tokens_seen": 72381070,
      "step": 3349,
      "time_per_iteration": 2.5344419479370117
    },
    {
      "auxiliary_loss_clip": 0.01151823,
      "auxiliary_loss_mlp": 0.0103358,
      "balance_loss_clip": 1.05239129,
      "balance_loss_mlp": 1.02533638,
      "epoch": 0.4028136836409547,
      "flos": 14389657614720.0,
      "grad_norm": 1.7204618575155866,
      "language_loss": 0.63533592,
      "learning_rate": 2.7108902977152825e-06,
      "loss": 0.65718997,
      "num_input_tokens_seen": 72398970,
      "step": 3350,
      "time_per_iteration": 2.4805235862731934
    },
    {
      "auxiliary_loss_clip": 0.01168873,
      "auxiliary_loss_mlp": 0.0102978,
      "balance_loss_clip": 1.05275476,
      "balance_loss_mlp": 1.02146256,
      "epoch": 0.4029339265315938,
      "flos": 26136037284480.0,
      "grad_norm": 2.2334036188005415,
      "language_loss": 0.75367528,
      "learning_rate": 2.7101621412409704e-06,
      "loss": 0.77566183,
      "num_input_tokens_seen": 72418455,
      "step": 3351,
      "time_per_iteration": 2.525186538696289
    },
    {
      "auxiliary_loss_clip": 0.01186413,
      "auxiliary_loss_mlp": 0.01032365,
      "balance_loss_clip": 1.05487633,
      "balance_loss_mlp": 1.02405632,
      "epoch": 0.40305416942223293,
      "flos": 23256325042560.0,
      "grad_norm": 1.8354931310081248,
      "language_loss": 0.85628814,
      "learning_rate": 2.7094338770356256e-06,
      "loss": 0.8784759,
      "num_input_tokens_seen": 72437540,
      "step": 3352,
      "time_per_iteration": 3.27976393699646
    },
    {
      "auxiliary_loss_clip": 0.01154016,
      "auxiliary_loss_mlp": 0.01030767,
      "balance_loss_clip": 1.05235922,
      "balance_loss_mlp": 1.02269602,
      "epoch": 0.403174412312872,
      "flos": 27089645506560.0,
      "grad_norm": 2.629118436396993,
      "language_loss": 0.63947296,
      "learning_rate": 2.708705505209726e-06,
      "loss": 0.66132081,
      "num_input_tokens_seen": 72458315,
      "step": 3353,
      "time_per_iteration": 2.5375888347625732
    },
    {
      "auxiliary_loss_clip": 0.01122439,
      "auxiliary_loss_mlp": 0.01026105,
      "balance_loss_clip": 1.04585087,
      "balance_loss_mlp": 1.01825547,
      "epoch": 0.4032946552035111,
      "flos": 21756336065280.0,
      "grad_norm": 1.9407194038834914,
      "language_loss": 0.91799098,
      "learning_rate": 2.7079770258737646e-06,
      "loss": 0.93947643,
      "num_input_tokens_seen": 72476225,
      "step": 3354,
      "time_per_iteration": 2.582780599594116
    },
    {
      "auxiliary_loss_clip": 0.01140371,
      "auxiliary_loss_mlp": 0.01030516,
      "balance_loss_clip": 1.04916334,
      "balance_loss_mlp": 1.02134299,
      "epoch": 0.4034148980941502,
      "flos": 17343956448000.0,
      "grad_norm": 3.0027357754337194,
      "language_loss": 0.75169086,
      "learning_rate": 2.707248439138251e-06,
      "loss": 0.77339977,
      "num_input_tokens_seen": 72492460,
      "step": 3355,
      "time_per_iteration": 3.221224069595337
    },
    {
      "auxiliary_loss_clip": 0.01156202,
      "auxiliary_loss_mlp": 0.01031533,
      "balance_loss_clip": 1.05601764,
      "balance_loss_mlp": 1.02368355,
      "epoch": 0.40353514098478926,
      "flos": 22017838055040.0,
      "grad_norm": 1.7240781659763247,
      "language_loss": 0.65557688,
      "learning_rate": 2.7065197451137114e-06,
      "loss": 0.67745423,
      "num_input_tokens_seen": 72513840,
      "step": 3356,
      "time_per_iteration": 2.5538177490234375
    },
    {
      "auxiliary_loss_clip": 0.01158999,
      "auxiliary_loss_mlp": 0.01030862,
      "balance_loss_clip": 1.0540297,
      "balance_loss_mlp": 1.02254725,
      "epoch": 0.4036553838754284,
      "flos": 14246446089600.0,
      "grad_norm": 2.2061355480707143,
      "language_loss": 0.67547947,
      "learning_rate": 2.7057909439106894e-06,
      "loss": 0.69737804,
      "num_input_tokens_seen": 72531695,
      "step": 3357,
      "time_per_iteration": 3.2442209720611572
    },
    {
      "auxiliary_loss_clip": 0.01164787,
      "auxiliary_loss_mlp": 0.00763371,
      "balance_loss_clip": 1.05338669,
      "balance_loss_mlp": 1.00018477,
      "epoch": 0.40377562676606743,
      "flos": 24790644443520.0,
      "grad_norm": 2.709797235465548,
      "language_loss": 0.78505218,
      "learning_rate": 2.7050620356397417e-06,
      "loss": 0.80433381,
      "num_input_tokens_seen": 72550645,
      "step": 3358,
      "time_per_iteration": 2.5643270015716553
    },
    {
      "auxiliary_loss_clip": 0.01184855,
      "auxiliary_loss_mlp": 0.01025171,
      "balance_loss_clip": 1.05713582,
      "balance_loss_mlp": 1.01741993,
      "epoch": 0.40389586965670654,
      "flos": 24061226958720.0,
      "grad_norm": 1.7007413928700912,
      "language_loss": 0.72321671,
      "learning_rate": 2.7043330204114437e-06,
      "loss": 0.74531698,
      "num_input_tokens_seen": 72569355,
      "step": 3359,
      "time_per_iteration": 2.469942092895508
    },
    {
      "auxiliary_loss_clip": 0.01181486,
      "auxiliary_loss_mlp": 0.01028704,
      "balance_loss_clip": 1.05370665,
      "balance_loss_mlp": 1.02064538,
      "epoch": 0.40401611254734565,
      "flos": 16399613934720.0,
      "grad_norm": 2.0329688455435386,
      "language_loss": 0.85432428,
      "learning_rate": 2.7036038983363862e-06,
      "loss": 0.87642622,
      "num_input_tokens_seen": 72585960,
      "step": 3360,
      "time_per_iteration": 2.3977837562561035
    },
    {
      "auxiliary_loss_clip": 0.01168561,
      "auxiliary_loss_mlp": 0.0102978,
      "balance_loss_clip": 1.05547059,
      "balance_loss_mlp": 1.02231503,
      "epoch": 0.4041363554379847,
      "flos": 23988220565760.0,
      "grad_norm": 1.7792993108996924,
      "language_loss": 0.84526694,
      "learning_rate": 2.702874669525177e-06,
      "loss": 0.86725038,
      "num_input_tokens_seen": 72604440,
      "step": 3361,
      "time_per_iteration": 2.482865571975708
    },
    {
      "auxiliary_loss_clip": 0.01149164,
      "auxiliary_loss_mlp": 0.01031117,
      "balance_loss_clip": 1.05765474,
      "balance_loss_mlp": 1.0224092,
      "epoch": 0.4042565983286238,
      "flos": 28401964899840.0,
      "grad_norm": 2.085179581183854,
      "language_loss": 0.69770318,
      "learning_rate": 2.7021453340884394e-06,
      "loss": 0.71950603,
      "num_input_tokens_seen": 72622165,
      "step": 3362,
      "time_per_iteration": 2.5826261043548584
    },
    {
      "auxiliary_loss_clip": 0.01149949,
      "auxiliary_loss_mlp": 0.00762972,
      "balance_loss_clip": 1.05432153,
      "balance_loss_mlp": 1.00017297,
      "epoch": 0.40437684121926293,
      "flos": 17710963660800.0,
      "grad_norm": 2.270351245482001,
      "language_loss": 0.73048806,
      "learning_rate": 2.7014158921368125e-06,
      "loss": 0.74961734,
      "num_input_tokens_seen": 72640490,
      "step": 3363,
      "time_per_iteration": 2.470259666442871
    },
    {
      "auxiliary_loss_clip": 0.01187606,
      "auxiliary_loss_mlp": 0.01032017,
      "balance_loss_clip": 1.05740142,
      "balance_loss_mlp": 1.02351713,
      "epoch": 0.404497084109902,
      "flos": 24018959629440.0,
      "grad_norm": 2.6346032838777114,
      "language_loss": 0.86016029,
      "learning_rate": 2.700686343780953e-06,
      "loss": 0.88235652,
      "num_input_tokens_seen": 72660360,
      "step": 3364,
      "time_per_iteration": 2.4738242626190186
    },
    {
      "auxiliary_loss_clip": 0.0115954,
      "auxiliary_loss_mlp": 0.01028306,
      "balance_loss_clip": 1.05222845,
      "balance_loss_mlp": 1.01986027,
      "epoch": 0.4046173270005411,
      "flos": 22929861306240.0,
      "grad_norm": 1.5938177062569232,
      "language_loss": 0.88310719,
      "learning_rate": 2.699956689131532e-06,
      "loss": 0.90498561,
      "num_input_tokens_seen": 72680345,
      "step": 3365,
      "time_per_iteration": 2.5097124576568604
    },
    {
      "auxiliary_loss_clip": 0.01162785,
      "auxiliary_loss_mlp": 0.01028834,
      "balance_loss_clip": 1.0553143,
      "balance_loss_mlp": 1.02044165,
      "epoch": 0.4047375698911802,
      "flos": 20668135582080.0,
      "grad_norm": 2.7726547696088293,
      "language_loss": 0.84795403,
      "learning_rate": 2.699226928299238e-06,
      "loss": 0.86987019,
      "num_input_tokens_seen": 72698365,
      "step": 3366,
      "time_per_iteration": 2.494499444961548
    },
    {
      "auxiliary_loss_clip": 0.01176626,
      "auxiliary_loss_mlp": 0.01032701,
      "balance_loss_clip": 1.05664659,
      "balance_loss_mlp": 1.02492237,
      "epoch": 0.40485781278181926,
      "flos": 28912865996160.0,
      "grad_norm": 2.337205935409372,
      "language_loss": 0.79043329,
      "learning_rate": 2.698497061394774e-06,
      "loss": 0.81252658,
      "num_input_tokens_seen": 72716850,
      "step": 3367,
      "time_per_iteration": 2.506425380706787
    },
    {
      "auxiliary_loss_clip": 0.01152816,
      "auxiliary_loss_mlp": 0.00762704,
      "balance_loss_clip": 1.05440688,
      "balance_loss_mlp": 1.00016999,
      "epoch": 0.40497805567245837,
      "flos": 23148377694720.0,
      "grad_norm": 3.083649654128188,
      "language_loss": 0.80556291,
      "learning_rate": 2.6977670885288627e-06,
      "loss": 0.82471812,
      "num_input_tokens_seen": 72738250,
      "step": 3368,
      "time_per_iteration": 2.554211139678955
    },
    {
      "auxiliary_loss_clip": 0.01147126,
      "auxiliary_loss_mlp": 0.01032159,
      "balance_loss_clip": 1.05112576,
      "balance_loss_mlp": 1.02369559,
      "epoch": 0.4050982985630975,
      "flos": 16289404030080.0,
      "grad_norm": 1.6620764289472865,
      "language_loss": 0.75343096,
      "learning_rate": 2.6970370098122378e-06,
      "loss": 0.77522385,
      "num_input_tokens_seen": 72755235,
      "step": 3369,
      "time_per_iteration": 2.4633595943450928
    },
    {
      "auxiliary_loss_clip": 0.01186289,
      "auxiliary_loss_mlp": 0.01027713,
      "balance_loss_clip": 1.05559802,
      "balance_loss_mlp": 1.0200001,
      "epoch": 0.40521854145373654,
      "flos": 34459484353920.0,
      "grad_norm": 1.5045027977382917,
      "language_loss": 0.86840677,
      "learning_rate": 2.6963068253556535e-06,
      "loss": 0.8905468,
      "num_input_tokens_seen": 72776620,
      "step": 3370,
      "time_per_iteration": 2.544238328933716
    },
    {
      "auxiliary_loss_clip": 0.01180854,
      "auxiliary_loss_mlp": 0.01031945,
      "balance_loss_clip": 1.05515289,
      "balance_loss_mlp": 1.02268815,
      "epoch": 0.40533878434437565,
      "flos": 25331099454720.0,
      "grad_norm": 2.1361957414435926,
      "language_loss": 0.85629225,
      "learning_rate": 2.6955765352698763e-06,
      "loss": 0.87842023,
      "num_input_tokens_seen": 72796765,
      "step": 3371,
      "time_per_iteration": 2.522141218185425
    },
    {
      "auxiliary_loss_clip": 0.01190627,
      "auxiliary_loss_mlp": 0.01027998,
      "balance_loss_clip": 1.05681491,
      "balance_loss_mlp": 1.01903987,
      "epoch": 0.40545902723501476,
      "flos": 15012061505280.0,
      "grad_norm": 2.0548145858599893,
      "language_loss": 0.73062837,
      "learning_rate": 2.6948461396656923e-06,
      "loss": 0.75281459,
      "num_input_tokens_seen": 72814175,
      "step": 3372,
      "time_per_iteration": 2.4054653644561768
    },
    {
      "auxiliary_loss_clip": 0.01179153,
      "auxiliary_loss_mlp": 0.01032196,
      "balance_loss_clip": 1.05636632,
      "balance_loss_mlp": 1.02388716,
      "epoch": 0.4055792701256538,
      "flos": 25521103422720.0,
      "grad_norm": 2.447431264885515,
      "language_loss": 0.74588716,
      "learning_rate": 2.6941156386539013e-06,
      "loss": 0.7680006,
      "num_input_tokens_seen": 72834125,
      "step": 3373,
      "time_per_iteration": 2.485348701477051
    },
    {
      "auxiliary_loss_clip": 0.01158329,
      "auxiliary_loss_mlp": 0.01033555,
      "balance_loss_clip": 1.05734408,
      "balance_loss_mlp": 1.02516556,
      "epoch": 0.4056995130162929,
      "flos": 19574583972480.0,
      "grad_norm": 2.137650033651343,
      "language_loss": 0.80775881,
      "learning_rate": 2.6933850323453203e-06,
      "loss": 0.82967764,
      "num_input_tokens_seen": 72852570,
      "step": 3374,
      "time_per_iteration": 2.4789068698883057
    },
    {
      "auxiliary_loss_clip": 0.01187417,
      "auxiliary_loss_mlp": 0.01027991,
      "balance_loss_clip": 1.05851412,
      "balance_loss_mlp": 1.01972985,
      "epoch": 0.405819755906932,
      "flos": 15413794191360.0,
      "grad_norm": 2.007796301734701,
      "language_loss": 0.74573904,
      "learning_rate": 2.6926543208507806e-06,
      "loss": 0.76789308,
      "num_input_tokens_seen": 72871250,
      "step": 3375,
      "time_per_iteration": 3.2331912517547607
    },
    {
      "auxiliary_loss_clip": 0.01172506,
      "auxiliary_loss_mlp": 0.01027991,
      "balance_loss_clip": 1.05546772,
      "balance_loss_mlp": 1.01922894,
      "epoch": 0.4059399987975711,
      "flos": 21433930565760.0,
      "grad_norm": 1.9151486571471033,
      "language_loss": 0.8007406,
      "learning_rate": 2.6919235042811316e-06,
      "loss": 0.82274562,
      "num_input_tokens_seen": 72890035,
      "step": 3376,
      "time_per_iteration": 2.4446563720703125
    },
    {
      "auxiliary_loss_clip": 0.01143438,
      "auxiliary_loss_mlp": 0.01033442,
      "balance_loss_clip": 1.05294466,
      "balance_loss_mlp": 1.02448356,
      "epoch": 0.4060602416882102,
      "flos": 25556942217600.0,
      "grad_norm": 3.089289128033694,
      "language_loss": 0.76336491,
      "learning_rate": 2.691192582747237e-06,
      "loss": 0.78513372,
      "num_input_tokens_seen": 72909665,
      "step": 3377,
      "time_per_iteration": 2.5499448776245117
    },
    {
      "auxiliary_loss_clip": 0.01191227,
      "auxiliary_loss_mlp": 0.01025928,
      "balance_loss_clip": 1.05952299,
      "balance_loss_mlp": 1.01813221,
      "epoch": 0.40618048457884925,
      "flos": 23766759262080.0,
      "grad_norm": 1.7072330383906646,
      "language_loss": 0.73954666,
      "learning_rate": 2.6904615563599765e-06,
      "loss": 0.76171815,
      "num_input_tokens_seen": 72929465,
      "step": 3378,
      "time_per_iteration": 2.4370408058166504
    },
    {
      "auxiliary_loss_clip": 0.01138206,
      "auxiliary_loss_mlp": 0.01025638,
      "balance_loss_clip": 1.0499233,
      "balance_loss_mlp": 1.0176599,
      "epoch": 0.40630072746948837,
      "flos": 17639681120640.0,
      "grad_norm": 2.5686773493979973,
      "language_loss": 0.83398104,
      "learning_rate": 2.6897304252302477e-06,
      "loss": 0.85561949,
      "num_input_tokens_seen": 72946785,
      "step": 3379,
      "time_per_iteration": 3.4193224906921387
    },
    {
      "auxiliary_loss_clip": 0.01047831,
      "auxiliary_loss_mlp": 0.01000571,
      "balance_loss_clip": 1.01428914,
      "balance_loss_mlp": 0.99953955,
      "epoch": 0.4064209703601275,
      "flos": 60836053063680.0,
      "grad_norm": 0.7849656510100202,
      "language_loss": 0.54760706,
      "learning_rate": 2.688999189468962e-06,
      "loss": 0.56809115,
      "num_input_tokens_seen": 73003215,
      "step": 3380,
      "time_per_iteration": 2.9865474700927734
    },
    {
      "auxiliary_loss_clip": 0.01172775,
      "auxiliary_loss_mlp": 0.01033578,
      "balance_loss_clip": 1.0563426,
      "balance_loss_mlp": 1.02579331,
      "epoch": 0.40654121325076653,
      "flos": 24024346669440.0,
      "grad_norm": 2.5553775502011926,
      "language_loss": 0.76012647,
      "learning_rate": 2.6882678491870464e-06,
      "loss": 0.78218997,
      "num_input_tokens_seen": 73023650,
      "step": 3381,
      "time_per_iteration": 3.308553695678711
    },
    {
      "auxiliary_loss_clip": 0.01176911,
      "auxiliary_loss_mlp": 0.01024662,
      "balance_loss_clip": 1.05644083,
      "balance_loss_mlp": 1.01563787,
      "epoch": 0.40666145614140564,
      "flos": 27344252085120.0,
      "grad_norm": 1.6598282134544586,
      "language_loss": 0.71465707,
      "learning_rate": 2.6875364044954453e-06,
      "loss": 0.73667282,
      "num_input_tokens_seen": 73043880,
      "step": 3382,
      "time_per_iteration": 2.5174951553344727
    },
    {
      "auxiliary_loss_clip": 0.01153549,
      "auxiliary_loss_mlp": 0.01027292,
      "balance_loss_clip": 1.0483793,
      "balance_loss_mlp": 1.01939476,
      "epoch": 0.40678169903204475,
      "flos": 26176724415360.0,
      "grad_norm": 1.5373131935642868,
      "language_loss": 0.82119489,
      "learning_rate": 2.6868048555051185e-06,
      "loss": 0.84300327,
      "num_input_tokens_seen": 73065410,
      "step": 3383,
      "time_per_iteration": 2.553271770477295
    },
    {
      "auxiliary_loss_clip": 0.01164721,
      "auxiliary_loss_mlp": 0.01032917,
      "balance_loss_clip": 1.05202174,
      "balance_loss_mlp": 1.02447939,
      "epoch": 0.4069019419226838,
      "flos": 28622420622720.0,
      "grad_norm": 2.7480694585289087,
      "language_loss": 0.85636979,
      "learning_rate": 2.686073202327041e-06,
      "loss": 0.87834615,
      "num_input_tokens_seen": 73084410,
      "step": 3384,
      "time_per_iteration": 3.3438169956207275
    },
    {
      "auxiliary_loss_clip": 0.01147696,
      "auxiliary_loss_mlp": 0.0103518,
      "balance_loss_clip": 1.04973865,
      "balance_loss_mlp": 1.02645969,
      "epoch": 0.4070221848133229,
      "flos": 25229006023680.0,
      "grad_norm": 1.671546420932362,
      "language_loss": 0.73064798,
      "learning_rate": 2.6853414450722043e-06,
      "loss": 0.75247675,
      "num_input_tokens_seen": 73104075,
      "step": 3385,
      "time_per_iteration": 2.544618844985962
    },
    {
      "auxiliary_loss_clip": 0.01171394,
      "auxiliary_loss_mlp": 0.01027367,
      "balance_loss_clip": 1.05489922,
      "balance_loss_mlp": 1.01890278,
      "epoch": 0.40714242770396203,
      "flos": 18405224709120.0,
      "grad_norm": 1.7653393830609239,
      "language_loss": 0.85523063,
      "learning_rate": 2.684609583851616e-06,
      "loss": 0.87721825,
      "num_input_tokens_seen": 73122250,
      "step": 3386,
      "time_per_iteration": 2.448624849319458
    },
    {
      "auxiliary_loss_clip": 0.01130067,
      "auxiliary_loss_mlp": 0.01030279,
      "balance_loss_clip": 1.05005169,
      "balance_loss_mlp": 1.02222037,
      "epoch": 0.4072626705946011,
      "flos": 30228920403840.0,
      "grad_norm": 1.5110664228785189,
      "language_loss": 0.80705535,
      "learning_rate": 2.683877618776297e-06,
      "loss": 0.82865882,
      "num_input_tokens_seen": 73144505,
      "step": 3387,
      "time_per_iteration": 2.6386239528656006
    },
    {
      "auxiliary_loss_clip": 0.01151296,
      "auxiliary_loss_mlp": 0.01033124,
      "balance_loss_clip": 1.04902768,
      "balance_loss_mlp": 1.02413559,
      "epoch": 0.4073829134852402,
      "flos": 21834549930240.0,
      "grad_norm": 2.329699160904908,
      "language_loss": 0.73962247,
      "learning_rate": 2.6831455499572876e-06,
      "loss": 0.76146668,
      "num_input_tokens_seen": 73162440,
      "step": 3388,
      "time_per_iteration": 2.504287004470825
    },
    {
      "auxiliary_loss_clip": 0.01189,
      "auxiliary_loss_mlp": 0.01028595,
      "balance_loss_clip": 1.05571914,
      "balance_loss_mlp": 1.02000022,
      "epoch": 0.40750315637587925,
      "flos": 25260211964160.0,
      "grad_norm": 1.7834216739595714,
      "language_loss": 0.77691239,
      "learning_rate": 2.682413377505641e-06,
      "loss": 0.79908836,
      "num_input_tokens_seen": 73181245,
      "step": 3389,
      "time_per_iteration": 2.4684269428253174
    },
    {
      "auxiliary_loss_clip": 0.01173233,
      "auxiliary_loss_mlp": 0.01028147,
      "balance_loss_clip": 1.0538801,
      "balance_loss_mlp": 1.01969528,
      "epoch": 0.40762339926651836,
      "flos": 19712767593600.0,
      "grad_norm": 2.2829060372494556,
      "language_loss": 0.7656498,
      "learning_rate": 2.6816811015324284e-06,
      "loss": 0.78766358,
      "num_input_tokens_seen": 73199295,
      "step": 3390,
      "time_per_iteration": 2.4656755924224854
    },
    {
      "auxiliary_loss_clip": 0.01080487,
      "auxiliary_loss_mlp": 0.01000646,
      "balance_loss_clip": 1.0143044,
      "balance_loss_mlp": 0.99968034,
      "epoch": 0.40774364215715747,
      "flos": 71449307314560.0,
      "grad_norm": 0.7269881696067815,
      "language_loss": 0.56734681,
      "learning_rate": 2.6809487221487343e-06,
      "loss": 0.58815813,
      "num_input_tokens_seen": 73258780,
      "step": 3391,
      "time_per_iteration": 2.9244205951690674
    },
    {
      "auxiliary_loss_clip": 0.01164119,
      "auxiliary_loss_mlp": 0.01022943,
      "balance_loss_clip": 1.05148816,
      "balance_loss_mlp": 1.01450348,
      "epoch": 0.4078638850477965,
      "flos": 15084134144640.0,
      "grad_norm": 2.7218408774333094,
      "language_loss": 0.81857866,
      "learning_rate": 2.6802162394656605e-06,
      "loss": 0.84044933,
      "num_input_tokens_seen": 73275490,
      "step": 3392,
      "time_per_iteration": 2.4352774620056152
    },
    {
      "auxiliary_loss_clip": 0.01153499,
      "auxiliary_loss_mlp": 0.01030939,
      "balance_loss_clip": 1.04879951,
      "balance_loss_mlp": 1.0229224,
      "epoch": 0.40798412793843564,
      "flos": 23842890138240.0,
      "grad_norm": 1.6492507823067226,
      "language_loss": 0.71605313,
      "learning_rate": 2.679483653594324e-06,
      "loss": 0.73789752,
      "num_input_tokens_seen": 73297260,
      "step": 3393,
      "time_per_iteration": 2.553133010864258
    },
    {
      "auxiliary_loss_clip": 0.01176084,
      "auxiliary_loss_mlp": 0.01030856,
      "balance_loss_clip": 1.05544233,
      "balance_loss_mlp": 1.02255297,
      "epoch": 0.40810437082907475,
      "flos": 21065774117760.0,
      "grad_norm": 4.107292579256087,
      "language_loss": 0.75852883,
      "learning_rate": 2.678750964645857e-06,
      "loss": 0.78059822,
      "num_input_tokens_seen": 73316340,
      "step": 3394,
      "time_per_iteration": 2.469615936279297
    },
    {
      "auxiliary_loss_clip": 0.0118087,
      "auxiliary_loss_mlp": 0.01033674,
      "balance_loss_clip": 1.06152833,
      "balance_loss_mlp": 1.02521014,
      "epoch": 0.4082246137197138,
      "flos": 11321377948800.0,
      "grad_norm": 2.4602551635508556,
      "language_loss": 0.8330512,
      "learning_rate": 2.6780181727314094e-06,
      "loss": 0.8551966,
      "num_input_tokens_seen": 73331245,
      "step": 3395,
      "time_per_iteration": 2.4695208072662354
    },
    {
      "auxiliary_loss_clip": 0.01146659,
      "auxiliary_loss_mlp": 0.00762777,
      "balance_loss_clip": 1.05050862,
      "balance_loss_mlp": 1.00010419,
      "epoch": 0.4083448566103529,
      "flos": 19062569554560.0,
      "grad_norm": 1.888437769656481,
      "language_loss": 0.78091037,
      "learning_rate": 2.6772852779621435e-06,
      "loss": 0.80000478,
      "num_input_tokens_seen": 73349105,
      "step": 3396,
      "time_per_iteration": 2.50990891456604
    },
    {
      "auxiliary_loss_clip": 0.01170266,
      "auxiliary_loss_mlp": 0.00762535,
      "balance_loss_clip": 1.05872989,
      "balance_loss_mlp": 1.00008106,
      "epoch": 0.408465099500992,
      "flos": 23550254035200.0,
      "grad_norm": 2.3388074612949405,
      "language_loss": 0.86841989,
      "learning_rate": 2.676552280449239e-06,
      "loss": 0.88774788,
      "num_input_tokens_seen": 73368990,
      "step": 3397,
      "time_per_iteration": 2.494401454925537
    },
    {
      "auxiliary_loss_clip": 0.01163707,
      "auxiliary_loss_mlp": 0.0103192,
      "balance_loss_clip": 1.05220544,
      "balance_loss_mlp": 1.02306247,
      "epoch": 0.4085853423916311,
      "flos": 12750012558720.0,
      "grad_norm": 2.2422508009404467,
      "language_loss": 0.76202452,
      "learning_rate": 2.6758191803038917e-06,
      "loss": 0.78398085,
      "num_input_tokens_seen": 73387485,
      "step": 3398,
      "time_per_iteration": 2.436962366104126
    },
    {
      "auxiliary_loss_clip": 0.01107744,
      "auxiliary_loss_mlp": 0.0102908,
      "balance_loss_clip": 1.04918516,
      "balance_loss_mlp": 1.02071166,
      "epoch": 0.4087055852822702,
      "flos": 24353072962560.0,
      "grad_norm": 1.6875299277262752,
      "language_loss": 0.82741207,
      "learning_rate": 2.6750859776373125e-06,
      "loss": 0.84878027,
      "num_input_tokens_seen": 73406940,
      "step": 3399,
      "time_per_iteration": 2.617870569229126
    },
    {
      "auxiliary_loss_clip": 0.01029972,
      "auxiliary_loss_mlp": 0.01006581,
      "balance_loss_clip": 1.01360369,
      "balance_loss_mlp": 1.00530553,
      "epoch": 0.4088258281729093,
      "flos": 66387950720640.0,
      "grad_norm": 0.7750689225180801,
      "language_loss": 0.604289,
      "learning_rate": 2.674352672560727e-06,
      "loss": 0.62465453,
      "num_input_tokens_seen": 73468385,
      "step": 3400,
      "time_per_iteration": 3.161284923553467
    },
    {
      "auxiliary_loss_clip": 0.011438,
      "auxiliary_loss_mlp": 0.01026294,
      "balance_loss_clip": 1.05002904,
      "balance_loss_mlp": 1.01752019,
      "epoch": 0.40894607106354836,
      "flos": 20449260057600.0,
      "grad_norm": 1.896731507224513,
      "language_loss": 0.76985514,
      "learning_rate": 2.673619265185377e-06,
      "loss": 0.79155612,
      "num_input_tokens_seen": 73488225,
      "step": 3401,
      "time_per_iteration": 2.543379306793213
    },
    {
      "auxiliary_loss_clip": 0.01175115,
      "auxiliary_loss_mlp": 0.01035021,
      "balance_loss_clip": 1.05433238,
      "balance_loss_mlp": 1.02643752,
      "epoch": 0.40906631395418747,
      "flos": 27053627143680.0,
      "grad_norm": 1.6220312349977828,
      "language_loss": 0.78030384,
      "learning_rate": 2.672885755622521e-06,
      "loss": 0.80240524,
      "num_input_tokens_seen": 73510640,
      "step": 3402,
      "time_per_iteration": 3.3825464248657227
    },
    {
      "auxiliary_loss_clip": 0.01128153,
      "auxiliary_loss_mlp": 0.01029671,
      "balance_loss_clip": 1.04652154,
      "balance_loss_mlp": 1.02126062,
      "epoch": 0.4091865568448266,
      "flos": 25484151306240.0,
      "grad_norm": 2.122783565191288,
      "language_loss": 0.70295417,
      "learning_rate": 2.67215214398343e-06,
      "loss": 0.72453237,
      "num_input_tokens_seen": 73530655,
      "step": 3403,
      "time_per_iteration": 2.606388807296753
    },
    {
      "auxiliary_loss_clip": 0.01131998,
      "auxiliary_loss_mlp": 0.01031105,
      "balance_loss_clip": 1.04633403,
      "balance_loss_mlp": 1.02207541,
      "epoch": 0.40930679973546563,
      "flos": 28657864368000.0,
      "grad_norm": 2.4252370637363017,
      "language_loss": 0.78026772,
      "learning_rate": 2.671418430379393e-06,
      "loss": 0.80189872,
      "num_input_tokens_seen": 73549340,
      "step": 3404,
      "time_per_iteration": 2.6188619136810303
    },
    {
      "auxiliary_loss_clip": 0.01186454,
      "auxiliary_loss_mlp": 0.0102691,
      "balance_loss_clip": 1.05508924,
      "balance_loss_mlp": 1.01846433,
      "epoch": 0.40942704262610474,
      "flos": 20886292834560.0,
      "grad_norm": 3.0820917807794084,
      "language_loss": 0.82934201,
      "learning_rate": 2.670684614921715e-06,
      "loss": 0.85147572,
      "num_input_tokens_seen": 73568315,
      "step": 3405,
      "time_per_iteration": 2.432753324508667
    },
    {
      "auxiliary_loss_clip": 0.01159272,
      "auxiliary_loss_mlp": 0.01031815,
      "balance_loss_clip": 1.0510478,
      "balance_loss_mlp": 1.02314281,
      "epoch": 0.4095472855167438,
      "flos": 21618080616960.0,
      "grad_norm": 3.0741149815124618,
      "language_loss": 0.69075906,
      "learning_rate": 2.6699506977217128e-06,
      "loss": 0.71266997,
      "num_input_tokens_seen": 73588490,
      "step": 3406,
      "time_per_iteration": 3.3827433586120605
    },
    {
      "auxiliary_loss_clip": 0.01171514,
      "auxiliary_loss_mlp": 0.01026027,
      "balance_loss_clip": 1.05772305,
      "balance_loss_mlp": 1.01787865,
      "epoch": 0.4096675284073829,
      "flos": 27926112499200.0,
      "grad_norm": 2.0136362758093114,
      "language_loss": 0.69915766,
      "learning_rate": 2.6692166788907233e-06,
      "loss": 0.72113311,
      "num_input_tokens_seen": 73608685,
      "step": 3407,
      "time_per_iteration": 2.5616419315338135
    },
    {
      "auxiliary_loss_clip": 0.01159127,
      "auxiliary_loss_mlp": 0.01031652,
      "balance_loss_clip": 1.05208826,
      "balance_loss_mlp": 1.022735,
      "epoch": 0.409787771298022,
      "flos": 19206607092480.0,
      "grad_norm": 1.8887130947309867,
      "language_loss": 0.76980293,
      "learning_rate": 2.6684825585400957e-06,
      "loss": 0.79171073,
      "num_input_tokens_seen": 73627630,
      "step": 3408,
      "time_per_iteration": 3.2694413661956787
    },
    {
      "auxiliary_loss_clip": 0.01053915,
      "auxiliary_loss_mlp": 0.01001098,
      "balance_loss_clip": 1.01382983,
      "balance_loss_mlp": 0.99999523,
      "epoch": 0.4099080141886611,
      "flos": 59269234832640.0,
      "grad_norm": 0.8230670138384596,
      "language_loss": 0.65103191,
      "learning_rate": 2.6677483367811947e-06,
      "loss": 0.67158204,
      "num_input_tokens_seen": 73687670,
      "step": 3409,
      "time_per_iteration": 3.1678109169006348
    },
    {
      "auxiliary_loss_clip": 0.01173871,
      "auxiliary_loss_mlp": 0.01025946,
      "balance_loss_clip": 1.05198944,
      "balance_loss_mlp": 1.0180124,
      "epoch": 0.4100282570793002,
      "flos": 21906443001600.0,
      "grad_norm": 1.6943569846300655,
      "language_loss": 0.75456792,
      "learning_rate": 2.6670140137254028e-06,
      "loss": 0.77656609,
      "num_input_tokens_seen": 73707145,
      "step": 3410,
      "time_per_iteration": 3.225379705429077
    },
    {
      "auxiliary_loss_clip": 0.01125115,
      "auxiliary_loss_mlp": 0.01027262,
      "balance_loss_clip": 1.04895771,
      "balance_loss_mlp": 1.01903653,
      "epoch": 0.4101484999699393,
      "flos": 18551596631040.0,
      "grad_norm": 2.175898670822533,
      "language_loss": 0.89666653,
      "learning_rate": 2.666279589484115e-06,
      "loss": 0.91819024,
      "num_input_tokens_seen": 73725045,
      "step": 3411,
      "time_per_iteration": 2.541153907775879
    },
    {
      "auxiliary_loss_clip": 0.01127034,
      "auxiliary_loss_mlp": 0.01024678,
      "balance_loss_clip": 1.04674268,
      "balance_loss_mlp": 1.01692891,
      "epoch": 0.41026874286057835,
      "flos": 19094529680640.0,
      "grad_norm": 1.9883902105830888,
      "language_loss": 0.80995059,
      "learning_rate": 2.6655450641687435e-06,
      "loss": 0.83146769,
      "num_input_tokens_seen": 73742610,
      "step": 3412,
      "time_per_iteration": 2.542750597000122
    },
    {
      "auxiliary_loss_clip": 0.01187967,
      "auxiliary_loss_mlp": 0.01029251,
      "balance_loss_clip": 1.05880988,
      "balance_loss_mlp": 1.02080488,
      "epoch": 0.41038898575121746,
      "flos": 31209568588800.0,
      "grad_norm": 1.64470787294491,
      "language_loss": 0.69458604,
      "learning_rate": 2.664810437890715e-06,
      "loss": 0.71675819,
      "num_input_tokens_seen": 73764280,
      "step": 3413,
      "time_per_iteration": 2.5264055728912354
    },
    {
      "auxiliary_loss_clip": 0.01109018,
      "auxiliary_loss_mlp": 0.01028718,
      "balance_loss_clip": 1.0532831,
      "balance_loss_mlp": 1.02134216,
      "epoch": 0.41050922864185657,
      "flos": 14355865895040.0,
      "grad_norm": 4.289189759480733,
      "language_loss": 0.79475296,
      "learning_rate": 2.6640757107614714e-06,
      "loss": 0.81613034,
      "num_input_tokens_seen": 73782375,
      "step": 3414,
      "time_per_iteration": 2.571340799331665
    },
    {
      "auxiliary_loss_clip": 0.0113986,
      "auxiliary_loss_mlp": 0.01026666,
      "balance_loss_clip": 1.05248928,
      "balance_loss_mlp": 1.01807749,
      "epoch": 0.4106294715324956,
      "flos": 30956290813440.0,
      "grad_norm": 2.09267630270332,
      "language_loss": 0.68926632,
      "learning_rate": 2.6633408828924697e-06,
      "loss": 0.7109316,
      "num_input_tokens_seen": 73801240,
      "step": 3415,
      "time_per_iteration": 2.6018478870391846
    },
    {
      "auxiliary_loss_clip": 0.01153017,
      "auxiliary_loss_mlp": 0.01036531,
      "balance_loss_clip": 1.05497253,
      "balance_loss_mlp": 1.02856743,
      "epoch": 0.41074971442313474,
      "flos": 24457321209600.0,
      "grad_norm": 1.5546567530053277,
      "language_loss": 0.7016114,
      "learning_rate": 2.662605954395185e-06,
      "loss": 0.72350681,
      "num_input_tokens_seen": 73821200,
      "step": 3416,
      "time_per_iteration": 2.5898351669311523
    },
    {
      "auxiliary_loss_clip": 0.01173896,
      "auxiliary_loss_mlp": 0.01025115,
      "balance_loss_clip": 1.0532558,
      "balance_loss_mlp": 1.01743186,
      "epoch": 0.41086995731377385,
      "flos": 21542991235200.0,
      "grad_norm": 1.7748157847477455,
      "language_loss": 0.83809614,
      "learning_rate": 2.6618709253811027e-06,
      "loss": 0.86008626,
      "num_input_tokens_seen": 73840655,
      "step": 3417,
      "time_per_iteration": 2.5107195377349854
    },
    {
      "auxiliary_loss_clip": 0.01185247,
      "auxiliary_loss_mlp": 0.01026453,
      "balance_loss_clip": 1.05799937,
      "balance_loss_mlp": 1.01951528,
      "epoch": 0.4109902002044129,
      "flos": 20702753314560.0,
      "grad_norm": 1.6345945953980663,
      "language_loss": 0.87304175,
      "learning_rate": 2.6611357959617277e-06,
      "loss": 0.89515877,
      "num_input_tokens_seen": 73860275,
      "step": 3418,
      "time_per_iteration": 2.4438281059265137
    },
    {
      "auxiliary_loss_clip": 0.01138697,
      "auxiliary_loss_mlp": 0.01035899,
      "balance_loss_clip": 1.04998684,
      "balance_loss_mlp": 1.02713096,
      "epoch": 0.411110443095052,
      "flos": 18179992477440.0,
      "grad_norm": 1.7857831501145531,
      "language_loss": 0.90923345,
      "learning_rate": 2.660400566248578e-06,
      "loss": 0.93097949,
      "num_input_tokens_seen": 73878400,
      "step": 3419,
      "time_per_iteration": 2.5350852012634277
    },
    {
      "auxiliary_loss_clip": 0.01144094,
      "auxiliary_loss_mlp": 0.0103262,
      "balance_loss_clip": 1.04989147,
      "balance_loss_mlp": 1.0232445,
      "epoch": 0.41123068598569107,
      "flos": 14575244209920.0,
      "grad_norm": 2.2793699006355044,
      "language_loss": 0.66624498,
      "learning_rate": 2.6596652363531876e-06,
      "loss": 0.68801212,
      "num_input_tokens_seen": 73894275,
      "step": 3420,
      "time_per_iteration": 2.5832343101501465
    },
    {
      "auxiliary_loss_clip": 0.01184643,
      "auxiliary_loss_mlp": 0.01026623,
      "balance_loss_clip": 1.05510128,
      "balance_loss_mlp": 1.01871634,
      "epoch": 0.4113509288763302,
      "flos": 21177995184000.0,
      "grad_norm": 1.4748049450188518,
      "language_loss": 0.78456891,
      "learning_rate": 2.6589298063871055e-06,
      "loss": 0.80668163,
      "num_input_tokens_seen": 73914450,
      "step": 3421,
      "time_per_iteration": 2.444382429122925
    },
    {
      "auxiliary_loss_clip": 0.0118418,
      "auxiliary_loss_mlp": 0.0102781,
      "balance_loss_clip": 1.05450559,
      "balance_loss_mlp": 1.01916766,
      "epoch": 0.4114711717669693,
      "flos": 18442212739200.0,
      "grad_norm": 1.898926543113717,
      "language_loss": 0.69418907,
      "learning_rate": 2.658194276461895e-06,
      "loss": 0.71630895,
      "num_input_tokens_seen": 73932375,
      "step": 3422,
      "time_per_iteration": 2.4294989109039307
    },
    {
      "auxiliary_loss_clip": 0.01159555,
      "auxiliary_loss_mlp": 0.01028998,
      "balance_loss_clip": 1.05047846,
      "balance_loss_mlp": 1.01993859,
      "epoch": 0.41159141465760835,
      "flos": 27233395735680.0,
      "grad_norm": 2.1380857066121157,
      "language_loss": 0.67110312,
      "learning_rate": 2.6574586466891368e-06,
      "loss": 0.69298863,
      "num_input_tokens_seen": 73952850,
      "step": 3423,
      "time_per_iteration": 2.564086675643921
    },
    {
      "auxiliary_loss_clip": 0.01155646,
      "auxiliary_loss_mlp": 0.00762543,
      "balance_loss_clip": 1.05102348,
      "balance_loss_mlp": 1.00011778,
      "epoch": 0.41171165754824746,
      "flos": 20006876154240.0,
      "grad_norm": 3.1246470547510246,
      "language_loss": 0.64620644,
      "learning_rate": 2.6567229171804247e-06,
      "loss": 0.66538835,
      "num_input_tokens_seen": 73970735,
      "step": 3424,
      "time_per_iteration": 2.495319128036499
    },
    {
      "auxiliary_loss_clip": 0.01149574,
      "auxiliary_loss_mlp": 0.01038115,
      "balance_loss_clip": 1.0475421,
      "balance_loss_mlp": 1.02885866,
      "epoch": 0.41183190043888657,
      "flos": 18004318035840.0,
      "grad_norm": 2.5008071495225725,
      "language_loss": 0.87688345,
      "learning_rate": 2.655987088047368e-06,
      "loss": 0.89876032,
      "num_input_tokens_seen": 73989080,
      "step": 3425,
      "time_per_iteration": 2.4999303817749023
    },
    {
      "auxiliary_loss_clip": 0.01150703,
      "auxiliary_loss_mlp": 0.01031948,
      "balance_loss_clip": 1.04946041,
      "balance_loss_mlp": 1.02320433,
      "epoch": 0.4119521433295256,
      "flos": 27163370171520.0,
      "grad_norm": 2.463150626199949,
      "language_loss": 0.78562689,
      "learning_rate": 2.6552511594015912e-06,
      "loss": 0.80745339,
      "num_input_tokens_seen": 74009470,
      "step": 3426,
      "time_per_iteration": 2.5535061359405518
    },
    {
      "auxiliary_loss_clip": 0.01155337,
      "auxiliary_loss_mlp": 0.01028563,
      "balance_loss_clip": 1.04829454,
      "balance_loss_mlp": 1.01948524,
      "epoch": 0.41207238622016473,
      "flos": 15122020014720.0,
      "grad_norm": 1.915374764288529,
      "language_loss": 0.85259998,
      "learning_rate": 2.654515131354735e-06,
      "loss": 0.874439,
      "num_input_tokens_seen": 74027735,
      "step": 3427,
      "time_per_iteration": 2.483039379119873
    },
    {
      "auxiliary_loss_clip": 0.01143483,
      "auxiliary_loss_mlp": 0.010282,
      "balance_loss_clip": 1.05152774,
      "balance_loss_mlp": 1.02076781,
      "epoch": 0.41219262911080384,
      "flos": 27052872958080.0,
      "grad_norm": 2.4105201343755893,
      "language_loss": 0.84782207,
      "learning_rate": 2.653779004018453e-06,
      "loss": 0.8695389,
      "num_input_tokens_seen": 74048300,
      "step": 3428,
      "time_per_iteration": 3.4264934062957764
    },
    {
      "auxiliary_loss_clip": 0.01149484,
      "auxiliary_loss_mlp": 0.01022603,
      "balance_loss_clip": 1.05072546,
      "balance_loss_mlp": 1.01456213,
      "epoch": 0.4123128720014429,
      "flos": 24686360282880.0,
      "grad_norm": 1.8627355113254442,
      "language_loss": 0.82521957,
      "learning_rate": 2.653042777504417e-06,
      "loss": 0.8469404,
      "num_input_tokens_seen": 74070890,
      "step": 3429,
      "time_per_iteration": 2.5591697692871094
    },
    {
      "auxiliary_loss_clip": 0.01165088,
      "auxiliary_loss_mlp": 0.01025107,
      "balance_loss_clip": 1.05193996,
      "balance_loss_mlp": 1.01668453,
      "epoch": 0.412433114892082,
      "flos": 26244774731520.0,
      "grad_norm": 1.7746401224407709,
      "language_loss": 0.79560387,
      "learning_rate": 2.6523064519243105e-06,
      "loss": 0.81750584,
      "num_input_tokens_seen": 74090460,
      "step": 3430,
      "time_per_iteration": 2.5351781845092773
    },
    {
      "auxiliary_loss_clip": 0.01171847,
      "auxiliary_loss_mlp": 0.01033445,
      "balance_loss_clip": 1.05414796,
      "balance_loss_mlp": 1.024647,
      "epoch": 0.4125533577827211,
      "flos": 21361031913600.0,
      "grad_norm": 2.4688279827538633,
      "language_loss": 0.79005921,
      "learning_rate": 2.6515700273898333e-06,
      "loss": 0.81211209,
      "num_input_tokens_seen": 74108335,
      "step": 3431,
      "time_per_iteration": 2.4668734073638916
    },
    {
      "auxiliary_loss_clip": 0.01146948,
      "auxiliary_loss_mlp": 0.01032093,
      "balance_loss_clip": 1.05278516,
      "balance_loss_mlp": 1.02311659,
      "epoch": 0.4126736006733602,
      "flos": 26067556005120.0,
      "grad_norm": 1.8607911076876527,
      "language_loss": 0.68730193,
      "learning_rate": 2.6508335040127018e-06,
      "loss": 0.70909238,
      "num_input_tokens_seen": 74128030,
      "step": 3432,
      "time_per_iteration": 3.360414981842041
    },
    {
      "auxiliary_loss_clip": 0.01175638,
      "auxiliary_loss_mlp": 0.01029592,
      "balance_loss_clip": 1.05609465,
      "balance_loss_mlp": 1.02158082,
      "epoch": 0.4127938435639993,
      "flos": 25666146541440.0,
      "grad_norm": 1.9936524485733829,
      "language_loss": 0.77026403,
      "learning_rate": 2.6500968819046446e-06,
      "loss": 0.79231644,
      "num_input_tokens_seen": 74148330,
      "step": 3433,
      "time_per_iteration": 2.518282413482666
    },
    {
      "auxiliary_loss_clip": 0.01128342,
      "auxiliary_loss_mlp": 0.01029444,
      "balance_loss_clip": 1.04418397,
      "balance_loss_mlp": 1.02111101,
      "epoch": 0.4129140864546384,
      "flos": 17995914253440.0,
      "grad_norm": 2.3148799500342108,
      "language_loss": 0.58738357,
      "learning_rate": 2.649360161177408e-06,
      "loss": 0.60896146,
      "num_input_tokens_seen": 74163390,
      "step": 3434,
      "time_per_iteration": 2.4858314990997314
    },
    {
      "auxiliary_loss_clip": 0.01178267,
      "auxiliary_loss_mlp": 0.0102849,
      "balance_loss_clip": 1.05378246,
      "balance_loss_mlp": 1.02030063,
      "epoch": 0.41303432934527745,
      "flos": 23732895715200.0,
      "grad_norm": 2.182380081991096,
      "language_loss": 0.73062766,
      "learning_rate": 2.6486233419427504e-06,
      "loss": 0.7526952,
      "num_input_tokens_seen": 74183205,
      "step": 3435,
      "time_per_iteration": 3.205916166305542
    },
    {
      "auxiliary_loss_clip": 0.01133665,
      "auxiliary_loss_mlp": 0.01025716,
      "balance_loss_clip": 1.04971623,
      "balance_loss_mlp": 1.0169239,
      "epoch": 0.41315457223591656,
      "flos": 19755286318080.0,
      "grad_norm": 2.440945576826226,
      "language_loss": 0.75222224,
      "learning_rate": 2.6478864243124484e-06,
      "loss": 0.77381611,
      "num_input_tokens_seen": 74202870,
      "step": 3436,
      "time_per_iteration": 2.5269200801849365
    },
    {
      "auxiliary_loss_clip": 0.01170382,
      "auxiliary_loss_mlp": 0.01022709,
      "balance_loss_clip": 1.05038667,
      "balance_loss_mlp": 1.01523471,
      "epoch": 0.4132748151265556,
      "flos": 20923316778240.0,
      "grad_norm": 3.817682558471954,
      "language_loss": 0.8523711,
      "learning_rate": 2.6471494083982903e-06,
      "loss": 0.87430197,
      "num_input_tokens_seen": 74222255,
      "step": 3437,
      "time_per_iteration": 3.264099597930908
    },
    {
      "auxiliary_loss_clip": 0.01144507,
      "auxiliary_loss_mlp": 0.0102554,
      "balance_loss_clip": 1.0489893,
      "balance_loss_mlp": 1.01801169,
      "epoch": 0.4133950580171947,
      "flos": 32232520016640.0,
      "grad_norm": 1.6900278272702725,
      "language_loss": 0.74777019,
      "learning_rate": 2.6464122943120818e-06,
      "loss": 0.76947063,
      "num_input_tokens_seen": 74242480,
      "step": 3438,
      "time_per_iteration": 2.6100597381591797
    },
    {
      "auxiliary_loss_clip": 0.01142309,
      "auxiliary_loss_mlp": 0.01025588,
      "balance_loss_clip": 1.05198514,
      "balance_loss_mlp": 1.01710606,
      "epoch": 0.41351530090783384,
      "flos": 23292487059840.0,
      "grad_norm": 2.8451429782575164,
      "language_loss": 0.82532728,
      "learning_rate": 2.645675082165642e-06,
      "loss": 0.8470062,
      "num_input_tokens_seen": 74258690,
      "step": 3439,
      "time_per_iteration": 2.526207447052002
    },
    {
      "auxiliary_loss_clip": 0.0115597,
      "auxiliary_loss_mlp": 0.01030741,
      "balance_loss_clip": 1.05181456,
      "balance_loss_mlp": 1.0220449,
      "epoch": 0.4136355437984729,
      "flos": 25593571111680.0,
      "grad_norm": 1.9902155610622383,
      "language_loss": 0.75291467,
      "learning_rate": 2.644937772070806e-06,
      "loss": 0.77478182,
      "num_input_tokens_seen": 74277135,
      "step": 3440,
      "time_per_iteration": 2.5322275161743164
    },
    {
      "auxiliary_loss_clip": 0.01185336,
      "auxiliary_loss_mlp": 0.0102605,
      "balance_loss_clip": 1.05458558,
      "balance_loss_mlp": 1.01757407,
      "epoch": 0.413755786689112,
      "flos": 19828615933440.0,
      "grad_norm": 2.225910037209359,
      "language_loss": 0.83527458,
      "learning_rate": 2.6442003641394225e-06,
      "loss": 0.8573885,
      "num_input_tokens_seen": 74294730,
      "step": 3441,
      "time_per_iteration": 2.4213078022003174
    },
    {
      "auxiliary_loss_clip": 0.01153504,
      "auxiliary_loss_mlp": 0.01027351,
      "balance_loss_clip": 1.04937196,
      "balance_loss_mlp": 1.01957822,
      "epoch": 0.4138760295797511,
      "flos": 26870446759680.0,
      "grad_norm": 1.564187576833067,
      "language_loss": 0.83831024,
      "learning_rate": 2.643462858483356e-06,
      "loss": 0.86011875,
      "num_input_tokens_seen": 74315015,
      "step": 3442,
      "time_per_iteration": 2.5670652389526367
    },
    {
      "auxiliary_loss_clip": 0.01125422,
      "auxiliary_loss_mlp": 0.01031912,
      "balance_loss_clip": 1.04939699,
      "balance_loss_mlp": 1.02306676,
      "epoch": 0.41399627247039017,
      "flos": 16399254798720.0,
      "grad_norm": 1.8494849239088171,
      "language_loss": 0.72705418,
      "learning_rate": 2.6427252552144856e-06,
      "loss": 0.74862754,
      "num_input_tokens_seen": 74333665,
      "step": 3443,
      "time_per_iteration": 2.5562806129455566
    },
    {
      "auxiliary_loss_clip": 0.0118492,
      "auxiliary_loss_mlp": 0.01033638,
      "balance_loss_clip": 1.05346036,
      "balance_loss_mlp": 1.02501965,
      "epoch": 0.4141165153610293,
      "flos": 22930220442240.0,
      "grad_norm": 2.140378889839288,
      "language_loss": 0.75185335,
      "learning_rate": 2.6419875544447044e-06,
      "loss": 0.77403891,
      "num_input_tokens_seen": 74355065,
      "step": 3444,
      "time_per_iteration": 2.5564467906951904
    },
    {
      "auxiliary_loss_clip": 0.01185332,
      "auxiliary_loss_mlp": 0.01032517,
      "balance_loss_clip": 1.05342913,
      "balance_loss_mlp": 1.02368903,
      "epoch": 0.4142367582516684,
      "flos": 25192556697600.0,
      "grad_norm": 1.764905075102477,
      "language_loss": 0.71788776,
      "learning_rate": 2.6412497562859218e-06,
      "loss": 0.74006623,
      "num_input_tokens_seen": 74376345,
      "step": 3445,
      "time_per_iteration": 2.4802005290985107
    },
    {
      "auxiliary_loss_clip": 0.01173914,
      "auxiliary_loss_mlp": 0.01028674,
      "balance_loss_clip": 1.05203605,
      "balance_loss_mlp": 1.02009642,
      "epoch": 0.41435700114230745,
      "flos": 21690476478720.0,
      "grad_norm": 2.1571402197682286,
      "language_loss": 0.76271755,
      "learning_rate": 2.6405118608500617e-06,
      "loss": 0.78474349,
      "num_input_tokens_seen": 74395170,
      "step": 3446,
      "time_per_iteration": 2.4645159244537354
    },
    {
      "auxiliary_loss_clip": 0.01136026,
      "auxiliary_loss_mlp": 0.01026862,
      "balance_loss_clip": 1.0526613,
      "balance_loss_mlp": 1.01902962,
      "epoch": 0.41447724403294656,
      "flos": 25995160143360.0,
      "grad_norm": 2.178869720562306,
      "language_loss": 0.81265402,
      "learning_rate": 2.6397738682490613e-06,
      "loss": 0.83428293,
      "num_input_tokens_seen": 74416070,
      "step": 3447,
      "time_per_iteration": 2.579881191253662
    },
    {
      "auxiliary_loss_clip": 0.01183748,
      "auxiliary_loss_mlp": 0.01026675,
      "balance_loss_clip": 1.05414104,
      "balance_loss_mlp": 1.01855087,
      "epoch": 0.41459748692358567,
      "flos": 18259678800000.0,
      "grad_norm": 2.4376251487091425,
      "language_loss": 0.75320101,
      "learning_rate": 2.6390357785948734e-06,
      "loss": 0.77530515,
      "num_input_tokens_seen": 74433185,
      "step": 3448,
      "time_per_iteration": 2.4142868518829346
    },
    {
      "auxiliary_loss_clip": 0.01171666,
      "auxiliary_loss_mlp": 0.01032044,
      "balance_loss_clip": 1.0545063,
      "balance_loss_mlp": 1.02364588,
      "epoch": 0.4147177298142247,
      "flos": 24168456034560.0,
      "grad_norm": 1.705961680271532,
      "language_loss": 0.80198503,
      "learning_rate": 2.6382975919994667e-06,
      "loss": 0.82402205,
      "num_input_tokens_seen": 74453760,
      "step": 3449,
      "time_per_iteration": 2.4977855682373047
    },
    {
      "auxiliary_loss_clip": 0.01157141,
      "auxiliary_loss_mlp": 0.01026213,
      "balance_loss_clip": 1.05161262,
      "balance_loss_mlp": 1.01876807,
      "epoch": 0.41483797270486383,
      "flos": 20084659056000.0,
      "grad_norm": 1.5213965130436213,
      "language_loss": 0.72941345,
      "learning_rate": 2.637559308574822e-06,
      "loss": 0.75124699,
      "num_input_tokens_seen": 74473505,
      "step": 3450,
      "time_per_iteration": 2.498060464859009
    },
    {
      "auxiliary_loss_clip": 0.01183914,
      "auxiliary_loss_mlp": 0.01030931,
      "balance_loss_clip": 1.0540688,
      "balance_loss_mlp": 1.02298522,
      "epoch": 0.4149582155955029,
      "flos": 30081040110720.0,
      "grad_norm": 2.125749644174213,
      "language_loss": 0.70971662,
      "learning_rate": 2.6368209284329376e-06,
      "loss": 0.73186505,
      "num_input_tokens_seen": 74494135,
      "step": 3451,
      "time_per_iteration": 2.496692657470703
    },
    {
      "auxiliary_loss_clip": 0.01167887,
      "auxiliary_loss_mlp": 0.01029783,
      "balance_loss_clip": 1.04943669,
      "balance_loss_mlp": 1.02143788,
      "epoch": 0.415078458486142,
      "flos": 16764394504320.0,
      "grad_norm": 1.8856325771597418,
      "language_loss": 0.74971163,
      "learning_rate": 2.636082451685825e-06,
      "loss": 0.77168834,
      "num_input_tokens_seen": 74512335,
      "step": 3452,
      "time_per_iteration": 2.426292657852173
    },
    {
      "auxiliary_loss_clip": 0.01159274,
      "auxiliary_loss_mlp": 0.01025832,
      "balance_loss_clip": 1.05352104,
      "balance_loss_mlp": 1.01776731,
      "epoch": 0.4151987013767811,
      "flos": 26033692458240.0,
      "grad_norm": 1.5080057706844285,
      "language_loss": 0.86405468,
      "learning_rate": 2.6353438784455094e-06,
      "loss": 0.88590568,
      "num_input_tokens_seen": 74535620,
      "step": 3453,
      "time_per_iteration": 2.5559446811676025
    },
    {
      "auxiliary_loss_clip": 0.01154006,
      "auxiliary_loss_mlp": 0.01030048,
      "balance_loss_clip": 1.05315661,
      "balance_loss_mlp": 1.02099359,
      "epoch": 0.41531894426742016,
      "flos": 24608002763520.0,
      "grad_norm": 2.2212199936848736,
      "language_loss": 0.72012186,
      "learning_rate": 2.6346052088240326e-06,
      "loss": 0.74196237,
      "num_input_tokens_seen": 74555140,
      "step": 3454,
      "time_per_iteration": 2.524017333984375
    },
    {
      "auxiliary_loss_clip": 0.01156956,
      "auxiliary_loss_mlp": 0.01029259,
      "balance_loss_clip": 1.04999161,
      "balance_loss_mlp": 1.02067614,
      "epoch": 0.4154391871580593,
      "flos": 14975791747200.0,
      "grad_norm": 2.11453114294564,
      "language_loss": 0.7770049,
      "learning_rate": 2.63386644293345e-06,
      "loss": 0.79886699,
      "num_input_tokens_seen": 74571485,
      "step": 3455,
      "time_per_iteration": 3.2934179306030273
    },
    {
      "auxiliary_loss_clip": 0.0113483,
      "auxiliary_loss_mlp": 0.01029339,
      "balance_loss_clip": 1.04548907,
      "balance_loss_mlp": 1.02170956,
      "epoch": 0.4155594300486984,
      "flos": 14647173194880.0,
      "grad_norm": 2.52187612389324,
      "language_loss": 0.82948858,
      "learning_rate": 2.633127580885833e-06,
      "loss": 0.85113031,
      "num_input_tokens_seen": 74585985,
      "step": 3456,
      "time_per_iteration": 2.497652769088745
    },
    {
      "auxiliary_loss_clip": 0.01183387,
      "auxiliary_loss_mlp": 0.01032019,
      "balance_loss_clip": 1.05622053,
      "balance_loss_mlp": 1.02385318,
      "epoch": 0.41567967293933744,
      "flos": 29497276275840.0,
      "grad_norm": 2.0215106515898817,
      "language_loss": 0.64998156,
      "learning_rate": 2.632388622793265e-06,
      "loss": 0.67213559,
      "num_input_tokens_seen": 74605140,
      "step": 3457,
      "time_per_iteration": 2.494825839996338
    },
    {
      "auxiliary_loss_clip": 0.01170144,
      "auxiliary_loss_mlp": 0.01029691,
      "balance_loss_clip": 1.05432391,
      "balance_loss_mlp": 1.02155519,
      "epoch": 0.41579991582997655,
      "flos": 19238387650560.0,
      "grad_norm": 1.7349252082218465,
      "language_loss": 0.67804873,
      "learning_rate": 2.6316495687678457e-06,
      "loss": 0.70004714,
      "num_input_tokens_seen": 74623790,
      "step": 3458,
      "time_per_iteration": 2.468646287918091
    },
    {
      "auxiliary_loss_clip": 0.01120027,
      "auxiliary_loss_mlp": 0.01024914,
      "balance_loss_clip": 1.04591346,
      "balance_loss_mlp": 1.01664686,
      "epoch": 0.41592015872061566,
      "flos": 24462061804800.0,
      "grad_norm": 2.88710394098374,
      "language_loss": 0.76728904,
      "learning_rate": 2.6309104189216887e-06,
      "loss": 0.78873843,
      "num_input_tokens_seen": 74641355,
      "step": 3459,
      "time_per_iteration": 3.4454345703125
    },
    {
      "auxiliary_loss_clip": 0.01128556,
      "auxiliary_loss_mlp": 0.00763053,
      "balance_loss_clip": 1.0470171,
      "balance_loss_mlp": 1.00022221,
      "epoch": 0.4160404016112547,
      "flos": 20775651966720.0,
      "grad_norm": 2.385908597627895,
      "language_loss": 0.74884462,
      "learning_rate": 2.630171173366923e-06,
      "loss": 0.76776075,
      "num_input_tokens_seen": 74657155,
      "step": 3460,
      "time_per_iteration": 2.5208818912506104
    },
    {
      "auxiliary_loss_clip": 0.01124696,
      "auxiliary_loss_mlp": 0.01025763,
      "balance_loss_clip": 1.04675901,
      "balance_loss_mlp": 1.0173943,
      "epoch": 0.41616064450189383,
      "flos": 13916462820480.0,
      "grad_norm": 2.436091787653902,
      "language_loss": 0.74587989,
      "learning_rate": 2.629431832215691e-06,
      "loss": 0.76738441,
      "num_input_tokens_seen": 74671960,
      "step": 3461,
      "time_per_iteration": 2.5265612602233887
    },
    {
      "auxiliary_loss_clip": 0.01148906,
      "auxiliary_loss_mlp": 0.01022683,
      "balance_loss_clip": 1.0486927,
      "balance_loss_mlp": 1.01473808,
      "epoch": 0.41628088739253294,
      "flos": 20010826650240.0,
      "grad_norm": 1.5957769260540788,
      "language_loss": 0.87647814,
      "learning_rate": 2.628692395580151e-06,
      "loss": 0.89819407,
      "num_input_tokens_seen": 74692050,
      "step": 3462,
      "time_per_iteration": 3.3256468772888184
    },
    {
      "auxiliary_loss_clip": 0.01095397,
      "auxiliary_loss_mlp": 0.01027883,
      "balance_loss_clip": 1.04392695,
      "balance_loss_mlp": 1.01952004,
      "epoch": 0.416401130283172,
      "flos": 29168801377920.0,
      "grad_norm": 1.9496326518963858,
      "language_loss": 0.79516274,
      "learning_rate": 2.6279528635724747e-06,
      "loss": 0.81639552,
      "num_input_tokens_seen": 74712205,
      "step": 3463,
      "time_per_iteration": 3.423001527786255
    },
    {
      "auxiliary_loss_clip": 0.01169522,
      "auxiliary_loss_mlp": 0.01027616,
      "balance_loss_clip": 1.05216312,
      "balance_loss_mlp": 1.01903939,
      "epoch": 0.4165213731738111,
      "flos": 16246813478400.0,
      "grad_norm": 3.344907597717156,
      "language_loss": 0.78575432,
      "learning_rate": 2.627213236304848e-06,
      "loss": 0.80772573,
      "num_input_tokens_seen": 74729005,
      "step": 3464,
      "time_per_iteration": 2.4564061164855957
    },
    {
      "auxiliary_loss_clip": 0.01173144,
      "auxiliary_loss_mlp": 0.01026151,
      "balance_loss_clip": 1.05422044,
      "balance_loss_mlp": 1.01782405,
      "epoch": 0.4166416160644502,
      "flos": 33765438787200.0,
      "grad_norm": 1.9955075768981707,
      "language_loss": 0.70592636,
      "learning_rate": 2.626473513889472e-06,
      "loss": 0.72791928,
      "num_input_tokens_seen": 74751385,
      "step": 3465,
      "time_per_iteration": 2.591647148132324
    },
    {
      "auxiliary_loss_clip": 0.01159776,
      "auxiliary_loss_mlp": 0.01033299,
      "balance_loss_clip": 1.05077195,
      "balance_loss_mlp": 1.02512693,
      "epoch": 0.41676185895508927,
      "flos": 20917498775040.0,
      "grad_norm": 1.806155721163052,
      "language_loss": 0.83063149,
      "learning_rate": 2.625733696438562e-06,
      "loss": 0.85256219,
      "num_input_tokens_seen": 74768890,
      "step": 3466,
      "time_per_iteration": 2.469381332397461
    },
    {
      "auxiliary_loss_clip": 0.01152033,
      "auxiliary_loss_mlp": 0.01032967,
      "balance_loss_clip": 1.05059648,
      "balance_loss_mlp": 1.0249387,
      "epoch": 0.4168821018457284,
      "flos": 18406122549120.0,
      "grad_norm": 1.6478267641127824,
      "language_loss": 0.75351608,
      "learning_rate": 2.6249937840643476e-06,
      "loss": 0.77536607,
      "num_input_tokens_seen": 74787195,
      "step": 3467,
      "time_per_iteration": 2.4977591037750244
    },
    {
      "auxiliary_loss_clip": 0.01183263,
      "auxiliary_loss_mlp": 0.0076253,
      "balance_loss_clip": 1.05469751,
      "balance_loss_mlp": 1.00021815,
      "epoch": 0.41700234473636744,
      "flos": 18698399516160.0,
      "grad_norm": 1.8950878755130762,
      "language_loss": 0.66750717,
      "learning_rate": 2.6242537768790733e-06,
      "loss": 0.68696517,
      "num_input_tokens_seen": 74806350,
      "step": 3468,
      "time_per_iteration": 2.4356980323791504
    },
    {
      "auxiliary_loss_clip": 0.01170564,
      "auxiliary_loss_mlp": 0.01028495,
      "balance_loss_clip": 1.05308676,
      "balance_loss_mlp": 1.01998937,
      "epoch": 0.41712258762700655,
      "flos": 31033283616000.0,
      "grad_norm": 1.8723668590831297,
      "language_loss": 0.68660635,
      "learning_rate": 2.6235136749949975e-06,
      "loss": 0.70859694,
      "num_input_tokens_seen": 74829800,
      "step": 3469,
      "time_per_iteration": 2.580603837966919
    },
    {
      "auxiliary_loss_clip": 0.01180828,
      "auxiliary_loss_mlp": 0.0102949,
      "balance_loss_clip": 1.05264091,
      "balance_loss_mlp": 1.02044451,
      "epoch": 0.41724283051764566,
      "flos": 35914763877120.0,
      "grad_norm": 2.1710000814226094,
      "language_loss": 0.61695391,
      "learning_rate": 2.6227734785243924e-06,
      "loss": 0.63905704,
      "num_input_tokens_seen": 74849760,
      "step": 3470,
      "time_per_iteration": 2.570401906967163
    },
    {
      "auxiliary_loss_clip": 0.01103893,
      "auxiliary_loss_mlp": 0.01025199,
      "balance_loss_clip": 1.04281986,
      "balance_loss_mlp": 1.0173018,
      "epoch": 0.4173630734082847,
      "flos": 25333649320320.0,
      "grad_norm": 11.067771064294295,
      "language_loss": 0.79270077,
      "learning_rate": 2.6220331875795466e-06,
      "loss": 0.81399167,
      "num_input_tokens_seen": 74869110,
      "step": 3471,
      "time_per_iteration": 2.623163938522339
    },
    {
      "auxiliary_loss_clip": 0.01165239,
      "auxiliary_loss_mlp": 0.01029891,
      "balance_loss_clip": 1.05230427,
      "balance_loss_mlp": 1.02164721,
      "epoch": 0.4174833162989238,
      "flos": 26685398868480.0,
      "grad_norm": 1.653054843406505,
      "language_loss": 0.7503953,
      "learning_rate": 2.62129280227276e-06,
      "loss": 0.77234662,
      "num_input_tokens_seen": 74889110,
      "step": 3472,
      "time_per_iteration": 2.5103964805603027
    },
    {
      "auxiliary_loss_clip": 0.01172667,
      "auxiliary_loss_mlp": 0.01034589,
      "balance_loss_clip": 1.05238307,
      "balance_loss_mlp": 1.0259347,
      "epoch": 0.41760355918956293,
      "flos": 74739584010240.0,
      "grad_norm": 1.9367760155530178,
      "language_loss": 0.68340445,
      "learning_rate": 2.62055232271635e-06,
      "loss": 0.705477,
      "num_input_tokens_seen": 74916260,
      "step": 3473,
      "time_per_iteration": 2.8592793941497803
    },
    {
      "auxiliary_loss_clip": 0.01130314,
      "auxiliary_loss_mlp": 0.01027395,
      "balance_loss_clip": 1.04696918,
      "balance_loss_mlp": 1.01933026,
      "epoch": 0.417723802080202,
      "flos": 14317513148160.0,
      "grad_norm": 1.945403972843583,
      "language_loss": 0.87755072,
      "learning_rate": 2.619811749022646e-06,
      "loss": 0.89912778,
      "num_input_tokens_seen": 74931570,
      "step": 3474,
      "time_per_iteration": 2.4995055198669434
    },
    {
      "auxiliary_loss_clip": 0.01170781,
      "auxiliary_loss_mlp": 0.01031733,
      "balance_loss_clip": 1.05418348,
      "balance_loss_mlp": 1.02320981,
      "epoch": 0.4178440449708411,
      "flos": 14643797316480.0,
      "grad_norm": 2.291969290459892,
      "language_loss": 0.71711504,
      "learning_rate": 2.6190710813039917e-06,
      "loss": 0.73914027,
      "num_input_tokens_seen": 74944695,
      "step": 3475,
      "time_per_iteration": 2.429954767227173
    },
    {
      "auxiliary_loss_clip": 0.01119317,
      "auxiliary_loss_mlp": 0.00763779,
      "balance_loss_clip": 1.04352081,
      "balance_loss_mlp": 1.00018811,
      "epoch": 0.4179642878614802,
      "flos": 21507296094720.0,
      "grad_norm": 2.1346174732178995,
      "language_loss": 0.83984888,
      "learning_rate": 2.618330319672747e-06,
      "loss": 0.85867989,
      "num_input_tokens_seen": 74964115,
      "step": 3476,
      "time_per_iteration": 2.5777759552001953
    },
    {
      "auxiliary_loss_clip": 0.01183848,
      "auxiliary_loss_mlp": 0.01027566,
      "balance_loss_clip": 1.0543288,
      "balance_loss_mlp": 1.01938236,
      "epoch": 0.41808453075211927,
      "flos": 18441997257600.0,
      "grad_norm": 2.360735482633372,
      "language_loss": 0.91842341,
      "learning_rate": 2.617589464241284e-06,
      "loss": 0.94053757,
      "num_input_tokens_seen": 74978515,
      "step": 3477,
      "time_per_iteration": 2.418271780014038
    },
    {
      "auxiliary_loss_clip": 0.01142993,
      "auxiliary_loss_mlp": 0.01023973,
      "balance_loss_clip": 1.05011487,
      "balance_loss_mlp": 1.01650465,
      "epoch": 0.4182047736427584,
      "flos": 20301020628480.0,
      "grad_norm": 1.989406817569932,
      "language_loss": 0.74753785,
      "learning_rate": 2.6168485151219914e-06,
      "loss": 0.76920748,
      "num_input_tokens_seen": 74998135,
      "step": 3478,
      "time_per_iteration": 2.545653820037842
    },
    {
      "auxiliary_loss_clip": 0.01169943,
      "auxiliary_loss_mlp": 0.01026444,
      "balance_loss_clip": 1.05275559,
      "balance_loss_mlp": 1.01809978,
      "epoch": 0.4183250165333975,
      "flos": 18876623823360.0,
      "grad_norm": 7.008066589815805,
      "language_loss": 0.71727669,
      "learning_rate": 2.616107472427269e-06,
      "loss": 0.73924047,
      "num_input_tokens_seen": 75012830,
      "step": 3479,
      "time_per_iteration": 2.4293010234832764
    },
    {
      "auxiliary_loss_clip": 0.01175003,
      "auxiliary_loss_mlp": 0.01025116,
      "balance_loss_clip": 1.05257463,
      "balance_loss_mlp": 1.01675367,
      "epoch": 0.41844525942403654,
      "flos": 17740050698880.0,
      "grad_norm": 2.9363689596502995,
      "language_loss": 0.76565349,
      "learning_rate": 2.615366336269533e-06,
      "loss": 0.78765464,
      "num_input_tokens_seen": 75026495,
      "step": 3480,
      "time_per_iteration": 2.436577796936035
    },
    {
      "auxiliary_loss_clip": 0.01185547,
      "auxiliary_loss_mlp": 0.0103608,
      "balance_loss_clip": 1.05353904,
      "balance_loss_mlp": 1.02694905,
      "epoch": 0.41856550231467565,
      "flos": 18361377181440.0,
      "grad_norm": 2.6185338129962386,
      "language_loss": 0.80420965,
      "learning_rate": 2.6146251067612126e-06,
      "loss": 0.82642591,
      "num_input_tokens_seen": 75041970,
      "step": 3481,
      "time_per_iteration": 2.412691354751587
    },
    {
      "auxiliary_loss_clip": 0.01168067,
      "auxiliary_loss_mlp": 0.01028688,
      "balance_loss_clip": 1.05354917,
      "balance_loss_mlp": 1.02039719,
      "epoch": 0.41868574520531476,
      "flos": 22781801445120.0,
      "grad_norm": 1.5559728647902487,
      "language_loss": 0.82525861,
      "learning_rate": 2.6138837840147525e-06,
      "loss": 0.84722614,
      "num_input_tokens_seen": 75061005,
      "step": 3482,
      "time_per_iteration": 3.317591428756714
    },
    {
      "auxiliary_loss_clip": 0.01138692,
      "auxiliary_loss_mlp": 0.01028026,
      "balance_loss_clip": 1.04896796,
      "balance_loss_mlp": 1.01991975,
      "epoch": 0.4188059880959538,
      "flos": 13699167494400.0,
      "grad_norm": 2.2546182438774176,
      "language_loss": 0.76345217,
      "learning_rate": 2.6131423681426103e-06,
      "loss": 0.7851193,
      "num_input_tokens_seen": 75076920,
      "step": 3483,
      "time_per_iteration": 2.500379800796509
    },
    {
      "auxiliary_loss_clip": 0.01183503,
      "auxiliary_loss_mlp": 0.01025574,
      "balance_loss_clip": 1.0551976,
      "balance_loss_mlp": 1.01827884,
      "epoch": 0.41892623098659293,
      "flos": 37818281220480.0,
      "grad_norm": 1.6276821832869575,
      "language_loss": 0.72819209,
      "learning_rate": 2.6124008592572587e-06,
      "loss": 0.75028282,
      "num_input_tokens_seen": 75100905,
      "step": 3484,
      "time_per_iteration": 2.60031795501709
    },
    {
      "auxiliary_loss_clip": 0.01186815,
      "auxiliary_loss_mlp": 0.01028761,
      "balance_loss_clip": 1.05375266,
      "balance_loss_mlp": 1.01998127,
      "epoch": 0.419046473877232,
      "flos": 23258874908160.0,
      "grad_norm": 3.085197331183568,
      "language_loss": 0.82063305,
      "learning_rate": 2.6116592574711835e-06,
      "loss": 0.84278882,
      "num_input_tokens_seen": 75119205,
      "step": 3485,
      "time_per_iteration": 2.4477596282958984
    },
    {
      "auxiliary_loss_clip": 0.01187818,
      "auxiliary_loss_mlp": 0.01037582,
      "balance_loss_clip": 1.05585194,
      "balance_loss_mlp": 1.02883172,
      "epoch": 0.4191667167678711,
      "flos": 20741034234240.0,
      "grad_norm": 1.7808387905703396,
      "language_loss": 0.84116077,
      "learning_rate": 2.6109175628968853e-06,
      "loss": 0.86341482,
      "num_input_tokens_seen": 75138970,
      "step": 3486,
      "time_per_iteration": 3.312459707260132
    },
    {
      "auxiliary_loss_clip": 0.01160745,
      "auxiliary_loss_mlp": 0.01028275,
      "balance_loss_clip": 1.05087364,
      "balance_loss_mlp": 1.02054977,
      "epoch": 0.4192869596585102,
      "flos": 23586416052480.0,
      "grad_norm": 1.8839716590873883,
      "language_loss": 0.82592857,
      "learning_rate": 2.610175775646878e-06,
      "loss": 0.84781879,
      "num_input_tokens_seen": 75157550,
      "step": 3487,
      "time_per_iteration": 2.485684394836426
    },
    {
      "auxiliary_loss_clip": 0.01152335,
      "auxiliary_loss_mlp": 0.01029144,
      "balance_loss_clip": 1.04782867,
      "balance_loss_mlp": 1.02097869,
      "epoch": 0.41940720254914926,
      "flos": 25081269384960.0,
      "grad_norm": 1.9322295648543002,
      "language_loss": 0.7250151,
      "learning_rate": 2.6094338958336907e-06,
      "loss": 0.74682987,
      "num_input_tokens_seen": 75176220,
      "step": 3488,
      "time_per_iteration": 3.348875045776367
    },
    {
      "auxiliary_loss_clip": 0.01155766,
      "auxiliary_loss_mlp": 0.01027057,
      "balance_loss_clip": 1.05182672,
      "balance_loss_mlp": 1.01898026,
      "epoch": 0.41952744543978837,
      "flos": 15554132628480.0,
      "grad_norm": 2.0032949221501264,
      "language_loss": 0.82409155,
      "learning_rate": 2.608691923569867e-06,
      "loss": 0.84591979,
      "num_input_tokens_seen": 75193095,
      "step": 3489,
      "time_per_iteration": 2.5133297443389893
    },
    {
      "auxiliary_loss_clip": 0.01171894,
      "auxiliary_loss_mlp": 0.01029811,
      "balance_loss_clip": 1.05367529,
      "balance_loss_mlp": 1.02166343,
      "epoch": 0.4196476883304275,
      "flos": 24644775312000.0,
      "grad_norm": 1.6185373153168154,
      "language_loss": 0.75680375,
      "learning_rate": 2.6079498589679616e-06,
      "loss": 0.77882075,
      "num_input_tokens_seen": 75214185,
      "step": 3490,
      "time_per_iteration": 3.2565455436706543
    },
    {
      "auxiliary_loss_clip": 0.01105126,
      "auxiliary_loss_mlp": 0.01034595,
      "balance_loss_clip": 1.0413866,
      "balance_loss_mlp": 1.02512932,
      "epoch": 0.41976793122106654,
      "flos": 24531333183360.0,
      "grad_norm": 2.767977131496339,
      "language_loss": 0.75896966,
      "learning_rate": 2.6072077021405465e-06,
      "loss": 0.7803669,
      "num_input_tokens_seen": 75233020,
      "step": 3491,
      "time_per_iteration": 2.621495485305786
    },
    {
      "auxiliary_loss_clip": 0.01148216,
      "auxiliary_loss_mlp": 0.01033635,
      "balance_loss_clip": 1.04925704,
      "balance_loss_mlp": 1.0259335,
      "epoch": 0.41988817411170565,
      "flos": 21175301664000.0,
      "grad_norm": 1.697596358884439,
      "language_loss": 0.68812382,
      "learning_rate": 2.6064654532002054e-06,
      "loss": 0.70994234,
      "num_input_tokens_seen": 75252030,
      "step": 3492,
      "time_per_iteration": 2.519561767578125
    },
    {
      "auxiliary_loss_clip": 0.01184474,
      "auxiliary_loss_mlp": 0.01033969,
      "balance_loss_clip": 1.05488694,
      "balance_loss_mlp": 1.02591014,
      "epoch": 0.42000841700234476,
      "flos": 31649402626560.0,
      "grad_norm": 3.9879169386297106,
      "language_loss": 0.75748897,
      "learning_rate": 2.6057231122595375e-06,
      "loss": 0.7796734,
      "num_input_tokens_seen": 75273340,
      "step": 3493,
      "time_per_iteration": 2.5316569805145264
    },
    {
      "auxiliary_loss_clip": 0.01155415,
      "auxiliary_loss_mlp": 0.01031245,
      "balance_loss_clip": 1.04857957,
      "balance_loss_mlp": 1.022614,
      "epoch": 0.4201286598929838,
      "flos": 21281525159040.0,
      "grad_norm": 1.6304732473793624,
      "language_loss": 0.72744763,
      "learning_rate": 2.604980679431154e-06,
      "loss": 0.74931419,
      "num_input_tokens_seen": 75291580,
      "step": 3494,
      "time_per_iteration": 2.5114362239837646
    },
    {
      "auxiliary_loss_clip": 0.01170537,
      "auxiliary_loss_mlp": 0.01027123,
      "balance_loss_clip": 1.04991305,
      "balance_loss_mlp": 1.01922584,
      "epoch": 0.4202489027836229,
      "flos": 18546532813440.0,
      "grad_norm": 1.990481141980804,
      "language_loss": 0.7480433,
      "learning_rate": 2.604238154827684e-06,
      "loss": 0.77002001,
      "num_input_tokens_seen": 75308205,
      "step": 3495,
      "time_per_iteration": 2.4571053981781006
    },
    {
      "auxiliary_loss_clip": 0.01169991,
      "auxiliary_loss_mlp": 0.01024945,
      "balance_loss_clip": 1.05244744,
      "balance_loss_mlp": 1.01691628,
      "epoch": 0.42036914567426203,
      "flos": 19317643009920.0,
      "grad_norm": 1.8816238367556346,
      "language_loss": 0.72657776,
      "learning_rate": 2.6034955385617656e-06,
      "loss": 0.74852717,
      "num_input_tokens_seen": 75326535,
      "step": 3496,
      "time_per_iteration": 2.450303554534912
    },
    {
      "auxiliary_loss_clip": 0.01051489,
      "auxiliary_loss_mlp": 0.0100441,
      "balance_loss_clip": 1.01800179,
      "balance_loss_mlp": 1.00332499,
      "epoch": 0.4204893885649011,
      "flos": 67842942935040.0,
      "grad_norm": 0.8101331624344884,
      "language_loss": 0.61675388,
      "learning_rate": 2.6027528307460544e-06,
      "loss": 0.63731289,
      "num_input_tokens_seen": 75390540,
      "step": 3497,
      "time_per_iteration": 3.181201219558716
    },
    {
      "auxiliary_loss_clip": 0.01183969,
      "auxiliary_loss_mlp": 0.01025552,
      "balance_loss_clip": 1.05389047,
      "balance_loss_mlp": 1.01807761,
      "epoch": 0.4206096314555402,
      "flos": 21908777385600.0,
      "grad_norm": 2.066648623615736,
      "language_loss": 0.86420363,
      "learning_rate": 2.602010031493217e-06,
      "loss": 0.88629889,
      "num_input_tokens_seen": 75408770,
      "step": 3498,
      "time_per_iteration": 2.477231740951538
    },
    {
      "auxiliary_loss_clip": 0.01136994,
      "auxiliary_loss_mlp": 0.01032969,
      "balance_loss_clip": 1.04853821,
      "balance_loss_mlp": 1.02462125,
      "epoch": 0.42072987434617926,
      "flos": 29278185269760.0,
      "grad_norm": 1.9352074136477548,
      "language_loss": 0.86761618,
      "learning_rate": 2.6012671409159367e-06,
      "loss": 0.88931578,
      "num_input_tokens_seen": 75430105,
      "step": 3499,
      "time_per_iteration": 2.5866153240203857
    },
    {
      "auxiliary_loss_clip": 0.01150318,
      "auxiliary_loss_mlp": 0.01029862,
      "balance_loss_clip": 1.05144715,
      "balance_loss_mlp": 1.02129054,
      "epoch": 0.42085011723681837,
      "flos": 27600726170880.0,
      "grad_norm": 1.8928820323263145,
      "language_loss": 0.81662506,
      "learning_rate": 2.6005241591269097e-06,
      "loss": 0.83842683,
      "num_input_tokens_seen": 75449475,
      "step": 3500,
      "time_per_iteration": 2.536463975906372
    },
    {
      "auxiliary_loss_clip": 0.01135685,
      "auxiliary_loss_mlp": 0.01030621,
      "balance_loss_clip": 1.05114555,
      "balance_loss_mlp": 1.02265811,
      "epoch": 0.4209703601274575,
      "flos": 27818632028160.0,
      "grad_norm": 1.7220329486448918,
      "language_loss": 0.79491782,
      "learning_rate": 2.5997810862388454e-06,
      "loss": 0.81658089,
      "num_input_tokens_seen": 75469315,
      "step": 3501,
      "time_per_iteration": 2.63895845413208
    },
    {
      "auxiliary_loss_clip": 0.01155156,
      "auxiliary_loss_mlp": 0.01028641,
      "balance_loss_clip": 1.04859161,
      "balance_loss_mlp": 1.02044547,
      "epoch": 0.42109060301809653,
      "flos": 27525529048320.0,
      "grad_norm": 1.9726864276113674,
      "language_loss": 0.75441056,
      "learning_rate": 2.599037922364467e-06,
      "loss": 0.77624857,
      "num_input_tokens_seen": 75488215,
      "step": 3502,
      "time_per_iteration": 2.5483884811401367
    },
    {
      "auxiliary_loss_clip": 0.01135715,
      "auxiliary_loss_mlp": 0.01025802,
      "balance_loss_clip": 1.04949701,
      "balance_loss_mlp": 1.01752949,
      "epoch": 0.42121084590873564,
      "flos": 29314275459840.0,
      "grad_norm": 2.1339329649153487,
      "language_loss": 0.75387096,
      "learning_rate": 2.5982946676165112e-06,
      "loss": 0.77548611,
      "num_input_tokens_seen": 75507985,
      "step": 3503,
      "time_per_iteration": 2.5771071910858154
    },
    {
      "auxiliary_loss_clip": 0.01057201,
      "auxiliary_loss_mlp": 0.01003984,
      "balance_loss_clip": 1.03015924,
      "balance_loss_mlp": 1.00238693,
      "epoch": 0.42133108879937475,
      "flos": 67398835178880.0,
      "grad_norm": 0.7289518483371384,
      "language_loss": 0.57607305,
      "learning_rate": 2.5975513221077313e-06,
      "loss": 0.59668487,
      "num_input_tokens_seen": 75571955,
      "step": 3504,
      "time_per_iteration": 3.1944680213928223
    },
    {
      "auxiliary_loss_clip": 0.01147057,
      "auxiliary_loss_mlp": 0.01032023,
      "balance_loss_clip": 1.05017352,
      "balance_loss_mlp": 1.02370226,
      "epoch": 0.4214513316900138,
      "flos": 23106038538240.0,
      "grad_norm": 2.582364925543384,
      "language_loss": 0.88772297,
      "learning_rate": 2.5968078859508897e-06,
      "loss": 0.90951371,
      "num_input_tokens_seen": 75589155,
      "step": 3505,
      "time_per_iteration": 2.4954302310943604
    },
    {
      "auxiliary_loss_clip": 0.0116895,
      "auxiliary_loss_mlp": 0.01028058,
      "balance_loss_clip": 1.05304158,
      "balance_loss_mlp": 1.02011251,
      "epoch": 0.4215715745806529,
      "flos": 15336190857600.0,
      "grad_norm": 2.0607230537098666,
      "language_loss": 0.80054593,
      "learning_rate": 2.5960643592587673e-06,
      "loss": 0.82251602,
      "num_input_tokens_seen": 75606565,
      "step": 3506,
      "time_per_iteration": 2.4433610439300537
    },
    {
      "auxiliary_loss_clip": 0.01141058,
      "auxiliary_loss_mlp": 0.0102741,
      "balance_loss_clip": 1.04900134,
      "balance_loss_mlp": 1.01985788,
      "epoch": 0.42169181747129203,
      "flos": 22127257860480.0,
      "grad_norm": 2.0123536169178435,
      "language_loss": 0.81577432,
      "learning_rate": 2.5953207421441553e-06,
      "loss": 0.83745903,
      "num_input_tokens_seen": 75625165,
      "step": 3507,
      "time_per_iteration": 2.540236473083496
    },
    {
      "auxiliary_loss_clip": 0.01145165,
      "auxiliary_loss_mlp": 0.01031624,
      "balance_loss_clip": 1.05258,
      "balance_loss_mlp": 1.0239414,
      "epoch": 0.4218120603619311,
      "flos": 22630724841600.0,
      "grad_norm": 2.151817061630336,
      "language_loss": 0.75429857,
      "learning_rate": 2.5945770347198603e-06,
      "loss": 0.77606642,
      "num_input_tokens_seen": 75643320,
      "step": 3508,
      "time_per_iteration": 3.3331243991851807
    },
    {
      "auxiliary_loss_clip": 0.01150369,
      "auxiliary_loss_mlp": 0.01022054,
      "balance_loss_clip": 1.04902506,
      "balance_loss_mlp": 1.01468706,
      "epoch": 0.4219323032525702,
      "flos": 19682818629120.0,
      "grad_norm": 1.6912967643455545,
      "language_loss": 0.81626594,
      "learning_rate": 2.593833237098701e-06,
      "loss": 0.83799016,
      "num_input_tokens_seen": 75660920,
      "step": 3509,
      "time_per_iteration": 2.510230779647827
    },
    {
      "auxiliary_loss_clip": 0.01166519,
      "auxiliary_loss_mlp": 0.01029738,
      "balance_loss_clip": 1.04867649,
      "balance_loss_mlp": 1.02115536,
      "epoch": 0.4220525461432093,
      "flos": 30190747224960.0,
      "grad_norm": 1.8069589798945016,
      "language_loss": 0.6215992,
      "learning_rate": 2.593089349393512e-06,
      "loss": 0.64356172,
      "num_input_tokens_seen": 75681410,
      "step": 3510,
      "time_per_iteration": 2.546098470687866
    },
    {
      "auxiliary_loss_clip": 0.01170095,
      "auxiliary_loss_mlp": 0.01025437,
      "balance_loss_clip": 1.05638552,
      "balance_loss_mlp": 1.01773643,
      "epoch": 0.42217278903384836,
      "flos": 24315941278080.0,
      "grad_norm": 2.119552249181925,
      "language_loss": 0.83804792,
      "learning_rate": 2.592345371717141e-06,
      "loss": 0.86000323,
      "num_input_tokens_seen": 75700940,
      "step": 3511,
      "time_per_iteration": 2.493624448776245
    },
    {
      "auxiliary_loss_clip": 0.01172858,
      "auxiliary_loss_mlp": 0.01029833,
      "balance_loss_clip": 1.05883813,
      "balance_loss_mlp": 1.02148247,
      "epoch": 0.42229303192448747,
      "flos": 17092474352640.0,
      "grad_norm": 2.330988490580442,
      "language_loss": 0.71804029,
      "learning_rate": 2.591601304182448e-06,
      "loss": 0.74006718,
      "num_input_tokens_seen": 75718910,
      "step": 3512,
      "time_per_iteration": 3.300139904022217
    },
    {
      "auxiliary_loss_clip": 0.01156208,
      "auxiliary_loss_mlp": 0.01027477,
      "balance_loss_clip": 1.05470657,
      "balance_loss_mlp": 1.02029443,
      "epoch": 0.4224132748151266,
      "flos": 22784530878720.0,
      "grad_norm": 1.7998042832617744,
      "language_loss": 0.79349375,
      "learning_rate": 2.5908571469023067e-06,
      "loss": 0.81533056,
      "num_input_tokens_seen": 75738395,
      "step": 3513,
      "time_per_iteration": 2.5223584175109863
    },
    {
      "auxiliary_loss_clip": 0.01184027,
      "auxiliary_loss_mlp": 0.01030342,
      "balance_loss_clip": 1.05492163,
      "balance_loss_mlp": 1.02240252,
      "epoch": 0.42253351770576564,
      "flos": 17819090576640.0,
      "grad_norm": 2.464266565644855,
      "language_loss": 0.75665057,
      "learning_rate": 2.5901128999896067e-06,
      "loss": 0.77879429,
      "num_input_tokens_seen": 75753825,
      "step": 3514,
      "time_per_iteration": 2.401898145675659
    },
    {
      "auxiliary_loss_clip": 0.01170024,
      "auxiliary_loss_mlp": 0.01027642,
      "balance_loss_clip": 1.05601716,
      "balance_loss_mlp": 1.01997733,
      "epoch": 0.42265376059640475,
      "flos": 28512390286080.0,
      "grad_norm": 1.591482796493947,
      "language_loss": 0.67968857,
      "learning_rate": 2.5893685635572487e-06,
      "loss": 0.70166528,
      "num_input_tokens_seen": 75774675,
      "step": 3515,
      "time_per_iteration": 3.2679989337921143
    },
    {
      "auxiliary_loss_clip": 0.01157476,
      "auxiliary_loss_mlp": 0.01029621,
      "balance_loss_clip": 1.05532002,
      "balance_loss_mlp": 1.02132416,
      "epoch": 0.4227740034870438,
      "flos": 16253349753600.0,
      "grad_norm": 1.8927749769020958,
      "language_loss": 0.68916708,
      "learning_rate": 2.5886241377181483e-06,
      "loss": 0.71103811,
      "num_input_tokens_seen": 75793545,
      "step": 3516,
      "time_per_iteration": 2.508251428604126
    },
    {
      "auxiliary_loss_clip": 0.01174479,
      "auxiliary_loss_mlp": 0.01029051,
      "balance_loss_clip": 1.05570459,
      "balance_loss_mlp": 1.0204556,
      "epoch": 0.4228942463776829,
      "flos": 25295691623040.0,
      "grad_norm": 1.7004153494485628,
      "language_loss": 0.81485415,
      "learning_rate": 2.587879622585234e-06,
      "loss": 0.83688939,
      "num_input_tokens_seen": 75812145,
      "step": 3517,
      "time_per_iteration": 3.2553234100341797
    },
    {
      "auxiliary_loss_clip": 0.01169647,
      "auxiliary_loss_mlp": 0.01031274,
      "balance_loss_clip": 1.05565786,
      "balance_loss_mlp": 1.02340662,
      "epoch": 0.423014489268322,
      "flos": 26395779507840.0,
      "grad_norm": 2.354801716907938,
      "language_loss": 0.75580376,
      "learning_rate": 2.5871350182714486e-06,
      "loss": 0.77781296,
      "num_input_tokens_seen": 75833025,
      "step": 3518,
      "time_per_iteration": 2.603123188018799
    },
    {
      "auxiliary_loss_clip": 0.01183822,
      "auxiliary_loss_mlp": 0.01026397,
      "balance_loss_clip": 1.0559907,
      "balance_loss_mlp": 1.01895237,
      "epoch": 0.4231347321589611,
      "flos": 17274002711040.0,
      "grad_norm": 2.167273863329668,
      "language_loss": 0.80323267,
      "learning_rate": 2.586390324889748e-06,
      "loss": 0.82533491,
      "num_input_tokens_seen": 75848925,
      "step": 3519,
      "time_per_iteration": 2.456437110900879
    },
    {
      "auxiliary_loss_clip": 0.01171034,
      "auxiliary_loss_mlp": 0.01031623,
      "balance_loss_clip": 1.05701935,
      "balance_loss_mlp": 1.02421975,
      "epoch": 0.4232549750496002,
      "flos": 22999635475200.0,
      "grad_norm": 1.8381712125618284,
      "language_loss": 0.67579174,
      "learning_rate": 2.5856455425531003e-06,
      "loss": 0.69781828,
      "num_input_tokens_seen": 75870400,
      "step": 3520,
      "time_per_iteration": 2.4923095703125
    },
    {
      "auxiliary_loss_clip": 0.01171643,
      "auxiliary_loss_mlp": 0.0102523,
      "balance_loss_clip": 1.05672765,
      "balance_loss_mlp": 1.01748681,
      "epoch": 0.4233752179402393,
      "flos": 21248343970560.0,
      "grad_norm": 1.7376983949816318,
      "language_loss": 0.80391335,
      "learning_rate": 2.5849006713744902e-06,
      "loss": 0.82588208,
      "num_input_tokens_seen": 75889195,
      "step": 3521,
      "time_per_iteration": 2.4542765617370605
    },
    {
      "auxiliary_loss_clip": 0.01154491,
      "auxiliary_loss_mlp": 0.01028206,
      "balance_loss_clip": 1.0523802,
      "balance_loss_mlp": 1.02002847,
      "epoch": 0.42349546083087836,
      "flos": 20704297599360.0,
      "grad_norm": 2.044537984655198,
      "language_loss": 0.72736812,
      "learning_rate": 2.5841557114669135e-06,
      "loss": 0.7491951,
      "num_input_tokens_seen": 75906055,
      "step": 3522,
      "time_per_iteration": 2.4920802116394043
    },
    {
      "auxiliary_loss_clip": 0.01189986,
      "auxiliary_loss_mlp": 0.01030601,
      "balance_loss_clip": 1.05591679,
      "balance_loss_mlp": 1.02198803,
      "epoch": 0.42361570372151747,
      "flos": 18585065128320.0,
      "grad_norm": 9.759805701092263,
      "language_loss": 0.67134333,
      "learning_rate": 2.58341066294338e-06,
      "loss": 0.69354922,
      "num_input_tokens_seen": 75922720,
      "step": 3523,
      "time_per_iteration": 2.4124362468719482
    },
    {
      "auxiliary_loss_clip": 0.0113476,
      "auxiliary_loss_mlp": 0.00762849,
      "balance_loss_clip": 1.05186999,
      "balance_loss_mlp": 1.00041032,
      "epoch": 0.4237359466121566,
      "flos": 20959478795520.0,
      "grad_norm": 2.3431021267201904,
      "language_loss": 0.85559934,
      "learning_rate": 2.5826655259169124e-06,
      "loss": 0.87457538,
      "num_input_tokens_seen": 75941375,
      "step": 3524,
      "time_per_iteration": 2.602137327194214
    },
    {
      "auxiliary_loss_clip": 0.01188391,
      "auxiliary_loss_mlp": 0.01031083,
      "balance_loss_clip": 1.05849552,
      "balance_loss_mlp": 1.02301896,
      "epoch": 0.42385618950279563,
      "flos": 18038181582720.0,
      "grad_norm": 2.1170359855964445,
      "language_loss": 0.90596938,
      "learning_rate": 2.5819203005005475e-06,
      "loss": 0.92816412,
      "num_input_tokens_seen": 75958710,
      "step": 3525,
      "time_per_iteration": 2.5590906143188477
    },
    {
      "auxiliary_loss_clip": 0.01152699,
      "auxiliary_loss_mlp": 0.01031834,
      "balance_loss_clip": 1.05451465,
      "balance_loss_mlp": 1.02406764,
      "epoch": 0.42397643239343474,
      "flos": 23769129559680.0,
      "grad_norm": 2.016743251961724,
      "language_loss": 0.78911197,
      "learning_rate": 2.581174986807336e-06,
      "loss": 0.81095731,
      "num_input_tokens_seen": 75978945,
      "step": 3526,
      "time_per_iteration": 2.5470805168151855
    },
    {
      "auxiliary_loss_clip": 0.01162197,
      "auxiliary_loss_mlp": 0.00762812,
      "balance_loss_clip": 1.05285668,
      "balance_loss_mlp": 1.00040948,
      "epoch": 0.42409667528407385,
      "flos": 16545088016640.0,
      "grad_norm": 2.2881186211263764,
      "language_loss": 0.9098134,
      "learning_rate": 2.580429584950341e-06,
      "loss": 0.92906344,
      "num_input_tokens_seen": 75994695,
      "step": 3527,
      "time_per_iteration": 2.4381277561187744
    },
    {
      "auxiliary_loss_clip": 0.0114904,
      "auxiliary_loss_mlp": 0.01026268,
      "balance_loss_clip": 1.05291605,
      "balance_loss_mlp": 1.01747012,
      "epoch": 0.4242169181747129,
      "flos": 16034186920320.0,
      "grad_norm": 2.0953409189123198,
      "language_loss": 0.66436046,
      "learning_rate": 2.5796840950426397e-06,
      "loss": 0.68611354,
      "num_input_tokens_seen": 76011780,
      "step": 3528,
      "time_per_iteration": 2.5010526180267334
    },
    {
      "auxiliary_loss_clip": 0.0116075,
      "auxiliary_loss_mlp": 0.01029284,
      "balance_loss_clip": 1.05131042,
      "balance_loss_mlp": 1.02163661,
      "epoch": 0.424337161065352,
      "flos": 20084012611200.0,
      "grad_norm": 1.8718419769513992,
      "language_loss": 0.65617955,
      "learning_rate": 2.578938517197322e-06,
      "loss": 0.67807996,
      "num_input_tokens_seen": 76029875,
      "step": 3529,
      "time_per_iteration": 2.463440179824829
    },
    {
      "auxiliary_loss_clip": 0.01146128,
      "auxiliary_loss_mlp": 0.01027713,
      "balance_loss_clip": 1.05065584,
      "balance_loss_mlp": 1.01976764,
      "epoch": 0.4244574039559911,
      "flos": 23878369797120.0,
      "grad_norm": 2.1711844819422317,
      "language_loss": 0.62226725,
      "learning_rate": 2.5781928515274916e-06,
      "loss": 0.6440056,
      "num_input_tokens_seen": 76048595,
      "step": 3530,
      "time_per_iteration": 2.522073268890381
    },
    {
      "auxiliary_loss_clip": 0.01175823,
      "auxiliary_loss_mlp": 0.01028923,
      "balance_loss_clip": 1.0578289,
      "balance_loss_mlp": 1.02146935,
      "epoch": 0.4245776468466302,
      "flos": 17565920542080.0,
      "grad_norm": 1.91141545402392,
      "language_loss": 0.67646921,
      "learning_rate": 2.577447098146265e-06,
      "loss": 0.69851661,
      "num_input_tokens_seen": 76065770,
      "step": 3531,
      "time_per_iteration": 2.4430432319641113
    },
    {
      "auxiliary_loss_clip": 0.01142812,
      "auxiliary_loss_mlp": 0.0103582,
      "balance_loss_clip": 1.05106425,
      "balance_loss_mlp": 1.02803016,
      "epoch": 0.4246978897372693,
      "flos": 27776256958080.0,
      "grad_norm": 1.749077554336442,
      "language_loss": 0.78845739,
      "learning_rate": 2.5767012571667724e-06,
      "loss": 0.81024373,
      "num_input_tokens_seen": 76085250,
      "step": 3532,
      "time_per_iteration": 2.579741954803467
    },
    {
      "auxiliary_loss_clip": 0.0117314,
      "auxiliary_loss_mlp": 0.01025766,
      "balance_loss_clip": 1.05347848,
      "balance_loss_mlp": 1.01669431,
      "epoch": 0.42481813262790835,
      "flos": 15596615439360.0,
      "grad_norm": 2.0263533953766473,
      "language_loss": 0.67991364,
      "learning_rate": 2.5759553287021587e-06,
      "loss": 0.70190275,
      "num_input_tokens_seen": 76103580,
      "step": 3533,
      "time_per_iteration": 2.4520928859710693
    },
    {
      "auxiliary_loss_clip": 0.01155958,
      "auxiliary_loss_mlp": 0.01027463,
      "balance_loss_clip": 1.05504012,
      "balance_loss_mlp": 1.01883841,
      "epoch": 0.42493837551854746,
      "flos": 23951088881280.0,
      "grad_norm": 2.0860933239739214,
      "language_loss": 0.77319604,
      "learning_rate": 2.5752093128655786e-06,
      "loss": 0.79503024,
      "num_input_tokens_seen": 76121825,
      "step": 3534,
      "time_per_iteration": 2.5165352821350098
    },
    {
      "auxiliary_loss_clip": 0.01148916,
      "auxiliary_loss_mlp": 0.01025351,
      "balance_loss_clip": 1.05022907,
      "balance_loss_mlp": 1.01698267,
      "epoch": 0.4250586184091866,
      "flos": 20813466009600.0,
      "grad_norm": 2.29887617168106,
      "language_loss": 0.73720282,
      "learning_rate": 2.574463209770204e-06,
      "loss": 0.75894547,
      "num_input_tokens_seen": 76141140,
      "step": 3535,
      "time_per_iteration": 3.3060848712921143
    },
    {
      "auxiliary_loss_clip": 0.0114067,
      "auxiliary_loss_mlp": 0.01031257,
      "balance_loss_clip": 1.05043602,
      "balance_loss_mlp": 1.02297163,
      "epoch": 0.42517886129982563,
      "flos": 30371018607360.0,
      "grad_norm": 1.719612265982136,
      "language_loss": 0.79694527,
      "learning_rate": 2.5737170195292165e-06,
      "loss": 0.81866461,
      "num_input_tokens_seen": 76164475,
      "step": 3536,
      "time_per_iteration": 2.6351466178894043
    },
    {
      "auxiliary_loss_clip": 0.01141957,
      "auxiliary_loss_mlp": 0.01028683,
      "balance_loss_clip": 1.04971194,
      "balance_loss_mlp": 1.01990306,
      "epoch": 0.42529910419046474,
      "flos": 20080636732800.0,
      "grad_norm": 2.000824768490275,
      "language_loss": 0.78247178,
      "learning_rate": 2.572970742255814e-06,
      "loss": 0.80417824,
      "num_input_tokens_seen": 76182965,
      "step": 3537,
      "time_per_iteration": 2.5373334884643555
    },
    {
      "auxiliary_loss_clip": 0.01171066,
      "auxiliary_loss_mlp": 0.01028203,
      "balance_loss_clip": 1.05716693,
      "balance_loss_mlp": 1.02056146,
      "epoch": 0.42541934708110385,
      "flos": 22632448694400.0,
      "grad_norm": 1.723061995218477,
      "language_loss": 0.81909847,
      "learning_rate": 2.5722243780632046e-06,
      "loss": 0.84109116,
      "num_input_tokens_seen": 76201230,
      "step": 3538,
      "time_per_iteration": 2.487924098968506
    },
    {
      "auxiliary_loss_clip": 0.01043385,
      "auxiliary_loss_mlp": 0.01004675,
      "balance_loss_clip": 1.02050674,
      "balance_loss_mlp": 1.00367403,
      "epoch": 0.4255395899717429,
      "flos": 66200676186240.0,
      "grad_norm": 0.7732521984192919,
      "language_loss": 0.6047954,
      "learning_rate": 2.5714779270646125e-06,
      "loss": 0.62527603,
      "num_input_tokens_seen": 76262000,
      "step": 3539,
      "time_per_iteration": 3.9092202186584473
    },
    {
      "auxiliary_loss_clip": 0.01162388,
      "auxiliary_loss_mlp": 0.0076273,
      "balance_loss_clip": 1.05590367,
      "balance_loss_mlp": 1.00038922,
      "epoch": 0.425659832862382,
      "flos": 17931814433280.0,
      "grad_norm": 2.6841389965686613,
      "language_loss": 0.77799147,
      "learning_rate": 2.5707313893732735e-06,
      "loss": 0.79724264,
      "num_input_tokens_seen": 76280540,
      "step": 3540,
      "time_per_iteration": 2.483970880508423
    },
    {
      "auxiliary_loss_clip": 0.01091331,
      "auxiliary_loss_mlp": 0.0102913,
      "balance_loss_clip": 1.04120362,
      "balance_loss_mlp": 1.02106535,
      "epoch": 0.4257800757530211,
      "flos": 24022550989440.0,
      "grad_norm": 1.6796403842919871,
      "language_loss": 0.77069861,
      "learning_rate": 2.5699847651024364e-06,
      "loss": 0.79190314,
      "num_input_tokens_seen": 76301180,
      "step": 3541,
      "time_per_iteration": 2.65840482711792
    },
    {
      "auxiliary_loss_clip": 0.01169351,
      "auxiliary_loss_mlp": 0.01028823,
      "balance_loss_clip": 1.0560087,
      "balance_loss_mlp": 1.02093124,
      "epoch": 0.4259003186436602,
      "flos": 23696015425920.0,
      "grad_norm": 3.6313339743253934,
      "language_loss": 0.76857674,
      "learning_rate": 2.5692380543653627e-06,
      "loss": 0.7905584,
      "num_input_tokens_seen": 76319335,
      "step": 3542,
      "time_per_iteration": 3.2987897396087646
    },
    {
      "auxiliary_loss_clip": 0.01175421,
      "auxiliary_loss_mlp": 0.00763123,
      "balance_loss_clip": 1.05565202,
      "balance_loss_mlp": 1.00047863,
      "epoch": 0.4260205615342993,
      "flos": 15259772672640.0,
      "grad_norm": 2.0202996225289915,
      "language_loss": 0.69757605,
      "learning_rate": 2.5684912572753293e-06,
      "loss": 0.7169615,
      "num_input_tokens_seen": 76335010,
      "step": 3543,
      "time_per_iteration": 2.4561078548431396
    },
    {
      "auxiliary_loss_clip": 0.01181584,
      "auxiliary_loss_mlp": 0.01022966,
      "balance_loss_clip": 1.0555141,
      "balance_loss_mlp": 1.01529479,
      "epoch": 0.4261408044249384,
      "flos": 30665306736000.0,
      "grad_norm": 1.7189432327803047,
      "language_loss": 0.84336442,
      "learning_rate": 2.5677443739456245e-06,
      "loss": 0.86540991,
      "num_input_tokens_seen": 76356670,
      "step": 3544,
      "time_per_iteration": 3.206366777420044
    },
    {
      "auxiliary_loss_clip": 0.01158528,
      "auxiliary_loss_mlp": 0.01023892,
      "balance_loss_clip": 1.05613399,
      "balance_loss_mlp": 1.01613712,
      "epoch": 0.42626104731557746,
      "flos": 23257905240960.0,
      "grad_norm": 2.3803050649695447,
      "language_loss": 0.79395318,
      "learning_rate": 2.5669974044895495e-06,
      "loss": 0.8157773,
      "num_input_tokens_seen": 76373065,
      "step": 3545,
      "time_per_iteration": 2.509317636489868
    },
    {
      "auxiliary_loss_clip": 0.01149294,
      "auxiliary_loss_mlp": 0.0102665,
      "balance_loss_clip": 1.051085,
      "balance_loss_mlp": 1.01846027,
      "epoch": 0.42638129020621657,
      "flos": 25884770670720.0,
      "grad_norm": 1.8138901651688484,
      "language_loss": 0.79966331,
      "learning_rate": 2.5662503490204187e-06,
      "loss": 0.82142276,
      "num_input_tokens_seen": 76393230,
      "step": 3546,
      "time_per_iteration": 2.57382869720459
    },
    {
      "auxiliary_loss_clip": 0.0115434,
      "auxiliary_loss_mlp": 0.01022785,
      "balance_loss_clip": 1.0512619,
      "balance_loss_mlp": 1.01500082,
      "epoch": 0.4265015330968556,
      "flos": 26502362138880.0,
      "grad_norm": 2.0335016009604905,
      "language_loss": 0.76303583,
      "learning_rate": 2.5655032076515603e-06,
      "loss": 0.78480709,
      "num_input_tokens_seen": 76412555,
      "step": 3547,
      "time_per_iteration": 2.564664840698242
    },
    {
      "auxiliary_loss_clip": 0.01158405,
      "auxiliary_loss_mlp": 0.01027115,
      "balance_loss_clip": 1.05305004,
      "balance_loss_mlp": 1.01913369,
      "epoch": 0.42662177598749473,
      "flos": 24389522288640.0,
      "grad_norm": 2.8284615523244767,
      "language_loss": 0.82254064,
      "learning_rate": 2.5647559804963155e-06,
      "loss": 0.84439582,
      "num_input_tokens_seen": 76432485,
      "step": 3548,
      "time_per_iteration": 2.5287795066833496
    },
    {
      "auxiliary_loss_clip": 0.01138534,
      "auxiliary_loss_mlp": 0.01032447,
      "balance_loss_clip": 1.05347896,
      "balance_loss_mlp": 1.02463269,
      "epoch": 0.42674201887813384,
      "flos": 23148629089920.0,
      "grad_norm": 1.981451402656454,
      "language_loss": 0.78943217,
      "learning_rate": 2.5640086676680364e-06,
      "loss": 0.81114197,
      "num_input_tokens_seen": 76453980,
      "step": 3549,
      "time_per_iteration": 2.6089608669281006
    },
    {
      "auxiliary_loss_clip": 0.01173815,
      "auxiliary_loss_mlp": 0.01027844,
      "balance_loss_clip": 1.05610466,
      "balance_loss_mlp": 1.0191896,
      "epoch": 0.4268622617687729,
      "flos": 21689614552320.0,
      "grad_norm": 2.294518990482537,
      "language_loss": 0.80787909,
      "learning_rate": 2.5632612692800923e-06,
      "loss": 0.82989573,
      "num_input_tokens_seen": 76473045,
      "step": 3550,
      "time_per_iteration": 2.4722511768341064
    },
    {
      "auxiliary_loss_clip": 0.01143307,
      "auxiliary_loss_mlp": 0.01036092,
      "balance_loss_clip": 1.05048192,
      "balance_loss_mlp": 1.02725244,
      "epoch": 0.426982504659412,
      "flos": 23440151871360.0,
      "grad_norm": 1.9535029950736376,
      "language_loss": 0.75238627,
      "learning_rate": 2.5625137854458603e-06,
      "loss": 0.77418023,
      "num_input_tokens_seen": 76492060,
      "step": 3551,
      "time_per_iteration": 2.5510425567626953
    },
    {
      "auxiliary_loss_clip": 0.01160055,
      "auxiliary_loss_mlp": 0.0102903,
      "balance_loss_clip": 1.05492651,
      "balance_loss_mlp": 1.02156794,
      "epoch": 0.4271027475500511,
      "flos": 18916556768640.0,
      "grad_norm": 1.9293302171028055,
      "language_loss": 0.80180645,
      "learning_rate": 2.561766216278735e-06,
      "loss": 0.82369727,
      "num_input_tokens_seen": 76509655,
      "step": 3552,
      "time_per_iteration": 2.4863080978393555
    },
    {
      "auxiliary_loss_clip": 0.01127058,
      "auxiliary_loss_mlp": 0.01028716,
      "balance_loss_clip": 1.05039573,
      "balance_loss_mlp": 1.02047825,
      "epoch": 0.4272229904406902,
      "flos": 26870554500480.0,
      "grad_norm": 1.7434203483395674,
      "language_loss": 0.80870008,
      "learning_rate": 2.561018561892121e-06,
      "loss": 0.83025783,
      "num_input_tokens_seen": 76528795,
      "step": 3553,
      "time_per_iteration": 2.5821099281311035
    },
    {
      "auxiliary_loss_clip": 0.01153879,
      "auxiliary_loss_mlp": 0.01033323,
      "balance_loss_clip": 1.05030298,
      "balance_loss_mlp": 1.02555037,
      "epoch": 0.4273432333313293,
      "flos": 23951376190080.0,
      "grad_norm": 1.5333451459252483,
      "language_loss": 0.76568747,
      "learning_rate": 2.5602708223994363e-06,
      "loss": 0.78755951,
      "num_input_tokens_seen": 76550660,
      "step": 3554,
      "time_per_iteration": 2.5269036293029785
    },
    {
      "auxiliary_loss_clip": 0.01143208,
      "auxiliary_loss_mlp": 0.01025685,
      "balance_loss_clip": 1.04758322,
      "balance_loss_mlp": 1.01794851,
      "epoch": 0.4274634762219684,
      "flos": 29570354496000.0,
      "grad_norm": 2.1282753049550367,
      "language_loss": 0.67153227,
      "learning_rate": 2.559522997914115e-06,
      "loss": 0.69322121,
      "num_input_tokens_seen": 76570240,
      "step": 3555,
      "time_per_iteration": 2.5892279148101807
    },
    {
      "auxiliary_loss_clip": 0.01184469,
      "auxiliary_loss_mlp": 0.01029943,
      "balance_loss_clip": 1.05815864,
      "balance_loss_mlp": 1.02251947,
      "epoch": 0.42758371911260745,
      "flos": 21434146047360.0,
      "grad_norm": 1.9272640727724966,
      "language_loss": 0.84807754,
      "learning_rate": 2.558775088549599e-06,
      "loss": 0.87022161,
      "num_input_tokens_seen": 76589820,
      "step": 3556,
      "time_per_iteration": 2.4482078552246094
    },
    {
      "auxiliary_loss_clip": 0.0117736,
      "auxiliary_loss_mlp": 0.01028798,
      "balance_loss_clip": 1.05546021,
      "balance_loss_mlp": 1.02057838,
      "epoch": 0.42770396200324656,
      "flos": 14752822072320.0,
      "grad_norm": 4.6819192163826555,
      "language_loss": 0.66522568,
      "learning_rate": 2.5580270944193467e-06,
      "loss": 0.68728727,
      "num_input_tokens_seen": 76606640,
      "step": 3557,
      "time_per_iteration": 2.419623851776123
    },
    {
      "auxiliary_loss_clip": 0.01084349,
      "auxiliary_loss_mlp": 0.01003471,
      "balance_loss_clip": 1.02027535,
      "balance_loss_mlp": 1.0024159,
      "epoch": 0.4278242048938857,
      "flos": 70654712601600.0,
      "grad_norm": 0.7595784660188609,
      "language_loss": 0.55505395,
      "learning_rate": 2.557279015636827e-06,
      "loss": 0.57593215,
      "num_input_tokens_seen": 76667050,
      "step": 3558,
      "time_per_iteration": 3.0204498767852783
    },
    {
      "auxiliary_loss_clip": 0.01070572,
      "auxiliary_loss_mlp": 0.01003582,
      "balance_loss_clip": 1.0200609,
      "balance_loss_mlp": 1.00245595,
      "epoch": 0.42794444778452473,
      "flos": 69366165033600.0,
      "grad_norm": 0.7784422243866626,
      "language_loss": 0.61264527,
      "learning_rate": 2.5565308523155245e-06,
      "loss": 0.63338685,
      "num_input_tokens_seen": 76726650,
      "step": 3559,
      "time_per_iteration": 2.9806787967681885
    },
    {
      "auxiliary_loss_clip": 0.01123368,
      "auxiliary_loss_mlp": 0.01029542,
      "balance_loss_clip": 1.04993415,
      "balance_loss_mlp": 1.02137637,
      "epoch": 0.42806469067516384,
      "flos": 18215328481920.0,
      "grad_norm": 2.2434340829085153,
      "language_loss": 0.81738192,
      "learning_rate": 2.5557826045689336e-06,
      "loss": 0.838911,
      "num_input_tokens_seen": 76742890,
      "step": 3560,
      "time_per_iteration": 2.5365028381347656
    },
    {
      "auxiliary_loss_clip": 0.01050918,
      "auxiliary_loss_mlp": 0.01005069,
      "balance_loss_clip": 1.02442741,
      "balance_loss_mlp": 1.00351954,
      "epoch": 0.4281849335658029,
      "flos": 54535814432640.0,
      "grad_norm": 0.818732295499881,
      "language_loss": 0.58867776,
      "learning_rate": 2.5550342725105643e-06,
      "loss": 0.60923767,
      "num_input_tokens_seen": 76801055,
      "step": 3561,
      "time_per_iteration": 3.0466389656066895
    },
    {
      "auxiliary_loss_clip": 0.0117374,
      "auxiliary_loss_mlp": 0.01034155,
      "balance_loss_clip": 1.0586741,
      "balance_loss_mlp": 1.02611387,
      "epoch": 0.428305176456442,
      "flos": 17274828723840.0,
      "grad_norm": 1.9251193512602596,
      "language_loss": 0.81000805,
      "learning_rate": 2.554285856253937e-06,
      "loss": 0.83208698,
      "num_input_tokens_seen": 76819890,
      "step": 3562,
      "time_per_iteration": 3.329968214035034
    },
    {
      "auxiliary_loss_clip": 0.01154903,
      "auxiliary_loss_mlp": 0.01030535,
      "balance_loss_clip": 1.05482984,
      "balance_loss_mlp": 1.02243209,
      "epoch": 0.4284254193470811,
      "flos": 26359509749760.0,
      "grad_norm": 2.103650819117505,
      "language_loss": 0.77569038,
      "learning_rate": 2.5535373559125855e-06,
      "loss": 0.79754472,
      "num_input_tokens_seen": 76840255,
      "step": 3563,
      "time_per_iteration": 2.5613696575164795
    },
    {
      "auxiliary_loss_clip": 0.01102388,
      "auxiliary_loss_mlp": 0.01024007,
      "balance_loss_clip": 1.04678929,
      "balance_loss_mlp": 1.01523948,
      "epoch": 0.42854566223772017,
      "flos": 29714248379520.0,
      "grad_norm": 1.80926251509629,
      "language_loss": 0.81813461,
      "learning_rate": 2.552788771600057e-06,
      "loss": 0.83939862,
      "num_input_tokens_seen": 76860565,
      "step": 3564,
      "time_per_iteration": 2.8759994506835938
    },
    {
      "auxiliary_loss_clip": 0.01146431,
      "auxiliary_loss_mlp": 0.01032696,
      "balance_loss_clip": 1.05467808,
      "balance_loss_mlp": 1.02420259,
      "epoch": 0.4286659051283593,
      "flos": 22018161277440.0,
      "grad_norm": 2.0687220530542243,
      "language_loss": 0.82019293,
      "learning_rate": 2.5520401034299118e-06,
      "loss": 0.84198427,
      "num_input_tokens_seen": 76878325,
      "step": 3565,
      "time_per_iteration": 2.961338758468628
    },
    {
      "auxiliary_loss_clip": 0.01175253,
      "auxiliary_loss_mlp": 0.01032268,
      "balance_loss_clip": 1.05667877,
      "balance_loss_mlp": 1.02370882,
      "epoch": 0.4287861480189984,
      "flos": 13334422838400.0,
      "grad_norm": 2.2319618979316096,
      "language_loss": 0.87335134,
      "learning_rate": 2.551291351515722e-06,
      "loss": 0.89542651,
      "num_input_tokens_seen": 76895340,
      "step": 3566,
      "time_per_iteration": 3.603695869445801
    },
    {
      "auxiliary_loss_clip": 0.01136179,
      "auxiliary_loss_mlp": 0.00763665,
      "balance_loss_clip": 1.04632497,
      "balance_loss_mlp": 1.00047851,
      "epoch": 0.42890639090963745,
      "flos": 26651535321600.0,
      "grad_norm": 1.6067015940367897,
      "language_loss": 0.85390478,
      "learning_rate": 2.5505425159710726e-06,
      "loss": 0.87290323,
      "num_input_tokens_seen": 76915150,
      "step": 3567,
      "time_per_iteration": 2.674638509750366
    },
    {
      "auxiliary_loss_clip": 0.01163183,
      "auxiliary_loss_mlp": 0.00763388,
      "balance_loss_clip": 1.05157232,
      "balance_loss_mlp": 1.00046396,
      "epoch": 0.42902663380027656,
      "flos": 24055768091520.0,
      "grad_norm": 3.9916115142268445,
      "language_loss": 0.83194745,
      "learning_rate": 2.549793596909561e-06,
      "loss": 0.85121316,
      "num_input_tokens_seen": 76933770,
      "step": 3568,
      "time_per_iteration": 2.529055118560791
    },
    {
      "auxiliary_loss_clip": 0.01154539,
      "auxiliary_loss_mlp": 0.01029289,
      "balance_loss_clip": 1.05505311,
      "balance_loss_mlp": 1.02096212,
      "epoch": 0.42914687669091567,
      "flos": 15632561975040.0,
      "grad_norm": 1.9891471433547236,
      "language_loss": 0.65879661,
      "learning_rate": 2.5490445944447976e-06,
      "loss": 0.68063486,
      "num_input_tokens_seen": 76952265,
      "step": 3569,
      "time_per_iteration": 3.33054518699646
    },
    {
      "auxiliary_loss_clip": 0.01169898,
      "auxiliary_loss_mlp": 0.01027862,
      "balance_loss_clip": 1.05364561,
      "balance_loss_mlp": 1.01987469,
      "epoch": 0.4292671195815547,
      "flos": 31467802440960.0,
      "grad_norm": 2.4909204966518628,
      "language_loss": 0.65166163,
      "learning_rate": 2.548295508690406e-06,
      "loss": 0.67363918,
      "num_input_tokens_seen": 76973560,
      "step": 3570,
      "time_per_iteration": 3.2226977348327637
    },
    {
      "auxiliary_loss_clip": 0.01173847,
      "auxiliary_loss_mlp": 0.01027592,
      "balance_loss_clip": 1.05355704,
      "balance_loss_mlp": 1.01922894,
      "epoch": 0.42938736247219383,
      "flos": 30257756046720.0,
      "grad_norm": 1.6986809529245595,
      "language_loss": 0.76484245,
      "learning_rate": 2.5475463397600217e-06,
      "loss": 0.78685689,
      "num_input_tokens_seen": 76993640,
      "step": 3571,
      "time_per_iteration": 2.5285048484802246
    },
    {
      "auxiliary_loss_clip": 0.01191717,
      "auxiliary_loss_mlp": 0.01029933,
      "balance_loss_clip": 1.05914819,
      "balance_loss_mlp": 1.02160597,
      "epoch": 0.42950760536283294,
      "flos": 29349683291520.0,
      "grad_norm": 2.443079276247546,
      "language_loss": 0.77560741,
      "learning_rate": 2.546797087767293e-06,
      "loss": 0.79782391,
      "num_input_tokens_seen": 77013765,
      "step": 3572,
      "time_per_iteration": 2.487550735473633
    },
    {
      "auxiliary_loss_clip": 0.01124443,
      "auxiliary_loss_mlp": 0.01034076,
      "balance_loss_clip": 1.04902542,
      "balance_loss_mlp": 1.02599955,
      "epoch": 0.429627848253472,
      "flos": 26869943969280.0,
      "grad_norm": 1.8032635849306458,
      "language_loss": 0.87161362,
      "learning_rate": 2.546047752825881e-06,
      "loss": 0.89319885,
      "num_input_tokens_seen": 77034370,
      "step": 3573,
      "time_per_iteration": 2.6012799739837646
    },
    {
      "auxiliary_loss_clip": 0.01132383,
      "auxiliary_loss_mlp": 0.01031203,
      "balance_loss_clip": 1.05017209,
      "balance_loss_mlp": 1.02296543,
      "epoch": 0.4297480911441111,
      "flos": 13881270470400.0,
      "grad_norm": 2.242408211206307,
      "language_loss": 0.93161988,
      "learning_rate": 2.5452983350494595e-06,
      "loss": 0.95325571,
      "num_input_tokens_seen": 77049925,
      "step": 3574,
      "time_per_iteration": 2.532256603240967
    },
    {
      "auxiliary_loss_clip": 0.01170554,
      "auxiliary_loss_mlp": 0.00763107,
      "balance_loss_clip": 1.05412567,
      "balance_loss_mlp": 1.00056624,
      "epoch": 0.4298683340347502,
      "flos": 20741141975040.0,
      "grad_norm": 2.1450186344560365,
      "language_loss": 0.65653044,
      "learning_rate": 2.544548834551713e-06,
      "loss": 0.67586708,
      "num_input_tokens_seen": 77068930,
      "step": 3575,
      "time_per_iteration": 2.500807762145996
    },
    {
      "auxiliary_loss_clip": 0.01139115,
      "auxiliary_loss_mlp": 0.00763196,
      "balance_loss_clip": 1.05166554,
      "balance_loss_mlp": 1.00045967,
      "epoch": 0.4299885769253893,
      "flos": 20882126856960.0,
      "grad_norm": 2.4004906846527336,
      "language_loss": 0.94127178,
      "learning_rate": 2.5437992514463424e-06,
      "loss": 0.9602949,
      "num_input_tokens_seen": 77082255,
      "step": 3576,
      "time_per_iteration": 2.515387773513794
    },
    {
      "auxiliary_loss_clip": 0.01174261,
      "auxiliary_loss_mlp": 0.01028046,
      "balance_loss_clip": 1.05637813,
      "balance_loss_mlp": 1.01946259,
      "epoch": 0.4301088198160284,
      "flos": 25484618183040.0,
      "grad_norm": 1.7281665072426735,
      "language_loss": 0.87893593,
      "learning_rate": 2.5430495858470565e-06,
      "loss": 0.90095901,
      "num_input_tokens_seen": 77101725,
      "step": 3577,
      "time_per_iteration": 2.4934825897216797
    },
    {
      "auxiliary_loss_clip": 0.01169641,
      "auxiliary_loss_mlp": 0.01032265,
      "balance_loss_clip": 1.05549073,
      "balance_loss_mlp": 1.0241704,
      "epoch": 0.43022906270666744,
      "flos": 18259427404800.0,
      "grad_norm": 2.78896037698433,
      "language_loss": 0.77113187,
      "learning_rate": 2.54229983786758e-06,
      "loss": 0.7931509,
      "num_input_tokens_seen": 77119670,
      "step": 3578,
      "time_per_iteration": 2.425658702850342
    },
    {
      "auxiliary_loss_clip": 0.01156303,
      "auxiliary_loss_mlp": 0.01029303,
      "balance_loss_clip": 1.05145121,
      "balance_loss_mlp": 1.02094698,
      "epoch": 0.43034930559730655,
      "flos": 23399536567680.0,
      "grad_norm": 1.8501987499236685,
      "language_loss": 0.85085428,
      "learning_rate": 2.541550007621651e-06,
      "loss": 0.87271035,
      "num_input_tokens_seen": 77138160,
      "step": 3579,
      "time_per_iteration": 2.4952118396759033
    },
    {
      "auxiliary_loss_clip": 0.01171266,
      "auxiliary_loss_mlp": 0.01028783,
      "balance_loss_clip": 1.05745423,
      "balance_loss_mlp": 1.02094483,
      "epoch": 0.43046954848794566,
      "flos": 28184382264960.0,
      "grad_norm": 2.139920714655677,
      "language_loss": 0.79701084,
      "learning_rate": 2.5408000952230156e-06,
      "loss": 0.81901133,
      "num_input_tokens_seen": 77156950,
      "step": 3580,
      "time_per_iteration": 2.5249781608581543
    },
    {
      "auxiliary_loss_clip": 0.01152759,
      "auxiliary_loss_mlp": 0.01027642,
      "balance_loss_clip": 1.05095124,
      "balance_loss_mlp": 1.01902294,
      "epoch": 0.4305897913785847,
      "flos": 28580476515840.0,
      "grad_norm": 2.0773310101012012,
      "language_loss": 0.90524906,
      "learning_rate": 2.5400501007854357e-06,
      "loss": 0.92705309,
      "num_input_tokens_seen": 77176395,
      "step": 3581,
      "time_per_iteration": 2.5997626781463623
    },
    {
      "auxiliary_loss_clip": 0.01126468,
      "auxiliary_loss_mlp": 0.01033263,
      "balance_loss_clip": 1.04613614,
      "balance_loss_mlp": 1.02560103,
      "epoch": 0.43071003426922383,
      "flos": 20448721353600.0,
      "grad_norm": 1.761173417858187,
      "language_loss": 0.75274009,
      "learning_rate": 2.539300024422685e-06,
      "loss": 0.77433741,
      "num_input_tokens_seen": 77194340,
      "step": 3582,
      "time_per_iteration": 2.556225299835205
    },
    {
      "auxiliary_loss_clip": 0.01050716,
      "auxiliary_loss_mlp": 0.01003189,
      "balance_loss_clip": 1.01982164,
      "balance_loss_mlp": 1.00197935,
      "epoch": 0.43083027715986294,
      "flos": 51997969883520.0,
      "grad_norm": 0.7906574745758691,
      "language_loss": 0.60928702,
      "learning_rate": 2.538549866248549e-06,
      "loss": 0.62982607,
      "num_input_tokens_seen": 77249320,
      "step": 3583,
      "time_per_iteration": 2.9549360275268555
    },
    {
      "auxiliary_loss_clip": 0.01173343,
      "auxiliary_loss_mlp": 0.0102745,
      "balance_loss_clip": 1.05350137,
      "balance_loss_mlp": 1.01874781,
      "epoch": 0.430950520050502,
      "flos": 16690885320960.0,
      "grad_norm": 2.3415440895236714,
      "language_loss": 0.80943739,
      "learning_rate": 2.5377996263768274e-06,
      "loss": 0.83144534,
      "num_input_tokens_seen": 77267400,
      "step": 3584,
      "time_per_iteration": 2.446281671524048
    },
    {
      "auxiliary_loss_clip": 0.01156413,
      "auxiliary_loss_mlp": 0.01033755,
      "balance_loss_clip": 1.04978085,
      "balance_loss_mlp": 1.02570295,
      "epoch": 0.4310707629411411,
      "flos": 24608433726720.0,
      "grad_norm": 1.7468793875230286,
      "language_loss": 0.68138158,
      "learning_rate": 2.5370493049213293e-06,
      "loss": 0.70328331,
      "num_input_tokens_seen": 77287045,
      "step": 3585,
      "time_per_iteration": 2.524519205093384
    },
    {
      "auxiliary_loss_clip": 0.010918,
      "auxiliary_loss_mlp": 0.01030482,
      "balance_loss_clip": 1.04665244,
      "balance_loss_mlp": 1.02145815,
      "epoch": 0.4311910058317802,
      "flos": 26432983019520.0,
      "grad_norm": 2.0215579887339024,
      "language_loss": 0.80402058,
      "learning_rate": 2.536298901995878e-06,
      "loss": 0.82524341,
      "num_input_tokens_seen": 77306255,
      "step": 3586,
      "time_per_iteration": 2.703721046447754
    },
    {
      "auxiliary_loss_clip": 0.01160498,
      "auxiliary_loss_mlp": 0.01027067,
      "balance_loss_clip": 1.05462074,
      "balance_loss_mlp": 1.0187639,
      "epoch": 0.43131124872241927,
      "flos": 25155891889920.0,
      "grad_norm": 1.614446675531778,
      "language_loss": 0.80473924,
      "learning_rate": 2.535548417714311e-06,
      "loss": 0.82661486,
      "num_input_tokens_seen": 77325555,
      "step": 3587,
      "time_per_iteration": 2.740952730178833
    },
    {
      "auxiliary_loss_clip": 0.01178517,
      "auxiliary_loss_mlp": 0.01029529,
      "balance_loss_clip": 1.05543554,
      "balance_loss_mlp": 1.02067208,
      "epoch": 0.4314314916130584,
      "flos": 21614812479360.0,
      "grad_norm": 1.533642986622583,
      "language_loss": 0.86980295,
      "learning_rate": 2.534797852190474e-06,
      "loss": 0.89188343,
      "num_input_tokens_seen": 77345735,
      "step": 3588,
      "time_per_iteration": 3.4060065746307373
    },
    {
      "auxiliary_loss_clip": 0.01169581,
      "auxiliary_loss_mlp": 0.01037813,
      "balance_loss_clip": 1.05319691,
      "balance_loss_mlp": 1.02904487,
      "epoch": 0.4315517345036975,
      "flos": 19275016544640.0,
      "grad_norm": 2.009137048748634,
      "language_loss": 0.81502658,
      "learning_rate": 2.5340472055382283e-06,
      "loss": 0.83710051,
      "num_input_tokens_seen": 77361765,
      "step": 3589,
      "time_per_iteration": 2.4752395153045654
    },
    {
      "auxiliary_loss_clip": 0.01139034,
      "auxiliary_loss_mlp": 0.01026132,
      "balance_loss_clip": 1.04677331,
      "balance_loss_mlp": 1.01827312,
      "epoch": 0.43167197739433655,
      "flos": 24273853516800.0,
      "grad_norm": 2.8294804541610383,
      "language_loss": 0.81204522,
      "learning_rate": 2.5332964778714468e-06,
      "loss": 0.83369684,
      "num_input_tokens_seen": 77378950,
      "step": 3590,
      "time_per_iteration": 2.5702152252197266
    },
    {
      "auxiliary_loss_clip": 0.0114217,
      "auxiliary_loss_mlp": 0.01026267,
      "balance_loss_clip": 1.05348587,
      "balance_loss_mlp": 1.0186739,
      "epoch": 0.43179222028497566,
      "flos": 16867816738560.0,
      "grad_norm": 1.5957164668142962,
      "language_loss": 0.66312826,
      "learning_rate": 2.5325456693040123e-06,
      "loss": 0.68481266,
      "num_input_tokens_seen": 77396145,
      "step": 3591,
      "time_per_iteration": 2.5571916103363037
    },
    {
      "auxiliary_loss_clip": 0.01179218,
      "auxiliary_loss_mlp": 0.01027707,
      "balance_loss_clip": 1.05383968,
      "balance_loss_mlp": 1.01914167,
      "epoch": 0.43191246317561477,
      "flos": 17639214243840.0,
      "grad_norm": 2.2300435099828704,
      "language_loss": 0.74680722,
      "learning_rate": 2.531794779949824e-06,
      "loss": 0.76887643,
      "num_input_tokens_seen": 77414045,
      "step": 3592,
      "time_per_iteration": 3.300826072692871
    },
    {
      "auxiliary_loss_clip": 0.01134499,
      "auxiliary_loss_mlp": 0.01027657,
      "balance_loss_clip": 1.04850733,
      "balance_loss_mlp": 1.01959825,
      "epoch": 0.4320327060662538,
      "flos": 23878800760320.0,
      "grad_norm": 1.6716766956956526,
      "language_loss": 0.87860763,
      "learning_rate": 2.5310438099227903e-06,
      "loss": 0.90022922,
      "num_input_tokens_seen": 77431310,
      "step": 3593,
      "time_per_iteration": 2.5510101318359375
    },
    {
      "auxiliary_loss_clip": 0.01072462,
      "auxiliary_loss_mlp": 0.01001891,
      "balance_loss_clip": 1.01909506,
      "balance_loss_mlp": 1.00081182,
      "epoch": 0.43215294895689293,
      "flos": 66394917959040.0,
      "grad_norm": 0.9191302928058255,
      "language_loss": 0.53381926,
      "learning_rate": 2.530292759336833e-06,
      "loss": 0.55456281,
      "num_input_tokens_seen": 77492045,
      "step": 3594,
      "time_per_iteration": 3.1054184436798096
    },
    {
      "auxiliary_loss_clip": 0.01157345,
      "auxiliary_loss_mlp": 0.01028151,
      "balance_loss_clip": 1.05487216,
      "balance_loss_mlp": 1.01938283,
      "epoch": 0.432273191847532,
      "flos": 20594267262720.0,
      "grad_norm": 2.3427555431985314,
      "language_loss": 0.69422531,
      "learning_rate": 2.5295416283058855e-06,
      "loss": 0.71608031,
      "num_input_tokens_seen": 77510910,
      "step": 3595,
      "time_per_iteration": 2.497631311416626
    },
    {
      "auxiliary_loss_clip": 0.01153369,
      "auxiliary_loss_mlp": 0.00762782,
      "balance_loss_clip": 1.05268729,
      "balance_loss_mlp": 1.00050402,
      "epoch": 0.4323934347381711,
      "flos": 19282127437440.0,
      "grad_norm": 1.5675448187427266,
      "language_loss": 0.65933645,
      "learning_rate": 2.5287904169438943e-06,
      "loss": 0.67849797,
      "num_input_tokens_seen": 77530115,
      "step": 3596,
      "time_per_iteration": 4.03898024559021
    },
    {
      "auxiliary_loss_clip": 0.01110511,
      "auxiliary_loss_mlp": 0.01038306,
      "balance_loss_clip": 1.05006278,
      "balance_loss_mlp": 1.02889407,
      "epoch": 0.4325136776288102,
      "flos": 21726315273600.0,
      "grad_norm": 3.051767216861184,
      "language_loss": 0.64282525,
      "learning_rate": 2.528039125364817e-06,
      "loss": 0.66431338,
      "num_input_tokens_seen": 77548920,
      "step": 3597,
      "time_per_iteration": 2.6449201107025146
    },
    {
      "auxiliary_loss_clip": 0.01145312,
      "auxiliary_loss_mlp": 0.01030879,
      "balance_loss_clip": 1.05067301,
      "balance_loss_mlp": 1.02229559,
      "epoch": 0.43263392051944927,
      "flos": 22340746344960.0,
      "grad_norm": 2.246116252924554,
      "language_loss": 0.75853992,
      "learning_rate": 2.5272877536826246e-06,
      "loss": 0.78030181,
      "num_input_tokens_seen": 77567715,
      "step": 3598,
      "time_per_iteration": 2.5189919471740723
    },
    {
      "auxiliary_loss_clip": 0.0113151,
      "auxiliary_loss_mlp": 0.01030876,
      "balance_loss_clip": 1.04659212,
      "balance_loss_mlp": 1.02204812,
      "epoch": 0.4327541634100884,
      "flos": 29168406328320.0,
      "grad_norm": 2.3783946157873723,
      "language_loss": 0.70116448,
      "learning_rate": 2.5265363020112986e-06,
      "loss": 0.72278833,
      "num_input_tokens_seen": 77588035,
      "step": 3599,
      "time_per_iteration": 2.6355364322662354
    },
    {
      "auxiliary_loss_clip": 0.01172324,
      "auxiliary_loss_mlp": 0.01034719,
      "balance_loss_clip": 1.05613291,
      "balance_loss_mlp": 1.02587986,
      "epoch": 0.4328744063007275,
      "flos": 26067448264320.0,
      "grad_norm": 3.323915847623867,
      "language_loss": 0.83648968,
      "learning_rate": 2.5257847704648344e-06,
      "loss": 0.85856009,
      "num_input_tokens_seen": 77609265,
      "step": 3600,
      "time_per_iteration": 2.5086514949798584
    },
    {
      "auxiliary_loss_clip": 0.01184066,
      "auxiliary_loss_mlp": 0.01028052,
      "balance_loss_clip": 1.05506432,
      "balance_loss_mlp": 1.02017796,
      "epoch": 0.43299464919136654,
      "flos": 16581357774720.0,
      "grad_norm": 1.9252807626380084,
      "language_loss": 0.75428838,
      "learning_rate": 2.525033159157239e-06,
      "loss": 0.77640957,
      "num_input_tokens_seen": 77625580,
      "step": 3601,
      "time_per_iteration": 2.3959898948669434
    },
    {
      "auxiliary_loss_clip": 0.01169547,
      "auxiliary_loss_mlp": 0.01039193,
      "balance_loss_clip": 1.05401242,
      "balance_loss_mlp": 1.02986491,
      "epoch": 0.43311489208200565,
      "flos": 16107265140480.0,
      "grad_norm": 2.006134952642327,
      "language_loss": 0.77326339,
      "learning_rate": 2.52428146820253e-06,
      "loss": 0.79535079,
      "num_input_tokens_seen": 77643835,
      "step": 3602,
      "time_per_iteration": 2.4670159816741943
    },
    {
      "auxiliary_loss_clip": 0.01146354,
      "auxiliary_loss_mlp": 0.01031665,
      "balance_loss_clip": 1.05238581,
      "balance_loss_mlp": 1.02189553,
      "epoch": 0.43323513497264476,
      "flos": 22930220442240.0,
      "grad_norm": 1.7307430856872235,
      "language_loss": 0.81849921,
      "learning_rate": 2.52352969771474e-06,
      "loss": 0.84027946,
      "num_input_tokens_seen": 77663060,
      "step": 3603,
      "time_per_iteration": 2.5484559535980225
    },
    {
      "auxiliary_loss_clip": 0.01159581,
      "auxiliary_loss_mlp": 0.01029624,
      "balance_loss_clip": 1.05276,
      "balance_loss_mlp": 1.02141619,
      "epoch": 0.4333553778632838,
      "flos": 25299031587840.0,
      "grad_norm": 2.356006697098909,
      "language_loss": 0.88193458,
      "learning_rate": 2.5227778478079106e-06,
      "loss": 0.90382659,
      "num_input_tokens_seen": 77682470,
      "step": 3604,
      "time_per_iteration": 2.5340633392333984
    },
    {
      "auxiliary_loss_clip": 0.01167145,
      "auxiliary_loss_mlp": 0.01033376,
      "balance_loss_clip": 1.05174792,
      "balance_loss_mlp": 1.02540648,
      "epoch": 0.43347562075392293,
      "flos": 19387165783680.0,
      "grad_norm": 1.5783330184256474,
      "language_loss": 0.76814699,
      "learning_rate": 2.522025918596098e-06,
      "loss": 0.79015219,
      "num_input_tokens_seen": 77700770,
      "step": 3605,
      "time_per_iteration": 2.449148654937744
    },
    {
      "auxiliary_loss_clip": 0.01173961,
      "auxiliary_loss_mlp": 0.0102963,
      "balance_loss_clip": 1.05479336,
      "balance_loss_mlp": 1.02148247,
      "epoch": 0.43359586364456204,
      "flos": 26325969425280.0,
      "grad_norm": 2.181342040543769,
      "language_loss": 0.65379238,
      "learning_rate": 2.521273910193368e-06,
      "loss": 0.67582834,
      "num_input_tokens_seen": 77723950,
      "step": 3606,
      "time_per_iteration": 2.523129940032959
    },
    {
      "auxiliary_loss_clip": 0.01178969,
      "auxiliary_loss_mlp": 0.01027876,
      "balance_loss_clip": 1.05606961,
      "balance_loss_mlp": 1.01939464,
      "epoch": 0.4337161065352011,
      "flos": 15989261984640.0,
      "grad_norm": 2.609366644306554,
      "language_loss": 0.87145531,
      "learning_rate": 2.5205218227138006e-06,
      "loss": 0.89352375,
      "num_input_tokens_seen": 77736905,
      "step": 3607,
      "time_per_iteration": 2.421447277069092
    },
    {
      "auxiliary_loss_clip": 0.0118613,
      "auxiliary_loss_mlp": 0.01027583,
      "balance_loss_clip": 1.05574131,
      "balance_loss_mlp": 1.01929164,
      "epoch": 0.4338363494258402,
      "flos": 20224710184320.0,
      "grad_norm": 2.258393103108662,
      "language_loss": 0.79083574,
      "learning_rate": 2.519769656271486e-06,
      "loss": 0.81297284,
      "num_input_tokens_seen": 77754325,
      "step": 3608,
      "time_per_iteration": 2.4304187297821045
    },
    {
      "auxiliary_loss_clip": 0.01120083,
      "auxiliary_loss_mlp": 0.01031254,
      "balance_loss_clip": 1.0486685,
      "balance_loss_mlp": 1.02252769,
      "epoch": 0.43395659231647926,
      "flos": 20083904870400.0,
      "grad_norm": 2.317198853509111,
      "language_loss": 0.67356873,
      "learning_rate": 2.5190174109805285e-06,
      "loss": 0.69508207,
      "num_input_tokens_seen": 77774150,
      "step": 3609,
      "time_per_iteration": 2.546675443649292
    },
    {
      "auxiliary_loss_clip": 0.01147647,
      "auxiliary_loss_mlp": 0.01026939,
      "balance_loss_clip": 1.04997301,
      "balance_loss_mlp": 1.01796293,
      "epoch": 0.43407683520711837,
      "flos": 19901801894400.0,
      "grad_norm": 2.10948214248333,
      "language_loss": 0.64168036,
      "learning_rate": 2.518265086955042e-06,
      "loss": 0.66342616,
      "num_input_tokens_seen": 77791870,
      "step": 3610,
      "time_per_iteration": 2.5092434883117676
    },
    {
      "auxiliary_loss_clip": 0.01185956,
      "auxiliary_loss_mlp": 0.01037496,
      "balance_loss_clip": 1.05473149,
      "balance_loss_mlp": 1.02936292,
      "epoch": 0.4341970780977575,
      "flos": 23108732058240.0,
      "grad_norm": 1.6843623205361662,
      "language_loss": 0.83678395,
      "learning_rate": 2.5175126843091534e-06,
      "loss": 0.85901845,
      "num_input_tokens_seen": 77811240,
      "step": 3611,
      "time_per_iteration": 2.4476075172424316
    },
    {
      "auxiliary_loss_clip": 0.01158178,
      "auxiliary_loss_mlp": 0.01023769,
      "balance_loss_clip": 1.05048442,
      "balance_loss_mlp": 1.01560867,
      "epoch": 0.43431732098839654,
      "flos": 37408288406400.0,
      "grad_norm": 2.1042005140210893,
      "language_loss": 0.75145274,
      "learning_rate": 2.5167602031570034e-06,
      "loss": 0.77327222,
      "num_input_tokens_seen": 77831425,
      "step": 3612,
      "time_per_iteration": 2.617098093032837
    },
    {
      "auxiliary_loss_clip": 0.0118606,
      "auxiliary_loss_mlp": 0.01025393,
      "balance_loss_clip": 1.05638385,
      "balance_loss_mlp": 1.01708436,
      "epoch": 0.43443756387903565,
      "flos": 31868206323840.0,
      "grad_norm": 1.6016382545762038,
      "language_loss": 0.73373705,
      "learning_rate": 2.51600764361274e-06,
      "loss": 0.75585163,
      "num_input_tokens_seen": 77852950,
      "step": 3613,
      "time_per_iteration": 2.501828908920288
    },
    {
      "auxiliary_loss_clip": 0.01188277,
      "auxiliary_loss_mlp": 0.01026601,
      "balance_loss_clip": 1.05649614,
      "balance_loss_mlp": 1.01847661,
      "epoch": 0.43455780676967476,
      "flos": 23477139901440.0,
      "grad_norm": 2.1886245207445425,
      "language_loss": 0.78908598,
      "learning_rate": 2.5152550057905283e-06,
      "loss": 0.81123483,
      "num_input_tokens_seen": 77872840,
      "step": 3614,
      "time_per_iteration": 2.4389216899871826
    },
    {
      "auxiliary_loss_clip": 0.01174111,
      "auxiliary_loss_mlp": 0.0076366,
      "balance_loss_clip": 1.05603015,
      "balance_loss_mlp": 1.00063324,
      "epoch": 0.4346780496603138,
      "flos": 24207060176640.0,
      "grad_norm": 2.2492235974098054,
      "language_loss": 0.76705825,
      "learning_rate": 2.5145022898045415e-06,
      "loss": 0.78643596,
      "num_input_tokens_seen": 77892025,
      "step": 3615,
      "time_per_iteration": 3.308350086212158
    },
    {
      "auxiliary_loss_clip": 0.01159266,
      "auxiliary_loss_mlp": 0.0103332,
      "balance_loss_clip": 1.04948032,
      "balance_loss_mlp": 1.02433205,
      "epoch": 0.4347982925509529,
      "flos": 17092366611840.0,
      "grad_norm": 2.026332633776362,
      "language_loss": 0.89337057,
      "learning_rate": 2.5137494957689664e-06,
      "loss": 0.91529644,
      "num_input_tokens_seen": 77907635,
      "step": 3616,
      "time_per_iteration": 2.4904677867889404
    },
    {
      "auxiliary_loss_clip": 0.01061576,
      "auxiliary_loss_mlp": 0.01005692,
      "balance_loss_clip": 1.01835418,
      "balance_loss_mlp": 1.00462496,
      "epoch": 0.43491853544159204,
      "flos": 60945544696320.0,
      "grad_norm": 0.7613973989573046,
      "language_loss": 0.5735485,
      "learning_rate": 2.5129966237980016e-06,
      "loss": 0.59422117,
      "num_input_tokens_seen": 77970630,
      "step": 3617,
      "time_per_iteration": 3.0938665866851807
    },
    {
      "auxiliary_loss_clip": 0.01142707,
      "auxiliary_loss_mlp": 0.01025509,
      "balance_loss_clip": 1.04907811,
      "balance_loss_mlp": 1.017272,
      "epoch": 0.4350387783322311,
      "flos": 21944652094080.0,
      "grad_norm": 1.8988439242893407,
      "language_loss": 0.77993286,
      "learning_rate": 2.512243674005857e-06,
      "loss": 0.80161512,
      "num_input_tokens_seen": 77989995,
      "step": 3618,
      "time_per_iteration": 2.55068302154541
    },
    {
      "auxiliary_loss_clip": 0.01113454,
      "auxiliary_loss_mlp": 0.01030697,
      "balance_loss_clip": 1.048311,
      "balance_loss_mlp": 1.02201319,
      "epoch": 0.4351590212228702,
      "flos": 25082705928960.0,
      "grad_norm": 1.7363392803079163,
      "language_loss": 0.85896677,
      "learning_rate": 2.5114906465067537e-06,
      "loss": 0.88040829,
      "num_input_tokens_seen": 78010980,
      "step": 3619,
      "time_per_iteration": 3.4609029293060303
    },
    {
      "auxiliary_loss_clip": 0.01172403,
      "auxiliary_loss_mlp": 0.01026266,
      "balance_loss_clip": 1.05167282,
      "balance_loss_mlp": 1.01766515,
      "epoch": 0.4352792641135093,
      "flos": 21506541909120.0,
      "grad_norm": 1.949125982000857,
      "language_loss": 0.75005817,
      "learning_rate": 2.5107375414149264e-06,
      "loss": 0.77204478,
      "num_input_tokens_seen": 78030225,
      "step": 3620,
      "time_per_iteration": 2.509326457977295
    },
    {
      "auxiliary_loss_clip": 0.01118833,
      "auxiliary_loss_mlp": 0.01029265,
      "balance_loss_clip": 1.04273868,
      "balance_loss_mlp": 1.02034283,
      "epoch": 0.43539950700414837,
      "flos": 16253457494400.0,
      "grad_norm": 2.311843330396648,
      "language_loss": 0.71927226,
      "learning_rate": 2.5099843588446197e-06,
      "loss": 0.74075323,
      "num_input_tokens_seen": 78048545,
      "step": 3621,
      "time_per_iteration": 2.5363385677337646
    },
    {
      "auxiliary_loss_clip": 0.01138254,
      "auxiliary_loss_mlp": 0.01031514,
      "balance_loss_clip": 1.05215633,
      "balance_loss_mlp": 1.02281761,
      "epoch": 0.4355197498947875,
      "flos": 16691819074560.0,
      "grad_norm": 1.6071659894416652,
      "language_loss": 0.61504698,
      "learning_rate": 2.509231098910091e-06,
      "loss": 0.63674462,
      "num_input_tokens_seen": 78068415,
      "step": 3622,
      "time_per_iteration": 3.354856252670288
    },
    {
      "auxiliary_loss_clip": 0.01155929,
      "auxiliary_loss_mlp": 0.01029368,
      "balance_loss_clip": 1.0559597,
      "balance_loss_mlp": 1.02075517,
      "epoch": 0.4356399927854266,
      "flos": 16362733645440.0,
      "grad_norm": 2.2405377597638565,
      "language_loss": 0.7461971,
      "learning_rate": 2.508477761725611e-06,
      "loss": 0.76805007,
      "num_input_tokens_seen": 78086690,
      "step": 3623,
      "time_per_iteration": 3.242828607559204
    },
    {
      "auxiliary_loss_clip": 0.01176179,
      "auxiliary_loss_mlp": 0.01029671,
      "balance_loss_clip": 1.05535257,
      "balance_loss_mlp": 1.02109432,
      "epoch": 0.43576023567606564,
      "flos": 17202037812480.0,
      "grad_norm": 2.402791545855353,
      "language_loss": 0.80649513,
      "learning_rate": 2.507724347405458e-06,
      "loss": 0.82855362,
      "num_input_tokens_seen": 78104640,
      "step": 3624,
      "time_per_iteration": 2.445854663848877
    },
    {
      "auxiliary_loss_clip": 0.01120554,
      "auxiliary_loss_mlp": 0.01029629,
      "balance_loss_clip": 1.04393983,
      "balance_loss_mlp": 1.02121329,
      "epoch": 0.43588047856670475,
      "flos": 15917656222080.0,
      "grad_norm": 2.233436473490909,
      "language_loss": 0.82006812,
      "learning_rate": 2.5069708560639243e-06,
      "loss": 0.84156996,
      "num_input_tokens_seen": 78122550,
      "step": 3625,
      "time_per_iteration": 2.5374538898468018
    },
    {
      "auxiliary_loss_clip": 0.01146066,
      "auxiliary_loss_mlp": 0.01028852,
      "balance_loss_clip": 1.05117929,
      "balance_loss_mlp": 1.02008963,
      "epoch": 0.4360007214573438,
      "flos": 23659566099840.0,
      "grad_norm": 2.705413983559013,
      "language_loss": 0.61605096,
      "learning_rate": 2.5062172878153158e-06,
      "loss": 0.6378001,
      "num_input_tokens_seen": 78141825,
      "step": 3626,
      "time_per_iteration": 2.5436885356903076
    },
    {
      "auxiliary_loss_clip": 0.01124464,
      "auxiliary_loss_mlp": 0.01032089,
      "balance_loss_clip": 1.05002499,
      "balance_loss_mlp": 1.02250445,
      "epoch": 0.4361209643479829,
      "flos": 21978767036160.0,
      "grad_norm": 1.9465188996827472,
      "language_loss": 0.87297785,
      "learning_rate": 2.505463642773947e-06,
      "loss": 0.89454341,
      "num_input_tokens_seen": 78161790,
      "step": 3627,
      "time_per_iteration": 2.6227526664733887
    },
    {
      "auxiliary_loss_clip": 0.01144854,
      "auxiliary_loss_mlp": 0.00763519,
      "balance_loss_clip": 1.05143547,
      "balance_loss_mlp": 1.00067449,
      "epoch": 0.43624120723862203,
      "flos": 17420159151360.0,
      "grad_norm": 2.514219648593826,
      "language_loss": 0.75091147,
      "learning_rate": 2.504709921054146e-06,
      "loss": 0.76999521,
      "num_input_tokens_seen": 78178605,
      "step": 3628,
      "time_per_iteration": 2.509385824203491
    },
    {
      "auxiliary_loss_clip": 0.01139182,
      "auxiliary_loss_mlp": 0.01031962,
      "balance_loss_clip": 1.04682803,
      "balance_loss_mlp": 1.02279472,
      "epoch": 0.4363614501292611,
      "flos": 17895293280000.0,
      "grad_norm": 4.513723805289859,
      "language_loss": 0.83737111,
      "learning_rate": 2.50395612277025e-06,
      "loss": 0.85908252,
      "num_input_tokens_seen": 78194460,
      "step": 3629,
      "time_per_iteration": 2.5140767097473145
    },
    {
      "auxiliary_loss_clip": 0.01160723,
      "auxiliary_loss_mlp": 0.01028449,
      "balance_loss_clip": 1.0512743,
      "balance_loss_mlp": 1.02008033,
      "epoch": 0.4364816930199002,
      "flos": 20302888135680.0,
      "grad_norm": 2.1144942238942255,
      "language_loss": 0.72881603,
      "learning_rate": 2.503202248036612e-06,
      "loss": 0.75070775,
      "num_input_tokens_seen": 78213315,
      "step": 3630,
      "time_per_iteration": 2.5045201778411865
    },
    {
      "auxiliary_loss_clip": 0.0118435,
      "auxiliary_loss_mlp": 0.01033718,
      "balance_loss_clip": 1.0543474,
      "balance_loss_mlp": 1.02489638,
      "epoch": 0.4366019359105393,
      "flos": 24061334699520.0,
      "grad_norm": 1.7372950643847136,
      "language_loss": 0.73373204,
      "learning_rate": 2.5024482969675927e-06,
      "loss": 0.75591266,
      "num_input_tokens_seen": 78233270,
      "step": 3631,
      "time_per_iteration": 2.45857310295105
    },
    {
      "auxiliary_loss_clip": 0.01133253,
      "auxiliary_loss_mlp": 0.01025508,
      "balance_loss_clip": 1.04852903,
      "balance_loss_mlp": 1.01764035,
      "epoch": 0.43672217880117836,
      "flos": 21754109422080.0,
      "grad_norm": 2.022253891462564,
      "language_loss": 0.84468782,
      "learning_rate": 2.501694269677566e-06,
      "loss": 0.86627543,
      "num_input_tokens_seen": 78251040,
      "step": 3632,
      "time_per_iteration": 2.5529282093048096
    },
    {
      "auxiliary_loss_clip": 0.01176267,
      "auxiliary_loss_mlp": 0.01026618,
      "balance_loss_clip": 1.05301142,
      "balance_loss_mlp": 1.01802921,
      "epoch": 0.4368424216918175,
      "flos": 18035200753920.0,
      "grad_norm": 1.9302069716994963,
      "language_loss": 0.80720055,
      "learning_rate": 2.500940166280918e-06,
      "loss": 0.82922935,
      "num_input_tokens_seen": 78269470,
      "step": 3633,
      "time_per_iteration": 2.469496011734009
    },
    {
      "auxiliary_loss_clip": 0.01167428,
      "auxiliary_loss_mlp": 0.01031377,
      "balance_loss_clip": 1.05148768,
      "balance_loss_mlp": 1.0229255,
      "epoch": 0.4369626645824566,
      "flos": 25447127362560.0,
      "grad_norm": 1.9898720563025336,
      "language_loss": 0.79129708,
      "learning_rate": 2.500185986892045e-06,
      "loss": 0.81328511,
      "num_input_tokens_seen": 78288955,
      "step": 3634,
      "time_per_iteration": 2.5015029907226562
    },
    {
      "auxiliary_loss_clip": 0.0116851,
      "auxiliary_loss_mlp": 0.01032771,
      "balance_loss_clip": 1.05206907,
      "balance_loss_mlp": 1.02424169,
      "epoch": 0.43708290747309564,
      "flos": 25302694775040.0,
      "grad_norm": 2.160090104061873,
      "language_loss": 0.77396762,
      "learning_rate": 2.499431731625355e-06,
      "loss": 0.79598045,
      "num_input_tokens_seen": 78307980,
      "step": 3635,
      "time_per_iteration": 2.495213031768799
    },
    {
      "auxiliary_loss_clip": 0.011879,
      "auxiliary_loss_mlp": 0.01029413,
      "balance_loss_clip": 1.05529296,
      "balance_loss_mlp": 1.01998973,
      "epoch": 0.43720315036373475,
      "flos": 31575103344000.0,
      "grad_norm": 1.8733933394012074,
      "language_loss": 0.79531556,
      "learning_rate": 2.4986774005952686e-06,
      "loss": 0.81748867,
      "num_input_tokens_seen": 78330355,
      "step": 3636,
      "time_per_iteration": 2.529128074645996
    },
    {
      "auxiliary_loss_clip": 0.01171313,
      "auxiliary_loss_mlp": 0.01029927,
      "balance_loss_clip": 1.05647755,
      "balance_loss_mlp": 1.02175498,
      "epoch": 0.43732339325437386,
      "flos": 23112000195840.0,
      "grad_norm": 1.9465508311630513,
      "language_loss": 0.84558129,
      "learning_rate": 2.4979229939162166e-06,
      "loss": 0.86759365,
      "num_input_tokens_seen": 78349135,
      "step": 3637,
      "time_per_iteration": 2.4494831562042236
    },
    {
      "auxiliary_loss_clip": 0.01168537,
      "auxiliary_loss_mlp": 0.01023937,
      "balance_loss_clip": 1.05422688,
      "balance_loss_mlp": 1.01566434,
      "epoch": 0.4374436361450129,
      "flos": 27746272080000.0,
      "grad_norm": 1.6012624121032137,
      "language_loss": 0.80400336,
      "learning_rate": 2.4971685117026433e-06,
      "loss": 0.82592809,
      "num_input_tokens_seen": 78368900,
      "step": 3638,
      "time_per_iteration": 2.498469114303589
    },
    {
      "auxiliary_loss_clip": 0.01172505,
      "auxiliary_loss_mlp": 0.01026223,
      "balance_loss_clip": 1.05378866,
      "balance_loss_mlp": 1.01771784,
      "epoch": 0.437563879035652,
      "flos": 24172370616960.0,
      "grad_norm": 1.467348615614611,
      "language_loss": 0.76309681,
      "learning_rate": 2.4964139540690018e-06,
      "loss": 0.78508413,
      "num_input_tokens_seen": 78392235,
      "step": 3639,
      "time_per_iteration": 2.51043701171875
    },
    {
      "auxiliary_loss_clip": 0.01145479,
      "auxiliary_loss_mlp": 0.01029246,
      "balance_loss_clip": 1.0515523,
      "balance_loss_mlp": 1.02043033,
      "epoch": 0.4376841219262911,
      "flos": 23477211728640.0,
      "grad_norm": 2.1019965662682916,
      "language_loss": 0.72508317,
      "learning_rate": 2.495659321129758e-06,
      "loss": 0.74683046,
      "num_input_tokens_seen": 78409980,
      "step": 3640,
      "time_per_iteration": 2.566718578338623
    },
    {
      "auxiliary_loss_clip": 0.01166287,
      "auxiliary_loss_mlp": 0.01036522,
      "balance_loss_clip": 1.04983711,
      "balance_loss_mlp": 1.02836776,
      "epoch": 0.4378043648169302,
      "flos": 25447809720960.0,
      "grad_norm": 1.7008280931151147,
      "language_loss": 0.75023711,
      "learning_rate": 2.494904612999389e-06,
      "loss": 0.77226526,
      "num_input_tokens_seen": 78428690,
      "step": 3641,
      "time_per_iteration": 2.5219779014587402
    },
    {
      "auxiliary_loss_clip": 0.01067198,
      "auxiliary_loss_mlp": 0.01003329,
      "balance_loss_clip": 1.0167594,
      "balance_loss_mlp": 1.00212479,
      "epoch": 0.4379246077075693,
      "flos": 53914056986880.0,
      "grad_norm": 0.748602284354078,
      "language_loss": 0.56544054,
      "learning_rate": 2.4941498297923843e-06,
      "loss": 0.58614576,
      "num_input_tokens_seen": 78489260,
      "step": 3642,
      "time_per_iteration": 3.7832112312316895
    },
    {
      "auxiliary_loss_clip": 0.0117018,
      "auxiliary_loss_mlp": 0.01025861,
      "balance_loss_clip": 1.05351484,
      "balance_loss_mlp": 1.01771307,
      "epoch": 0.43804485059820836,
      "flos": 20588305605120.0,
      "grad_norm": 1.9664160691605947,
      "language_loss": 0.6983552,
      "learning_rate": 2.4933949716232424e-06,
      "loss": 0.72031558,
      "num_input_tokens_seen": 78506785,
      "step": 3643,
      "time_per_iteration": 2.4410908222198486
    },
    {
      "auxiliary_loss_clip": 0.01142508,
      "auxiliary_loss_mlp": 0.0103006,
      "balance_loss_clip": 1.05175757,
      "balance_loss_mlp": 1.02110696,
      "epoch": 0.43816509348884747,
      "flos": 23876214981120.0,
      "grad_norm": 2.329121432420207,
      "language_loss": 0.73704511,
      "learning_rate": 2.492640038606476e-06,
      "loss": 0.75877076,
      "num_input_tokens_seen": 78525150,
      "step": 3644,
      "time_per_iteration": 2.539618730545044
    },
    {
      "auxiliary_loss_clip": 0.0117145,
      "auxiliary_loss_mlp": 0.01030598,
      "balance_loss_clip": 1.05210304,
      "balance_loss_mlp": 1.02188373,
      "epoch": 0.4382853363794866,
      "flos": 14684448533760.0,
      "grad_norm": 1.9267487199619864,
      "language_loss": 0.78487146,
      "learning_rate": 2.491885030856608e-06,
      "loss": 0.80689198,
      "num_input_tokens_seen": 78543245,
      "step": 3645,
      "time_per_iteration": 2.428431272506714
    },
    {
      "auxiliary_loss_clip": 0.01160944,
      "auxiliary_loss_mlp": 0.01028709,
      "balance_loss_clip": 1.05413651,
      "balance_loss_mlp": 1.02057326,
      "epoch": 0.43840557927012563,
      "flos": 17165301177600.0,
      "grad_norm": 2.085209038771853,
      "language_loss": 0.82645285,
      "learning_rate": 2.4911299484881713e-06,
      "loss": 0.84834945,
      "num_input_tokens_seen": 78560775,
      "step": 3646,
      "time_per_iteration": 3.336498498916626
    },
    {
      "auxiliary_loss_clip": 0.01150848,
      "auxiliary_loss_mlp": 0.01025502,
      "balance_loss_clip": 1.04993951,
      "balance_loss_mlp": 1.01747894,
      "epoch": 0.43852582216076474,
      "flos": 19390685316480.0,
      "grad_norm": 5.4379245278726245,
      "language_loss": 0.81367928,
      "learning_rate": 2.490374791615712e-06,
      "loss": 0.83544278,
      "num_input_tokens_seen": 78580800,
      "step": 3647,
      "time_per_iteration": 2.5091891288757324
    },
    {
      "auxiliary_loss_clip": 0.0119299,
      "auxiliary_loss_mlp": 0.00763706,
      "balance_loss_clip": 1.05729151,
      "balance_loss_mlp": 1.00067329,
      "epoch": 0.43864606505140386,
      "flos": 18075133699200.0,
      "grad_norm": 2.7574399281404096,
      "language_loss": 0.77927428,
      "learning_rate": 2.4896195603537867e-06,
      "loss": 0.79884124,
      "num_input_tokens_seen": 78595410,
      "step": 3648,
      "time_per_iteration": 2.397385358810425
    },
    {
      "auxiliary_loss_clip": 0.01125462,
      "auxiliary_loss_mlp": 0.01028919,
      "balance_loss_clip": 1.05315149,
      "balance_loss_mlp": 1.02045488,
      "epoch": 0.4387663079420429,
      "flos": 19644896845440.0,
      "grad_norm": 2.0292940268414967,
      "language_loss": 0.73760599,
      "learning_rate": 2.488864254816964e-06,
      "loss": 0.75914979,
      "num_input_tokens_seen": 78614100,
      "step": 3649,
      "time_per_iteration": 3.2876951694488525
    },
    {
      "auxiliary_loss_clip": 0.01174093,
      "auxiliary_loss_mlp": 0.01036956,
      "balance_loss_clip": 1.05545819,
      "balance_loss_mlp": 1.02809286,
      "epoch": 0.438886550832682,
      "flos": 19719339782400.0,
      "grad_norm": 6.459611713590826,
      "language_loss": 0.6875475,
      "learning_rate": 2.4881088751198218e-06,
      "loss": 0.70965803,
      "num_input_tokens_seen": 78632260,
      "step": 3650,
      "time_per_iteration": 3.2314679622650146
    },
    {
      "auxiliary_loss_clip": 0.01160188,
      "auxiliary_loss_mlp": 0.01028267,
      "balance_loss_clip": 1.05129576,
      "balance_loss_mlp": 1.01964176,
      "epoch": 0.43900679372332113,
      "flos": 14536675981440.0,
      "grad_norm": 2.725906032242637,
      "language_loss": 0.64348853,
      "learning_rate": 2.4873534213769517e-06,
      "loss": 0.66537309,
      "num_input_tokens_seen": 78647490,
      "step": 3651,
      "time_per_iteration": 2.4474642276763916
    },
    {
      "auxiliary_loss_clip": 0.01139304,
      "auxiliary_loss_mlp": 0.01031814,
      "balance_loss_clip": 1.05315351,
      "balance_loss_mlp": 1.02360094,
      "epoch": 0.4391270366139602,
      "flos": 24056234968320.0,
      "grad_norm": 1.6682082602519543,
      "language_loss": 0.71810389,
      "learning_rate": 2.4865978937029547e-06,
      "loss": 0.73981506,
      "num_input_tokens_seen": 78666470,
      "step": 3652,
      "time_per_iteration": 2.5351741313934326
    },
    {
      "auxiliary_loss_clip": 0.01119484,
      "auxiliary_loss_mlp": 0.01032613,
      "balance_loss_clip": 1.04930758,
      "balance_loss_mlp": 1.0236547,
      "epoch": 0.4392472795045993,
      "flos": 31538510363520.0,
      "grad_norm": 1.5912349278894915,
      "language_loss": 0.6605354,
      "learning_rate": 2.485842292212445e-06,
      "loss": 0.68205631,
      "num_input_tokens_seen": 78687685,
      "step": 3653,
      "time_per_iteration": 2.638922929763794
    },
    {
      "auxiliary_loss_clip": 0.011882,
      "auxiliary_loss_mlp": 0.01030056,
      "balance_loss_clip": 1.05671668,
      "balance_loss_mlp": 1.02149725,
      "epoch": 0.4393675223952384,
      "flos": 14866300114560.0,
      "grad_norm": 1.9302416867016723,
      "language_loss": 0.80380476,
      "learning_rate": 2.485086617020045e-06,
      "loss": 0.82598734,
      "num_input_tokens_seen": 78706180,
      "step": 3654,
      "time_per_iteration": 2.4113941192626953
    },
    {
      "auxiliary_loss_clip": 0.01149401,
      "auxiliary_loss_mlp": 0.010245,
      "balance_loss_clip": 1.04959059,
      "balance_loss_mlp": 1.01587486,
      "epoch": 0.43948776528587746,
      "flos": 14825900292480.0,
      "grad_norm": 2.1060905431072774,
      "language_loss": 0.81650442,
      "learning_rate": 2.4843308682403903e-06,
      "loss": 0.83824337,
      "num_input_tokens_seen": 78723095,
      "step": 3655,
      "time_per_iteration": 2.4563703536987305
    },
    {
      "auxiliary_loss_clip": 0.01185694,
      "auxiliary_loss_mlp": 0.01028623,
      "balance_loss_clip": 1.05482495,
      "balance_loss_mlp": 1.02050447,
      "epoch": 0.4396080081765166,
      "flos": 13914523486080.0,
      "grad_norm": 1.771312483835851,
      "language_loss": 0.82842958,
      "learning_rate": 2.4835750459881294e-06,
      "loss": 0.85057271,
      "num_input_tokens_seen": 78739720,
      "step": 3656,
      "time_per_iteration": 2.3916516304016113
    },
    {
      "auxiliary_loss_clip": 0.01149079,
      "auxiliary_loss_mlp": 0.01036869,
      "balance_loss_clip": 1.04897952,
      "balance_loss_mlp": 1.0275346,
      "epoch": 0.43972825106715563,
      "flos": 18222978078720.0,
      "grad_norm": 1.883646452895321,
      "language_loss": 0.81660694,
      "learning_rate": 2.4828191503779177e-06,
      "loss": 0.83846641,
      "num_input_tokens_seen": 78757820,
      "step": 3657,
      "time_per_iteration": 2.466453790664673
    },
    {
      "auxiliary_loss_clip": 0.01142291,
      "auxiliary_loss_mlp": 0.01026673,
      "balance_loss_clip": 1.04992008,
      "balance_loss_mlp": 1.01807773,
      "epoch": 0.43984849395779474,
      "flos": 16873239692160.0,
      "grad_norm": 2.1980635967594306,
      "language_loss": 0.89555985,
      "learning_rate": 2.482063181524425e-06,
      "loss": 0.9172495,
      "num_input_tokens_seen": 78773720,
      "step": 3658,
      "time_per_iteration": 2.5018508434295654
    },
    {
      "auxiliary_loss_clip": 0.01189443,
      "auxiliary_loss_mlp": 0.01038289,
      "balance_loss_clip": 1.0567528,
      "balance_loss_mlp": 1.02955055,
      "epoch": 0.43996873684843385,
      "flos": 18691504104960.0,
      "grad_norm": 2.2157591733945674,
      "language_loss": 0.81577098,
      "learning_rate": 2.4813071395423307e-06,
      "loss": 0.83804834,
      "num_input_tokens_seen": 78791285,
      "step": 3659,
      "time_per_iteration": 2.408621311187744
    },
    {
      "auxiliary_loss_clip": 0.01173055,
      "auxiliary_loss_mlp": 0.01033051,
      "balance_loss_clip": 1.05435991,
      "balance_loss_mlp": 1.0237174,
      "epoch": 0.4400889797390729,
      "flos": 23653460787840.0,
      "grad_norm": 1.8011518847937302,
      "language_loss": 0.64649594,
      "learning_rate": 2.4805510245463263e-06,
      "loss": 0.66855699,
      "num_input_tokens_seen": 78811440,
      "step": 3660,
      "time_per_iteration": 2.487126588821411
    },
    {
      "auxiliary_loss_clip": 0.01170902,
      "auxiliary_loss_mlp": 0.01033402,
      "balance_loss_clip": 1.05266976,
      "balance_loss_mlp": 1.02436018,
      "epoch": 0.440209222629712,
      "flos": 23149203707520.0,
      "grad_norm": 2.164366007879305,
      "language_loss": 0.60653222,
      "learning_rate": 2.4797948366511137e-06,
      "loss": 0.62857521,
      "num_input_tokens_seen": 78831150,
      "step": 3661,
      "time_per_iteration": 2.4744760990142822
    },
    {
      "auxiliary_loss_clip": 0.01144475,
      "auxiliary_loss_mlp": 0.01034919,
      "balance_loss_clip": 1.04944253,
      "balance_loss_mlp": 1.02641368,
      "epoch": 0.4403294655203511,
      "flos": 24823394668800.0,
      "grad_norm": 2.3243586023971243,
      "language_loss": 0.76402169,
      "learning_rate": 2.4790385759714055e-06,
      "loss": 0.7858156,
      "num_input_tokens_seen": 78850215,
      "step": 3662,
      "time_per_iteration": 2.62807559967041
    },
    {
      "auxiliary_loss_clip": 0.01171244,
      "auxiliary_loss_mlp": 0.01029695,
      "balance_loss_clip": 1.05690086,
      "balance_loss_mlp": 1.02120781,
      "epoch": 0.4404497084109902,
      "flos": 22565080736640.0,
      "grad_norm": 1.6277729916584753,
      "language_loss": 0.71002841,
      "learning_rate": 2.478282242621926e-06,
      "loss": 0.73203778,
      "num_input_tokens_seen": 78870675,
      "step": 3663,
      "time_per_iteration": 2.535227060317993
    },
    {
      "auxiliary_loss_clip": 0.01051195,
      "auxiliary_loss_mlp": 0.0100454,
      "balance_loss_clip": 1.01939011,
      "balance_loss_mlp": 1.00321126,
      "epoch": 0.4405699513016293,
      "flos": 64967073448320.0,
      "grad_norm": 0.8409831315378775,
      "language_loss": 0.59559989,
      "learning_rate": 2.477525836717411e-06,
      "loss": 0.61615723,
      "num_input_tokens_seen": 78938440,
      "step": 3664,
      "time_per_iteration": 3.2285568714141846
    },
    {
      "auxiliary_loss_clip": 0.01170862,
      "auxiliary_loss_mlp": 0.01030515,
      "balance_loss_clip": 1.05177999,
      "balance_loss_mlp": 1.02216434,
      "epoch": 0.4406901941922684,
      "flos": 35661952978560.0,
      "grad_norm": 2.331508199438346,
      "language_loss": 0.79460138,
      "learning_rate": 2.476769358372606e-06,
      "loss": 0.81661516,
      "num_input_tokens_seen": 78960090,
      "step": 3665,
      "time_per_iteration": 2.5901997089385986
    },
    {
      "auxiliary_loss_clip": 0.01139117,
      "auxiliary_loss_mlp": 0.01026376,
      "balance_loss_clip": 1.05162907,
      "balance_loss_mlp": 1.01894355,
      "epoch": 0.44081043708290746,
      "flos": 18040767361920.0,
      "grad_norm": 3.17328299839823,
      "language_loss": 0.75045919,
      "learning_rate": 2.4760128077022683e-06,
      "loss": 0.7721141,
      "num_input_tokens_seen": 78978225,
      "step": 3666,
      "time_per_iteration": 2.565070629119873
    },
    {
      "auxiliary_loss_clip": 0.01121405,
      "auxiliary_loss_mlp": 0.01025168,
      "balance_loss_clip": 1.05060518,
      "balance_loss_mlp": 1.01710391,
      "epoch": 0.44093067997354657,
      "flos": 30153507799680.0,
      "grad_norm": 1.682342936976867,
      "language_loss": 0.6852901,
      "learning_rate": 2.4752561848211672e-06,
      "loss": 0.70675582,
      "num_input_tokens_seen": 79000625,
      "step": 3667,
      "time_per_iteration": 2.6126208305358887
    },
    {
      "auxiliary_loss_clip": 0.011731,
      "auxiliary_loss_mlp": 0.01034368,
      "balance_loss_clip": 1.05914748,
      "balance_loss_mlp": 1.02599335,
      "epoch": 0.4410509228641857,
      "flos": 23255068066560.0,
      "grad_norm": 1.7725359982161295,
      "language_loss": 0.71227455,
      "learning_rate": 2.4744994898440797e-06,
      "loss": 0.73434925,
      "num_input_tokens_seen": 79019415,
      "step": 3668,
      "time_per_iteration": 2.4678609371185303
    },
    {
      "auxiliary_loss_clip": 0.0114822,
      "auxiliary_loss_mlp": 0.0103604,
      "balance_loss_clip": 1.05159271,
      "balance_loss_mlp": 1.02711105,
      "epoch": 0.44117116575482473,
      "flos": 19500571998720.0,
      "grad_norm": 1.958928770721048,
      "language_loss": 0.83773434,
      "learning_rate": 2.473742722885797e-06,
      "loss": 0.85957688,
      "num_input_tokens_seen": 79038435,
      "step": 3669,
      "time_per_iteration": 3.371425151824951
    },
    {
      "auxiliary_loss_clip": 0.01176063,
      "auxiliary_loss_mlp": 0.00763588,
      "balance_loss_clip": 1.05887735,
      "balance_loss_mlp": 1.00067163,
      "epoch": 0.44129140864546385,
      "flos": 27053124353280.0,
      "grad_norm": 2.0819537096438743,
      "language_loss": 0.65102518,
      "learning_rate": 2.4729858840611197e-06,
      "loss": 0.67042172,
      "num_input_tokens_seen": 79057345,
      "step": 3670,
      "time_per_iteration": 2.549913167953491
    },
    {
      "auxiliary_loss_clip": 0.01187426,
      "auxiliary_loss_mlp": 0.01026531,
      "balance_loss_clip": 1.05753946,
      "balance_loss_mlp": 1.01855588,
      "epoch": 0.4414116515361029,
      "flos": 26102101910400.0,
      "grad_norm": 4.809380728779214,
      "language_loss": 0.72867262,
      "learning_rate": 2.4722289734848605e-06,
      "loss": 0.75081217,
      "num_input_tokens_seen": 79077810,
      "step": 3671,
      "time_per_iteration": 2.4851484298706055
    },
    {
      "auxiliary_loss_clip": 0.01143818,
      "auxiliary_loss_mlp": 0.01028882,
      "balance_loss_clip": 1.05617523,
      "balance_loss_mlp": 1.02051973,
      "epoch": 0.441531894426742,
      "flos": 21906083865600.0,
      "grad_norm": 2.0540421767820445,
      "language_loss": 0.77928531,
      "learning_rate": 2.471471991271841e-06,
      "loss": 0.80101234,
      "num_input_tokens_seen": 79094935,
      "step": 3672,
      "time_per_iteration": 3.3719658851623535
    },
    {
      "auxiliary_loss_clip": 0.01164399,
      "auxiliary_loss_mlp": 0.01026875,
      "balance_loss_clip": 1.05276191,
      "balance_loss_mlp": 1.01832795,
      "epoch": 0.4416521373173811,
      "flos": 23437099215360.0,
      "grad_norm": 1.919010769417049,
      "language_loss": 0.79262859,
      "learning_rate": 2.470714937536896e-06,
      "loss": 0.81454134,
      "num_input_tokens_seen": 79113660,
      "step": 3673,
      "time_per_iteration": 2.470294237136841
    },
    {
      "auxiliary_loss_clip": 0.01125172,
      "auxiliary_loss_mlp": 0.0103085,
      "balance_loss_clip": 1.0489316,
      "balance_loss_mlp": 1.02226722,
      "epoch": 0.4417723802080202,
      "flos": 20334345471360.0,
      "grad_norm": 1.661876865895167,
      "language_loss": 0.70376062,
      "learning_rate": 2.469957812394868e-06,
      "loss": 0.72532082,
      "num_input_tokens_seen": 79132470,
      "step": 3674,
      "time_per_iteration": 2.565774917602539
    },
    {
      "auxiliary_loss_clip": 0.01186638,
      "auxiliary_loss_mlp": 0.01029041,
      "balance_loss_clip": 1.05809879,
      "balance_loss_mlp": 1.02052379,
      "epoch": 0.4418926230986593,
      "flos": 18880682060160.0,
      "grad_norm": 1.8807257481700441,
      "language_loss": 0.76321942,
      "learning_rate": 2.4692006159606148e-06,
      "loss": 0.78537619,
      "num_input_tokens_seen": 79150000,
      "step": 3675,
      "time_per_iteration": 2.4257280826568604
    },
    {
      "auxiliary_loss_clip": 0.01185678,
      "auxiliary_loss_mlp": 0.01030094,
      "balance_loss_clip": 1.05552435,
      "balance_loss_mlp": 1.02163005,
      "epoch": 0.4420128659892984,
      "flos": 19464409981440.0,
      "grad_norm": 1.8759780191007915,
      "language_loss": 0.78543627,
      "learning_rate": 2.468443348349e-06,
      "loss": 0.80759394,
      "num_input_tokens_seen": 79167875,
      "step": 3676,
      "time_per_iteration": 3.1443326473236084
    },
    {
      "auxiliary_loss_clip": 0.01127999,
      "auxiliary_loss_mlp": 0.01033494,
      "balance_loss_clip": 1.0489912,
      "balance_loss_mlp": 1.02381372,
      "epoch": 0.44213310887993745,
      "flos": 17894359526400.0,
      "grad_norm": 2.4823724357011616,
      "language_loss": 0.82399458,
      "learning_rate": 2.467686009674902e-06,
      "loss": 0.84560943,
      "num_input_tokens_seen": 79182325,
      "step": 3677,
      "time_per_iteration": 3.320093870162964
    },
    {
      "auxiliary_loss_clip": 0.01166445,
      "auxiliary_loss_mlp": 0.01029892,
      "balance_loss_clip": 1.05101466,
      "balance_loss_mlp": 1.02099323,
      "epoch": 0.44225335177057656,
      "flos": 19204667758080.0,
      "grad_norm": 1.8654601966993722,
      "language_loss": 0.85179412,
      "learning_rate": 2.466928600053209e-06,
      "loss": 0.87375748,
      "num_input_tokens_seen": 79197630,
      "step": 3678,
      "time_per_iteration": 2.4332807064056396
    },
    {
      "auxiliary_loss_clip": 0.0115601,
      "auxiliary_loss_mlp": 0.01026681,
      "balance_loss_clip": 1.05135798,
      "balance_loss_mlp": 1.01846766,
      "epoch": 0.4423735946612157,
      "flos": 23471321898240.0,
      "grad_norm": 1.8248486645074782,
      "language_loss": 0.71393889,
      "learning_rate": 2.466171119598818e-06,
      "loss": 0.73576581,
      "num_input_tokens_seen": 79217600,
      "step": 3679,
      "time_per_iteration": 2.5327563285827637
    },
    {
      "auxiliary_loss_clip": 0.01177879,
      "auxiliary_loss_mlp": 0.01033968,
      "balance_loss_clip": 1.05263925,
      "balance_loss_mlp": 1.02529597,
      "epoch": 0.44249383755185473,
      "flos": 26685398868480.0,
      "grad_norm": 1.8872031919742798,
      "language_loss": 0.77459937,
      "learning_rate": 2.465413568426639e-06,
      "loss": 0.79671788,
      "num_input_tokens_seen": 79238550,
      "step": 3680,
      "time_per_iteration": 2.516883134841919
    },
    {
      "auxiliary_loss_clip": 0.01167014,
      "auxiliary_loss_mlp": 0.010246,
      "balance_loss_clip": 1.05362177,
      "balance_loss_mlp": 1.01685166,
      "epoch": 0.44261408044249384,
      "flos": 23147659422720.0,
      "grad_norm": 2.471646533675412,
      "language_loss": 0.81122637,
      "learning_rate": 2.464655946651591e-06,
      "loss": 0.83314252,
      "num_input_tokens_seen": 79257555,
      "step": 3681,
      "time_per_iteration": 2.476849317550659
    },
    {
      "auxiliary_loss_clip": 0.01177036,
      "auxiliary_loss_mlp": 0.01029152,
      "balance_loss_clip": 1.0574224,
      "balance_loss_mlp": 1.0207355,
      "epoch": 0.44273432333313295,
      "flos": 24462564595200.0,
      "grad_norm": 2.0268041719936543,
      "language_loss": 0.81043327,
      "learning_rate": 2.4638982543886065e-06,
      "loss": 0.83249509,
      "num_input_tokens_seen": 79277595,
      "step": 3682,
      "time_per_iteration": 2.4976933002471924
    },
    {
      "auxiliary_loss_clip": 0.01174802,
      "auxiliary_loss_mlp": 0.0103535,
      "balance_loss_clip": 1.0558691,
      "balance_loss_mlp": 1.02689242,
      "epoch": 0.442854566223772,
      "flos": 17528932512000.0,
      "grad_norm": 9.69083777776178,
      "language_loss": 0.87227261,
      "learning_rate": 2.4631404917526254e-06,
      "loss": 0.89437413,
      "num_input_tokens_seen": 79294550,
      "step": 3683,
      "time_per_iteration": 2.4609546661376953
    },
    {
      "auxiliary_loss_clip": 0.01165195,
      "auxiliary_loss_mlp": 0.01027326,
      "balance_loss_clip": 1.05210638,
      "balance_loss_mlp": 1.01943398,
      "epoch": 0.4429748091144111,
      "flos": 24896293320960.0,
      "grad_norm": 1.6920251277605036,
      "language_loss": 0.79115558,
      "learning_rate": 2.4623826588586e-06,
      "loss": 0.81308079,
      "num_input_tokens_seen": 79314820,
      "step": 3684,
      "time_per_iteration": 2.491969108581543
    },
    {
      "auxiliary_loss_clip": 0.01151848,
      "auxiliary_loss_mlp": 0.01031889,
      "balance_loss_clip": 1.04962802,
      "balance_loss_mlp": 1.02289438,
      "epoch": 0.4430950520050502,
      "flos": 21614704738560.0,
      "grad_norm": 1.7836736265672408,
      "language_loss": 0.82626003,
      "learning_rate": 2.461624755821492e-06,
      "loss": 0.84809738,
      "num_input_tokens_seen": 79334300,
      "step": 3685,
      "time_per_iteration": 2.506840229034424
    },
    {
      "auxiliary_loss_clip": 0.01142849,
      "auxiliary_loss_mlp": 0.01024839,
      "balance_loss_clip": 1.05109644,
      "balance_loss_mlp": 1.01698887,
      "epoch": 0.4432152948956893,
      "flos": 24572271709440.0,
      "grad_norm": 1.7522927979709753,
      "language_loss": 0.76716673,
      "learning_rate": 2.4608667827562763e-06,
      "loss": 0.78884363,
      "num_input_tokens_seen": 79353630,
      "step": 3686,
      "time_per_iteration": 2.5531423091888428
    },
    {
      "auxiliary_loss_clip": 0.01176659,
      "auxiliary_loss_mlp": 0.01031689,
      "balance_loss_clip": 1.055722,
      "balance_loss_mlp": 1.02314723,
      "epoch": 0.4433355377863284,
      "flos": 21762261809280.0,
      "grad_norm": 2.12396996293802,
      "language_loss": 0.89869851,
      "learning_rate": 2.460108739777936e-06,
      "loss": 0.92078197,
      "num_input_tokens_seen": 79372765,
      "step": 3687,
      "time_per_iteration": 2.4826157093048096
    },
    {
      "auxiliary_loss_clip": 0.01157357,
      "auxiliary_loss_mlp": 0.01029924,
      "balance_loss_clip": 1.0546813,
      "balance_loss_mlp": 1.02138841,
      "epoch": 0.44345578067696745,
      "flos": 20084479488000.0,
      "grad_norm": 1.5974800149121162,
      "language_loss": 0.76480985,
      "learning_rate": 2.4593506270014656e-06,
      "loss": 0.78668267,
      "num_input_tokens_seen": 79391735,
      "step": 3688,
      "time_per_iteration": 2.4870541095733643
    },
    {
      "auxiliary_loss_clip": 0.01161638,
      "auxiliary_loss_mlp": 0.01027759,
      "balance_loss_clip": 1.05108809,
      "balance_loss_mlp": 1.01942611,
      "epoch": 0.44357602356760656,
      "flos": 24169497528960.0,
      "grad_norm": 1.6461131333415144,
      "language_loss": 0.82080132,
      "learning_rate": 2.45859244454187e-06,
      "loss": 0.84269536,
      "num_input_tokens_seen": 79411525,
      "step": 3689,
      "time_per_iteration": 2.550636053085327
    },
    {
      "auxiliary_loss_clip": 0.01169545,
      "auxiliary_loss_mlp": 0.01027946,
      "balance_loss_clip": 1.05495358,
      "balance_loss_mlp": 1.02015519,
      "epoch": 0.44369626645824567,
      "flos": 22707717644160.0,
      "grad_norm": 1.741849472187036,
      "language_loss": 0.66401744,
      "learning_rate": 2.4578341925141655e-06,
      "loss": 0.68599236,
      "num_input_tokens_seen": 79430740,
      "step": 3690,
      "time_per_iteration": 2.4910192489624023
    },
    {
      "auxiliary_loss_clip": 0.01179462,
      "auxiliary_loss_mlp": 0.01025428,
      "balance_loss_clip": 1.05479777,
      "balance_loss_mlp": 1.01664853,
      "epoch": 0.4438165093488847,
      "flos": 38030225420160.0,
      "grad_norm": 1.9121724741111004,
      "language_loss": 0.71961272,
      "learning_rate": 2.457075871033378e-06,
      "loss": 0.74166155,
      "num_input_tokens_seen": 79452615,
      "step": 3691,
      "time_per_iteration": 2.6078453063964844
    },
    {
      "auxiliary_loss_clip": 0.01143557,
      "auxiliary_loss_mlp": 0.01025529,
      "balance_loss_clip": 1.05234361,
      "balance_loss_mlp": 1.01735091,
      "epoch": 0.44393675223952384,
      "flos": 15523213996800.0,
      "grad_norm": 1.9985631047790897,
      "language_loss": 0.88896966,
      "learning_rate": 2.4563174802145445e-06,
      "loss": 0.91066051,
      "num_input_tokens_seen": 79469865,
      "step": 3692,
      "time_per_iteration": 2.485480546951294
    },
    {
      "auxiliary_loss_clip": 0.01063936,
      "auxiliary_loss_mlp": 0.01003626,
      "balance_loss_clip": 1.02172112,
      "balance_loss_mlp": 1.00226104,
      "epoch": 0.44405699513016295,
      "flos": 64574893779840.0,
      "grad_norm": 0.6334892616799727,
      "language_loss": 0.48645616,
      "learning_rate": 2.455559020172712e-06,
      "loss": 0.50713181,
      "num_input_tokens_seen": 79537220,
      "step": 3693,
      "time_per_iteration": 3.1618378162384033
    },
    {
      "auxiliary_loss_clip": 0.01136697,
      "auxiliary_loss_mlp": 0.01038172,
      "balance_loss_clip": 1.05596483,
      "balance_loss_mlp": 1.02953506,
      "epoch": 0.444177238020802,
      "flos": 23987394552960.0,
      "grad_norm": 2.397392671542597,
      "language_loss": 0.89748502,
      "learning_rate": 2.4548004910229385e-06,
      "loss": 0.91923368,
      "num_input_tokens_seen": 79554795,
      "step": 3694,
      "time_per_iteration": 2.628120183944702
    },
    {
      "auxiliary_loss_clip": 0.01176037,
      "auxiliary_loss_mlp": 0.00764013,
      "balance_loss_clip": 1.05644131,
      "balance_loss_mlp": 1.00070786,
      "epoch": 0.4442974809114411,
      "flos": 22563069575040.0,
      "grad_norm": 2.2172444591353107,
      "language_loss": 0.87127411,
      "learning_rate": 2.4540418928802913e-06,
      "loss": 0.89067459,
      "num_input_tokens_seen": 79573530,
      "step": 3695,
      "time_per_iteration": 2.4922924041748047
    },
    {
      "auxiliary_loss_clip": 0.01156449,
      "auxiliary_loss_mlp": 0.01031828,
      "balance_loss_clip": 1.0512948,
      "balance_loss_mlp": 1.02289367,
      "epoch": 0.4444177238020802,
      "flos": 17675699483520.0,
      "grad_norm": 2.8643462519079854,
      "language_loss": 0.65817106,
      "learning_rate": 2.4532832258598506e-06,
      "loss": 0.68005383,
      "num_input_tokens_seen": 79591360,
      "step": 3696,
      "time_per_iteration": 3.226816415786743
    },
    {
      "auxiliary_loss_clip": 0.01184374,
      "auxiliary_loss_mlp": 0.01026062,
      "balance_loss_clip": 1.05612481,
      "balance_loss_mlp": 1.01759839,
      "epoch": 0.4445379666927193,
      "flos": 28621594609920.0,
      "grad_norm": 1.7962066310276905,
      "language_loss": 0.80778468,
      "learning_rate": 2.4525244900767047e-06,
      "loss": 0.82988906,
      "num_input_tokens_seen": 79612175,
      "step": 3697,
      "time_per_iteration": 2.503439426422119
    },
    {
      "auxiliary_loss_clip": 0.01072983,
      "auxiliary_loss_mlp": 0.01001959,
      "balance_loss_clip": 1.02577829,
      "balance_loss_mlp": 1.00073075,
      "epoch": 0.4446582095833584,
      "flos": 70487370115200.0,
      "grad_norm": 0.7726120850558645,
      "language_loss": 0.60552257,
      "learning_rate": 2.4517656856459536e-06,
      "loss": 0.62627202,
      "num_input_tokens_seen": 79678020,
      "step": 3698,
      "time_per_iteration": 3.151158332824707
    },
    {
      "auxiliary_loss_clip": 0.01170439,
      "auxiliary_loss_mlp": 0.01033914,
      "balance_loss_clip": 1.05225992,
      "balance_loss_mlp": 1.02581656,
      "epoch": 0.4447784524739975,
      "flos": 26505199313280.0,
      "grad_norm": 1.656308441090315,
      "language_loss": 0.68097639,
      "learning_rate": 2.4510068126827073e-06,
      "loss": 0.70301998,
      "num_input_tokens_seen": 79699020,
      "step": 3699,
      "time_per_iteration": 3.3736932277679443
    },
    {
      "auxiliary_loss_clip": 0.01158756,
      "auxiliary_loss_mlp": 0.01035414,
      "balance_loss_clip": 1.05345583,
      "balance_loss_mlp": 1.02707577,
      "epoch": 0.44489869536463655,
      "flos": 11656209553920.0,
      "grad_norm": 6.3595659630556005,
      "language_loss": 0.81421053,
      "learning_rate": 2.450247871302086e-06,
      "loss": 0.8361522,
      "num_input_tokens_seen": 79716795,
      "step": 3700,
      "time_per_iteration": 2.4832677841186523
    },
    {
      "auxiliary_loss_clip": 0.01175798,
      "auxiliary_loss_mlp": 0.01027704,
      "balance_loss_clip": 1.0545212,
      "balance_loss_mlp": 1.01982427,
      "epoch": 0.44501893825527566,
      "flos": 20448469958400.0,
      "grad_norm": 2.313075096495458,
      "language_loss": 0.83370793,
      "learning_rate": 2.44948886161922e-06,
      "loss": 0.85574299,
      "num_input_tokens_seen": 79735810,
      "step": 3701,
      "time_per_iteration": 2.5322635173797607
    },
    {
      "auxiliary_loss_clip": 0.01175748,
      "auxiliary_loss_mlp": 0.01026403,
      "balance_loss_clip": 1.05587935,
      "balance_loss_mlp": 1.01868701,
      "epoch": 0.4451391811459148,
      "flos": 18261079430400.0,
      "grad_norm": 1.6084210201101368,
      "language_loss": 0.85110223,
      "learning_rate": 2.4487297837492524e-06,
      "loss": 0.87312371,
      "num_input_tokens_seen": 79754975,
      "step": 3702,
      "time_per_iteration": 3.270439863204956
    },
    {
      "auxiliary_loss_clip": 0.01142784,
      "auxiliary_loss_mlp": 0.01029359,
      "balance_loss_clip": 1.05190527,
      "balance_loss_mlp": 1.02097893,
      "epoch": 0.44525942403655383,
      "flos": 16910155895040.0,
      "grad_norm": 1.9247161420488583,
      "language_loss": 0.62377113,
      "learning_rate": 2.4479706378073323e-06,
      "loss": 0.64549261,
      "num_input_tokens_seen": 79773515,
      "step": 3703,
      "time_per_iteration": 2.6739790439605713
    },
    {
      "auxiliary_loss_clip": 0.01131238,
      "auxiliary_loss_mlp": 0.01027209,
      "balance_loss_clip": 1.0463264,
      "balance_loss_mlp": 1.01921606,
      "epoch": 0.44537966692719294,
      "flos": 23258838994560.0,
      "grad_norm": 1.5727117342734176,
      "language_loss": 0.83750415,
      "learning_rate": 2.447211423908623e-06,
      "loss": 0.8590886,
      "num_input_tokens_seen": 79793560,
      "step": 3704,
      "time_per_iteration": 3.283668279647827
    },
    {
      "auxiliary_loss_clip": 0.01173876,
      "auxiliary_loss_mlp": 0.01027256,
      "balance_loss_clip": 1.05376112,
      "balance_loss_mlp": 1.01920366,
      "epoch": 0.445499909817832,
      "flos": 21724160457600.0,
      "grad_norm": 2.295431497373646,
      "language_loss": 0.7439239,
      "learning_rate": 2.4464521421682966e-06,
      "loss": 0.7659353,
      "num_input_tokens_seen": 79811150,
      "step": 3705,
      "time_per_iteration": 2.464444875717163
    },
    {
      "auxiliary_loss_clip": 0.01166224,
      "auxiliary_loss_mlp": 0.01024693,
      "balance_loss_clip": 1.05462337,
      "balance_loss_mlp": 1.01706958,
      "epoch": 0.4456201527084711,
      "flos": 23987969170560.0,
      "grad_norm": 1.4100867967009847,
      "language_loss": 0.87436712,
      "learning_rate": 2.4456927927015345e-06,
      "loss": 0.89627624,
      "num_input_tokens_seen": 79832190,
      "step": 3706,
      "time_per_iteration": 2.4772579669952393
    },
    {
      "auxiliary_loss_clip": 0.0116715,
      "auxiliary_loss_mlp": 0.0103186,
      "balance_loss_clip": 1.0576086,
      "balance_loss_mlp": 1.02279973,
      "epoch": 0.4457403955991102,
      "flos": 18807065136000.0,
      "grad_norm": 2.034032312400158,
      "language_loss": 0.76659322,
      "learning_rate": 2.4449333756235307e-06,
      "loss": 0.78858328,
      "num_input_tokens_seen": 79848905,
      "step": 3707,
      "time_per_iteration": 2.479116439819336
    },
    {
      "auxiliary_loss_clip": 0.01176042,
      "auxiliary_loss_mlp": 0.01036432,
      "balance_loss_clip": 1.05490422,
      "balance_loss_mlp": 1.02784932,
      "epoch": 0.4458606384897493,
      "flos": 19207756327680.0,
      "grad_norm": 2.225344106043708,
      "language_loss": 0.78542346,
      "learning_rate": 2.4441738910494876e-06,
      "loss": 0.80754817,
      "num_input_tokens_seen": 79863640,
      "step": 3708,
      "time_per_iteration": 2.4359796047210693
    },
    {
      "auxiliary_loss_clip": 0.01164405,
      "auxiliary_loss_mlp": 0.01033269,
      "balance_loss_clip": 1.05187988,
      "balance_loss_mlp": 1.02489424,
      "epoch": 0.4459808813803884,
      "flos": 21361283308800.0,
      "grad_norm": 1.6699179867984635,
      "language_loss": 0.82155168,
      "learning_rate": 2.4434143390946176e-06,
      "loss": 0.84352839,
      "num_input_tokens_seen": 79882450,
      "step": 3709,
      "time_per_iteration": 2.4934604167938232
    },
    {
      "auxiliary_loss_clip": 0.01140056,
      "auxiliary_loss_mlp": 0.01029277,
      "balance_loss_clip": 1.04868197,
      "balance_loss_mlp": 1.02072358,
      "epoch": 0.4461011242710275,
      "flos": 23288967527040.0,
      "grad_norm": 1.8439956291296107,
      "language_loss": 0.85271466,
      "learning_rate": 2.4426547198741457e-06,
      "loss": 0.87440795,
      "num_input_tokens_seen": 79900655,
      "step": 3710,
      "time_per_iteration": 2.5815532207489014
    },
    {
      "auxiliary_loss_clip": 0.01129064,
      "auxiliary_loss_mlp": 0.0103097,
      "balance_loss_clip": 1.05202425,
      "balance_loss_mlp": 1.0228343,
      "epoch": 0.44622136716166655,
      "flos": 20193001453440.0,
      "grad_norm": 2.3639526633457377,
      "language_loss": 0.74590313,
      "learning_rate": 2.441895033503305e-06,
      "loss": 0.76750344,
      "num_input_tokens_seen": 79918575,
      "step": 3711,
      "time_per_iteration": 2.5526912212371826
    },
    {
      "auxiliary_loss_clip": 0.0117043,
      "auxiliary_loss_mlp": 0.01032191,
      "balance_loss_clip": 1.05340171,
      "balance_loss_mlp": 1.0232501,
      "epoch": 0.44634161005230566,
      "flos": 21283033530240.0,
      "grad_norm": 1.6951554902840211,
      "language_loss": 0.81890422,
      "learning_rate": 2.4411352800973375e-06,
      "loss": 0.84093046,
      "num_input_tokens_seen": 79937010,
      "step": 3712,
      "time_per_iteration": 2.4535343647003174
    },
    {
      "auxiliary_loss_clip": 0.01138319,
      "auxiliary_loss_mlp": 0.01027984,
      "balance_loss_clip": 1.0491575,
      "balance_loss_mlp": 1.0192399,
      "epoch": 0.44646185294294477,
      "flos": 22929358515840.0,
      "grad_norm": 2.282796522320087,
      "language_loss": 0.7557168,
      "learning_rate": 2.4403754597715005e-06,
      "loss": 0.77737975,
      "num_input_tokens_seen": 79956455,
      "step": 3713,
      "time_per_iteration": 2.531940460205078
    },
    {
      "auxiliary_loss_clip": 0.01159664,
      "auxiliary_loss_mlp": 0.01034928,
      "balance_loss_clip": 1.04897785,
      "balance_loss_mlp": 1.0256598,
      "epoch": 0.4465820958335838,
      "flos": 22637692080000.0,
      "grad_norm": 2.360868212339568,
      "language_loss": 0.92910033,
      "learning_rate": 2.4396155726410553e-06,
      "loss": 0.95104623,
      "num_input_tokens_seen": 79975065,
      "step": 3714,
      "time_per_iteration": 2.5019724369049072
    },
    {
      "auxiliary_loss_clip": 0.01177362,
      "auxiliary_loss_mlp": 0.01027234,
      "balance_loss_clip": 1.0527432,
      "balance_loss_mlp": 1.01925325,
      "epoch": 0.44670233872422294,
      "flos": 22672525294080.0,
      "grad_norm": 2.4848116740977697,
      "language_loss": 0.91114068,
      "learning_rate": 2.438855618821278e-06,
      "loss": 0.93318659,
      "num_input_tokens_seen": 79990865,
      "step": 3715,
      "time_per_iteration": 2.5113203525543213
    },
    {
      "auxiliary_loss_clip": 0.01162222,
      "auxiliary_loss_mlp": 0.01031138,
      "balance_loss_clip": 1.04904628,
      "balance_loss_mlp": 1.02264428,
      "epoch": 0.44682258161486205,
      "flos": 23582178247680.0,
      "grad_norm": 1.6282300358166526,
      "language_loss": 0.67165083,
      "learning_rate": 2.4380955984274517e-06,
      "loss": 0.69358444,
      "num_input_tokens_seen": 80009520,
      "step": 3716,
      "time_per_iteration": 2.4700863361358643
    },
    {
      "auxiliary_loss_clip": 0.01169011,
      "auxiliary_loss_mlp": 0.01035671,
      "balance_loss_clip": 1.05113196,
      "balance_loss_mlp": 1.02758896,
      "epoch": 0.4469428245055011,
      "flos": 26501356558080.0,
      "grad_norm": 1.658853914927923,
      "language_loss": 0.76939428,
      "learning_rate": 2.4373355115748716e-06,
      "loss": 0.79144108,
      "num_input_tokens_seen": 80030350,
      "step": 3717,
      "time_per_iteration": 2.5420291423797607
    },
    {
      "auxiliary_loss_clip": 0.01150738,
      "auxiliary_loss_mlp": 0.01031496,
      "balance_loss_clip": 1.05203068,
      "balance_loss_mlp": 1.02274573,
      "epoch": 0.4470630673961402,
      "flos": 21504925797120.0,
      "grad_norm": 1.686356699989055,
      "language_loss": 0.72169089,
      "learning_rate": 2.436575358378842e-06,
      "loss": 0.74351323,
      "num_input_tokens_seen": 80049840,
      "step": 3718,
      "time_per_iteration": 2.50437331199646
    },
    {
      "auxiliary_loss_clip": 0.01167251,
      "auxiliary_loss_mlp": 0.0103191,
      "balance_loss_clip": 1.05460501,
      "balance_loss_mlp": 1.02271855,
      "epoch": 0.44718331028677927,
      "flos": 16173986653440.0,
      "grad_norm": 3.6961995642420487,
      "language_loss": 0.83117914,
      "learning_rate": 2.4358151389546782e-06,
      "loss": 0.85317075,
      "num_input_tokens_seen": 80066525,
      "step": 3719,
      "time_per_iteration": 2.4982588291168213
    },
    {
      "auxiliary_loss_clip": 0.01185726,
      "auxiliary_loss_mlp": 0.01031734,
      "balance_loss_clip": 1.05546641,
      "balance_loss_mlp": 1.02307987,
      "epoch": 0.4473035531774184,
      "flos": 19681238430720.0,
      "grad_norm": 2.215861306443257,
      "language_loss": 0.76690769,
      "learning_rate": 2.4350548534177035e-06,
      "loss": 0.78908229,
      "num_input_tokens_seen": 80083355,
      "step": 3720,
      "time_per_iteration": 2.399819850921631
    },
    {
      "auxiliary_loss_clip": 0.01142642,
      "auxiliary_loss_mlp": 0.01032894,
      "balance_loss_clip": 1.0513339,
      "balance_loss_mlp": 1.02521062,
      "epoch": 0.4474237960680575,
      "flos": 41427590515200.0,
      "grad_norm": 1.5486249133206893,
      "language_loss": 0.6682502,
      "learning_rate": 2.434294501883254e-06,
      "loss": 0.69000554,
      "num_input_tokens_seen": 80106450,
      "step": 3721,
      "time_per_iteration": 2.7164411544799805
    },
    {
      "auxiliary_loss_clip": 0.01146654,
      "auxiliary_loss_mlp": 0.0103011,
      "balance_loss_clip": 1.04784989,
      "balance_loss_mlp": 1.02148521,
      "epoch": 0.44754403895869654,
      "flos": 22891328991360.0,
      "grad_norm": 1.6817321486877723,
      "language_loss": 0.65677023,
      "learning_rate": 2.433534084466674e-06,
      "loss": 0.67853791,
      "num_input_tokens_seen": 80125670,
      "step": 3722,
      "time_per_iteration": 3.3199877738952637
    },
    {
      "auxiliary_loss_clip": 0.01181532,
      "auxiliary_loss_mlp": 0.01027752,
      "balance_loss_clip": 1.05417609,
      "balance_loss_mlp": 1.01946068,
      "epoch": 0.44766428184933565,
      "flos": 25630271832960.0,
      "grad_norm": 1.6413005810510426,
      "language_loss": 0.7104544,
      "learning_rate": 2.4327736012833178e-06,
      "loss": 0.73254716,
      "num_input_tokens_seen": 80147390,
      "step": 3723,
      "time_per_iteration": 2.4865431785583496
    },
    {
      "auxiliary_loss_clip": 0.01171549,
      "auxiliary_loss_mlp": 0.01033204,
      "balance_loss_clip": 1.05430961,
      "balance_loss_mlp": 1.02479935,
      "epoch": 0.44778452473997477,
      "flos": 20448972748800.0,
      "grad_norm": 2.0468483539001556,
      "language_loss": 0.76963258,
      "learning_rate": 2.4320130524485506e-06,
      "loss": 0.7916801,
      "num_input_tokens_seen": 80166185,
      "step": 3724,
      "time_per_iteration": 2.446181058883667
    },
    {
      "auxiliary_loss_clip": 0.01151242,
      "auxiliary_loss_mlp": 0.010251,
      "balance_loss_clip": 1.05543447,
      "balance_loss_mlp": 1.01764607,
      "epoch": 0.4479047676306138,
      "flos": 21975462984960.0,
      "grad_norm": 4.294670207796363,
      "language_loss": 0.79626513,
      "learning_rate": 2.431252438077746e-06,
      "loss": 0.81802857,
      "num_input_tokens_seen": 80185685,
      "step": 3725,
      "time_per_iteration": 3.3264570236206055
    },
    {
      "auxiliary_loss_clip": 0.01174841,
      "auxiliary_loss_mlp": 0.00763459,
      "balance_loss_clip": 1.05286789,
      "balance_loss_mlp": 1.00054598,
      "epoch": 0.44802501052125293,
      "flos": 21467219495040.0,
      "grad_norm": 2.3404871547673305,
      "language_loss": 0.7730754,
      "learning_rate": 2.4304917582862906e-06,
      "loss": 0.79245836,
      "num_input_tokens_seen": 80204865,
      "step": 3726,
      "time_per_iteration": 2.4551732540130615
    },
    {
      "auxiliary_loss_clip": 0.01183236,
      "auxiliary_loss_mlp": 0.01028018,
      "balance_loss_clip": 1.05398405,
      "balance_loss_mlp": 1.01988196,
      "epoch": 0.44814525341189204,
      "flos": 22126970551680.0,
      "grad_norm": 1.904719991320197,
      "language_loss": 0.8765772,
      "learning_rate": 2.4297310131895774e-06,
      "loss": 0.89868975,
      "num_input_tokens_seen": 80223410,
      "step": 3727,
      "time_per_iteration": 2.4239656925201416
    },
    {
      "auxiliary_loss_clip": 0.01169893,
      "auxiliary_loss_mlp": 0.01031785,
      "balance_loss_clip": 1.05285311,
      "balance_loss_mlp": 1.02307653,
      "epoch": 0.4482654963025311,
      "flos": 16653933204480.0,
      "grad_norm": 1.94850857925264,
      "language_loss": 0.74924505,
      "learning_rate": 2.4289702029030113e-06,
      "loss": 0.77126175,
      "num_input_tokens_seen": 80240880,
      "step": 3728,
      "time_per_iteration": 3.241889715194702
    },
    {
      "auxiliary_loss_clip": 0.01171947,
      "auxiliary_loss_mlp": 0.01027051,
      "balance_loss_clip": 1.05626512,
      "balance_loss_mlp": 1.01886129,
      "epoch": 0.4483857391931702,
      "flos": 18841251905280.0,
      "grad_norm": 1.7704911897689553,
      "language_loss": 0.83301461,
      "learning_rate": 2.4282093275420057e-06,
      "loss": 0.85500461,
      "num_input_tokens_seen": 80259910,
      "step": 3729,
      "time_per_iteration": 2.490283489227295
    },
    {
      "auxiliary_loss_clip": 0.01175912,
      "auxiliary_loss_mlp": 0.01031034,
      "balance_loss_clip": 1.05572116,
      "balance_loss_mlp": 1.02324331,
      "epoch": 0.4485059820838093,
      "flos": 20372590477440.0,
      "grad_norm": 2.1172448537654196,
      "language_loss": 0.70855886,
      "learning_rate": 2.4274483872219863e-06,
      "loss": 0.73062837,
      "num_input_tokens_seen": 80277270,
      "step": 3730,
      "time_per_iteration": 2.428694009780884
    },
    {
      "auxiliary_loss_clip": 0.01166611,
      "auxiliary_loss_mlp": 0.01028713,
      "balance_loss_clip": 1.05191278,
      "balance_loss_mlp": 1.02068686,
      "epoch": 0.4486262249744484,
      "flos": 20047742853120.0,
      "grad_norm": 1.7888091077398294,
      "language_loss": 0.93992376,
      "learning_rate": 2.426687382058386e-06,
      "loss": 0.96187705,
      "num_input_tokens_seen": 80295550,
      "step": 3731,
      "time_per_iteration": 3.2148828506469727
    },
    {
      "auxiliary_loss_clip": 0.0107036,
      "auxiliary_loss_mlp": 0.01004385,
      "balance_loss_clip": 1.02395248,
      "balance_loss_mlp": 1.00322914,
      "epoch": 0.4487464678650875,
      "flos": 64595684776320.0,
      "grad_norm": 0.8617764825105491,
      "language_loss": 0.59841961,
      "learning_rate": 2.425926312166649e-06,
      "loss": 0.61916709,
      "num_input_tokens_seen": 80348425,
      "step": 3732,
      "time_per_iteration": 2.925603151321411
    },
    {
      "auxiliary_loss_clip": 0.01161711,
      "auxiliary_loss_mlp": 0.01025832,
      "balance_loss_clip": 1.05454564,
      "balance_loss_mlp": 1.01705253,
      "epoch": 0.4488667107557266,
      "flos": 20769798049920.0,
      "grad_norm": 2.0758542749481492,
      "language_loss": 0.72705811,
      "learning_rate": 2.42516517766223e-06,
      "loss": 0.74893355,
      "num_input_tokens_seen": 80366505,
      "step": 3733,
      "time_per_iteration": 2.46803617477417
    },
    {
      "auxiliary_loss_clip": 0.01184597,
      "auxiliary_loss_mlp": 0.01027602,
      "balance_loss_clip": 1.05712795,
      "balance_loss_mlp": 1.01927507,
      "epoch": 0.44898695364636565,
      "flos": 23951735326080.0,
      "grad_norm": 2.402057180054235,
      "language_loss": 0.68073934,
      "learning_rate": 2.4244039786605907e-06,
      "loss": 0.70286131,
      "num_input_tokens_seen": 80387510,
      "step": 3734,
      "time_per_iteration": 2.461834669113159
    },
    {
      "auxiliary_loss_clip": 0.01126024,
      "auxiliary_loss_mlp": 0.01027005,
      "balance_loss_clip": 1.04622388,
      "balance_loss_mlp": 1.01857138,
      "epoch": 0.44910719653700476,
      "flos": 18624351628800.0,
      "grad_norm": 2.326495437943824,
      "language_loss": 0.82343304,
      "learning_rate": 2.4236427152772055e-06,
      "loss": 0.84496337,
      "num_input_tokens_seen": 80405915,
      "step": 3735,
      "time_per_iteration": 2.5284368991851807
    },
    {
      "auxiliary_loss_clip": 0.01036357,
      "auxiliary_loss_mlp": 0.01002037,
      "balance_loss_clip": 1.01779127,
      "balance_loss_mlp": 1.00080955,
      "epoch": 0.4492274394276438,
      "flos": 57033435749760.0,
      "grad_norm": 0.828588795770772,
      "language_loss": 0.57338703,
      "learning_rate": 2.422881387627557e-06,
      "loss": 0.59377098,
      "num_input_tokens_seen": 80458365,
      "step": 3736,
      "time_per_iteration": 2.827345371246338
    },
    {
      "auxiliary_loss_clip": 0.01159183,
      "auxiliary_loss_mlp": 0.01024103,
      "balance_loss_clip": 1.05367064,
      "balance_loss_mlp": 1.01621163,
      "epoch": 0.4493476823182829,
      "flos": 23254888498560.0,
      "grad_norm": 1.5475513603628974,
      "language_loss": 0.7744534,
      "learning_rate": 2.422119995827139e-06,
      "loss": 0.79628628,
      "num_input_tokens_seen": 80478490,
      "step": 3737,
      "time_per_iteration": 2.501750946044922
    },
    {
      "auxiliary_loss_clip": 0.01173609,
      "auxiliary_loss_mlp": 0.01028821,
      "balance_loss_clip": 1.05444133,
      "balance_loss_mlp": 1.02088773,
      "epoch": 0.44946792520892204,
      "flos": 15815131827840.0,
      "grad_norm": 2.839449450650751,
      "language_loss": 0.74221724,
      "learning_rate": 2.4213585399914528e-06,
      "loss": 0.76424158,
      "num_input_tokens_seen": 80495695,
      "step": 3738,
      "time_per_iteration": 2.4529948234558105
    },
    {
      "auxiliary_loss_clip": 0.01171071,
      "auxiliary_loss_mlp": 0.01028793,
      "balance_loss_clip": 1.05458117,
      "balance_loss_mlp": 1.02071047,
      "epoch": 0.4495881680995611,
      "flos": 19610063631360.0,
      "grad_norm": 1.738964745627871,
      "language_loss": 0.85120904,
      "learning_rate": 2.4205970202360113e-06,
      "loss": 0.87320769,
      "num_input_tokens_seen": 80515260,
      "step": 3739,
      "time_per_iteration": 2.488502025604248
    },
    {
      "auxiliary_loss_clip": 0.01118295,
      "auxiliary_loss_mlp": 0.01027346,
      "balance_loss_clip": 1.04790878,
      "balance_loss_mlp": 1.01862633,
      "epoch": 0.4497084109902002,
      "flos": 26031465815040.0,
      "grad_norm": 1.8509698700314674,
      "language_loss": 0.77969623,
      "learning_rate": 2.4198354366763354e-06,
      "loss": 0.80115259,
      "num_input_tokens_seen": 80533900,
      "step": 3740,
      "time_per_iteration": 2.5950775146484375
    },
    {
      "auxiliary_loss_clip": 0.01160249,
      "auxiliary_loss_mlp": 0.01026637,
      "balance_loss_clip": 1.05328619,
      "balance_loss_mlp": 1.01825058,
      "epoch": 0.4498286538808393,
      "flos": 14793688771200.0,
      "grad_norm": 2.2453774512902474,
      "language_loss": 0.78749019,
      "learning_rate": 2.4190737894279587e-06,
      "loss": 0.80935907,
      "num_input_tokens_seen": 80551270,
      "step": 3741,
      "time_per_iteration": 2.5710997581481934
    },
    {
      "auxiliary_loss_clip": 0.01130927,
      "auxiliary_loss_mlp": 0.01027482,
      "balance_loss_clip": 1.04537284,
      "balance_loss_mlp": 1.01971555,
      "epoch": 0.44994889677147837,
      "flos": 15450171690240.0,
      "grad_norm": 2.0451407789837037,
      "language_loss": 0.80230498,
      "learning_rate": 2.4183120786064203e-06,
      "loss": 0.82388908,
      "num_input_tokens_seen": 80568145,
      "step": 3742,
      "time_per_iteration": 2.496428966522217
    },
    {
      "auxiliary_loss_clip": 0.01170937,
      "auxiliary_loss_mlp": 0.00762754,
      "balance_loss_clip": 1.05686092,
      "balance_loss_mlp": 1.00051403,
      "epoch": 0.4500691396621175,
      "flos": 21798316085760.0,
      "grad_norm": 2.2590118283325,
      "language_loss": 0.85631835,
      "learning_rate": 2.417550304327273e-06,
      "loss": 0.87565523,
      "num_input_tokens_seen": 80586185,
      "step": 3743,
      "time_per_iteration": 2.535747766494751
    },
    {
      "auxiliary_loss_clip": 0.01185947,
      "auxiliary_loss_mlp": 0.01033036,
      "balance_loss_clip": 1.05553615,
      "balance_loss_mlp": 1.02451897,
      "epoch": 0.4501893825527566,
      "flos": 32382016421760.0,
      "grad_norm": 1.6238272100642819,
      "language_loss": 0.75857115,
      "learning_rate": 2.4167884667060763e-06,
      "loss": 0.780761,
      "num_input_tokens_seen": 80608895,
      "step": 3744,
      "time_per_iteration": 2.560973882675171
    },
    {
      "auxiliary_loss_clip": 0.011562,
      "auxiliary_loss_mlp": 0.01031031,
      "balance_loss_clip": 1.05149508,
      "balance_loss_mlp": 1.02273417,
      "epoch": 0.45030962544339564,
      "flos": 16544944362240.0,
      "grad_norm": 2.0841093455484585,
      "language_loss": 0.87233818,
      "learning_rate": 2.4160265658584e-06,
      "loss": 0.89421046,
      "num_input_tokens_seen": 80623785,
      "step": 3745,
      "time_per_iteration": 2.479842185974121
    },
    {
      "auxiliary_loss_clip": 0.01175097,
      "auxiliary_loss_mlp": 0.01026984,
      "balance_loss_clip": 1.05516124,
      "balance_loss_mlp": 1.01866317,
      "epoch": 0.45042986833403476,
      "flos": 19573039687680.0,
      "grad_norm": 1.9375556610580837,
      "language_loss": 0.68178022,
      "learning_rate": 2.4152646018998253e-06,
      "loss": 0.7038011,
      "num_input_tokens_seen": 80642735,
      "step": 3746,
      "time_per_iteration": 2.4767673015594482
    },
    {
      "auxiliary_loss_clip": 0.01167417,
      "auxiliary_loss_mlp": 0.01036901,
      "balance_loss_clip": 1.05326366,
      "balance_loss_mlp": 1.02888441,
      "epoch": 0.45055011122467387,
      "flos": 23112467072640.0,
      "grad_norm": 2.4059667216935856,
      "language_loss": 0.71638024,
      "learning_rate": 2.4145025749459403e-06,
      "loss": 0.73842341,
      "num_input_tokens_seen": 80663760,
      "step": 3747,
      "time_per_iteration": 2.540201187133789
    },
    {
      "auxiliary_loss_clip": 0.01100442,
      "auxiliary_loss_mlp": 0.01037028,
      "balance_loss_clip": 1.04926336,
      "balance_loss_mlp": 1.02842689,
      "epoch": 0.4506703541153129,
      "flos": 19934623946880.0,
      "grad_norm": 3.1361848550496743,
      "language_loss": 0.70132393,
      "learning_rate": 2.413740485112344e-06,
      "loss": 0.72269857,
      "num_input_tokens_seen": 80682100,
      "step": 3748,
      "time_per_iteration": 2.6768319606781006
    },
    {
      "auxiliary_loss_clip": 0.01149807,
      "auxiliary_loss_mlp": 0.0102574,
      "balance_loss_clip": 1.05397427,
      "balance_loss_mlp": 1.01729381,
      "epoch": 0.45079059700595203,
      "flos": 19499530504320.0,
      "grad_norm": 1.6708286229857756,
      "language_loss": 0.82146597,
      "learning_rate": 2.412978332514646e-06,
      "loss": 0.84322149,
      "num_input_tokens_seen": 80700880,
      "step": 3749,
      "time_per_iteration": 3.3277199268341064
    },
    {
      "auxiliary_loss_clip": 0.01160279,
      "auxiliary_loss_mlp": 0.01025385,
      "balance_loss_clip": 1.05437994,
      "balance_loss_mlp": 1.01671219,
      "epoch": 0.4509108398965911,
      "flos": 27636313570560.0,
      "grad_norm": 2.5039223583520607,
      "language_loss": 0.71925116,
      "learning_rate": 2.4122161172684623e-06,
      "loss": 0.74110776,
      "num_input_tokens_seen": 80721675,
      "step": 3750,
      "time_per_iteration": 2.653538942337036
    },
    {
      "auxiliary_loss_clip": 0.01159929,
      "auxiliary_loss_mlp": 0.0103592,
      "balance_loss_clip": 1.05353808,
      "balance_loss_mlp": 1.02708101,
      "epoch": 0.4510310827872302,
      "flos": 20995712640000.0,
      "grad_norm": 4.125961595740625,
      "language_loss": 0.83858192,
      "learning_rate": 2.4114538394894216e-06,
      "loss": 0.86054045,
      "num_input_tokens_seen": 80739315,
      "step": 3751,
      "time_per_iteration": 2.5357472896575928
    },
    {
      "auxiliary_loss_clip": 0.01152362,
      "auxiliary_loss_mlp": 0.01025031,
      "balance_loss_clip": 1.04778242,
      "balance_loss_mlp": 1.01704431,
      "epoch": 0.4511513256778693,
      "flos": 16216684945920.0,
      "grad_norm": 1.9909336664787296,
      "language_loss": 0.83124518,
      "learning_rate": 2.410691499293161e-06,
      "loss": 0.85301912,
      "num_input_tokens_seen": 80757470,
      "step": 3752,
      "time_per_iteration": 3.3600518703460693
    },
    {
      "auxiliary_loss_clip": 0.0116877,
      "auxiliary_loss_mlp": 0.01026002,
      "balance_loss_clip": 1.05273366,
      "balance_loss_mlp": 1.01769948,
      "epoch": 0.45127156856850836,
      "flos": 25186702780800.0,
      "grad_norm": 1.586291049664566,
      "language_loss": 0.74620748,
      "learning_rate": 2.409929096795326e-06,
      "loss": 0.76815522,
      "num_input_tokens_seen": 80777840,
      "step": 3753,
      "time_per_iteration": 2.5574533939361572
    },
    {
      "auxiliary_loss_clip": 0.01170796,
      "auxiliary_loss_mlp": 0.0102956,
      "balance_loss_clip": 1.05291188,
      "balance_loss_mlp": 1.02073836,
      "epoch": 0.4513918114591475,
      "flos": 20412523422720.0,
      "grad_norm": 1.8514644660422697,
      "language_loss": 0.79435682,
      "learning_rate": 2.409166632111573e-06,
      "loss": 0.81636047,
      "num_input_tokens_seen": 80795975,
      "step": 3754,
      "time_per_iteration": 2.5205471515655518
    },
    {
      "auxiliary_loss_clip": 0.01178116,
      "auxiliary_loss_mlp": 0.01026187,
      "balance_loss_clip": 1.05422628,
      "balance_loss_mlp": 1.01752615,
      "epoch": 0.4515120543497866,
      "flos": 26648482665600.0,
      "grad_norm": 1.8717835396622062,
      "language_loss": 0.80601835,
      "learning_rate": 2.4084041053575674e-06,
      "loss": 0.82806146,
      "num_input_tokens_seen": 80815395,
      "step": 3755,
      "time_per_iteration": 3.294748544692993
    },
    {
      "auxiliary_loss_clip": 0.0116126,
      "auxiliary_loss_mlp": 0.01025496,
      "balance_loss_clip": 1.05414999,
      "balance_loss_mlp": 1.01704431,
      "epoch": 0.45163229724042564,
      "flos": 20595093275520.0,
      "grad_norm": 2.241044103456221,
      "language_loss": 0.7234416,
      "learning_rate": 2.4076415166489834e-06,
      "loss": 0.74530923,
      "num_input_tokens_seen": 80834805,
      "step": 3756,
      "time_per_iteration": 2.4837145805358887
    },
    {
      "auxiliary_loss_clip": 0.01133074,
      "auxiliary_loss_mlp": 0.0103185,
      "balance_loss_clip": 1.05036211,
      "balance_loss_mlp": 1.02398205,
      "epoch": 0.45175254013106475,
      "flos": 21689004021120.0,
      "grad_norm": 2.168584034022373,
      "language_loss": 0.78813004,
      "learning_rate": 2.406878866101506e-06,
      "loss": 0.80977929,
      "num_input_tokens_seen": 80853770,
      "step": 3757,
      "time_per_iteration": 2.6080996990203857
    },
    {
      "auxiliary_loss_clip": 0.0118539,
      "auxiliary_loss_mlp": 0.01028538,
      "balance_loss_clip": 1.05763328,
      "balance_loss_mlp": 1.02081299,
      "epoch": 0.45187278302170386,
      "flos": 18878850466560.0,
      "grad_norm": 1.9961197332170317,
      "language_loss": 0.78091234,
      "learning_rate": 2.4061161538308273e-06,
      "loss": 0.80305159,
      "num_input_tokens_seen": 80870615,
      "step": 3758,
      "time_per_iteration": 3.177151679992676
    },
    {
      "auxiliary_loss_clip": 0.0116982,
      "auxiliary_loss_mlp": 0.01027524,
      "balance_loss_clip": 1.05452859,
      "balance_loss_mlp": 1.01926851,
      "epoch": 0.4519930259123429,
      "flos": 18582479349120.0,
      "grad_norm": 1.8890221530485913,
      "language_loss": 0.88957685,
      "learning_rate": 2.4053533799526523e-06,
      "loss": 0.91155028,
      "num_input_tokens_seen": 80886335,
      "step": 3759,
      "time_per_iteration": 2.453939199447632
    },
    {
      "auxiliary_loss_clip": 0.01150136,
      "auxiliary_loss_mlp": 0.01031065,
      "balance_loss_clip": 1.05201554,
      "balance_loss_mlp": 1.02288747,
      "epoch": 0.452113268802982,
      "flos": 25192377129600.0,
      "grad_norm": 1.6118747846566377,
      "language_loss": 0.8601687,
      "learning_rate": 2.404590544582691e-06,
      "loss": 0.88198078,
      "num_input_tokens_seen": 80904570,
      "step": 3760,
      "time_per_iteration": 2.5175836086273193
    },
    {
      "auxiliary_loss_clip": 0.01129549,
      "auxiliary_loss_mlp": 0.0103248,
      "balance_loss_clip": 1.04423451,
      "balance_loss_mlp": 1.02413559,
      "epoch": 0.45223351169362114,
      "flos": 39378922312320.0,
      "grad_norm": 1.7108712405252762,
      "language_loss": 0.80917883,
      "learning_rate": 2.403827647836666e-06,
      "loss": 0.8307991,
      "num_input_tokens_seen": 80925125,
      "step": 3761,
      "time_per_iteration": 2.7732365131378174
    },
    {
      "auxiliary_loss_clip": 0.01185998,
      "auxiliary_loss_mlp": 0.01029848,
      "balance_loss_clip": 1.05456853,
      "balance_loss_mlp": 1.02126551,
      "epoch": 0.4523537545842602,
      "flos": 21582169994880.0,
      "grad_norm": 1.8969500765232759,
      "language_loss": 0.69223797,
      "learning_rate": 2.4030646898303075e-06,
      "loss": 0.71439648,
      "num_input_tokens_seen": 80946615,
      "step": 3762,
      "time_per_iteration": 2.5373876094818115
    },
    {
      "auxiliary_loss_clip": 0.01161509,
      "auxiliary_loss_mlp": 0.0103565,
      "balance_loss_clip": 1.05302978,
      "balance_loss_mlp": 1.02747226,
      "epoch": 0.4524739974748993,
      "flos": 28439527547520.0,
      "grad_norm": 2.2145640261037527,
      "language_loss": 0.81676447,
      "learning_rate": 2.4023016706793566e-06,
      "loss": 0.83873606,
      "num_input_tokens_seen": 80966410,
      "step": 3763,
      "time_per_iteration": 2.604886531829834
    },
    {
      "auxiliary_loss_clip": 0.01056349,
      "auxiliary_loss_mlp": 0.01003428,
      "balance_loss_clip": 1.01989651,
      "balance_loss_mlp": 1.00230718,
      "epoch": 0.4525942403655384,
      "flos": 61556492148480.0,
      "grad_norm": 0.7605593372196167,
      "language_loss": 0.56882799,
      "learning_rate": 2.401538590499561e-06,
      "loss": 0.58942574,
      "num_input_tokens_seen": 81026865,
      "step": 3764,
      "time_per_iteration": 3.249357223510742
    },
    {
      "auxiliary_loss_clip": 0.01173899,
      "auxiliary_loss_mlp": 0.00763177,
      "balance_loss_clip": 1.05513811,
      "balance_loss_mlp": 1.00044131,
      "epoch": 0.45271448325617747,
      "flos": 27529838680320.0,
      "grad_norm": 2.130759408348567,
      "language_loss": 0.71776503,
      "learning_rate": 2.400775449406682e-06,
      "loss": 0.73713577,
      "num_input_tokens_seen": 81050060,
      "step": 3765,
      "time_per_iteration": 2.6272900104522705
    },
    {
      "auxiliary_loss_clip": 0.01169285,
      "auxiliary_loss_mlp": 0.01030882,
      "balance_loss_clip": 1.0516479,
      "balance_loss_mlp": 1.02304423,
      "epoch": 0.4528347261468166,
      "flos": 22452608275200.0,
      "grad_norm": 1.7924407815249894,
      "language_loss": 0.72836196,
      "learning_rate": 2.400012247516485e-06,
      "loss": 0.75036359,
      "num_input_tokens_seen": 81070625,
      "step": 3766,
      "time_per_iteration": 2.55890154838562
    },
    {
      "auxiliary_loss_clip": 0.01144676,
      "auxiliary_loss_mlp": 0.01028603,
      "balance_loss_clip": 1.04866433,
      "balance_loss_mlp": 1.02052093,
      "epoch": 0.45295496903745563,
      "flos": 21103875469440.0,
      "grad_norm": 1.8522468763111193,
      "language_loss": 0.90215921,
      "learning_rate": 2.3992489849447484e-06,
      "loss": 0.92389202,
      "num_input_tokens_seen": 81089080,
      "step": 3767,
      "time_per_iteration": 2.5342369079589844
    },
    {
      "auxiliary_loss_clip": 0.01146481,
      "auxiliary_loss_mlp": 0.01028071,
      "balance_loss_clip": 1.04963183,
      "balance_loss_mlp": 1.01989055,
      "epoch": 0.45307521192809475,
      "flos": 23221168606080.0,
      "grad_norm": 1.593612532900772,
      "language_loss": 0.78877389,
      "learning_rate": 2.3984856618072584e-06,
      "loss": 0.8105194,
      "num_input_tokens_seen": 81109115,
      "step": 3768,
      "time_per_iteration": 2.526691198348999
    },
    {
      "auxiliary_loss_clip": 0.01147603,
      "auxiliary_loss_mlp": 0.01032843,
      "balance_loss_clip": 1.05243802,
      "balance_loss_mlp": 1.02458751,
      "epoch": 0.45319545481873386,
      "flos": 15560094286080.0,
      "grad_norm": 1.9856900933397736,
      "language_loss": 0.73791873,
      "learning_rate": 2.3977222782198098e-06,
      "loss": 0.75972319,
      "num_input_tokens_seen": 81127750,
      "step": 3769,
      "time_per_iteration": 2.4951794147491455
    },
    {
      "auxiliary_loss_clip": 0.01133546,
      "auxiliary_loss_mlp": 0.010372,
      "balance_loss_clip": 1.04953289,
      "balance_loss_mlp": 1.02744913,
      "epoch": 0.4533156977093729,
      "flos": 21944759834880.0,
      "grad_norm": 1.5617537540052147,
      "language_loss": 0.75110316,
      "learning_rate": 2.3969588342982077e-06,
      "loss": 0.77281058,
      "num_input_tokens_seen": 81147125,
      "step": 3770,
      "time_per_iteration": 2.5324506759643555
    },
    {
      "auxiliary_loss_clip": 0.01168656,
      "auxiliary_loss_mlp": 0.01029871,
      "balance_loss_clip": 1.05515385,
      "balance_loss_mlp": 1.02136564,
      "epoch": 0.453435940600012,
      "flos": 24242180699520.0,
      "grad_norm": 1.5061330962328552,
      "language_loss": 0.72658408,
      "learning_rate": 2.396195330158267e-06,
      "loss": 0.74856937,
      "num_input_tokens_seen": 81167015,
      "step": 3771,
      "time_per_iteration": 2.49704909324646
    },
    {
      "auxiliary_loss_clip": 0.01184709,
      "auxiliary_loss_mlp": 0.01027536,
      "balance_loss_clip": 1.05495811,
      "balance_loss_mlp": 1.01910233,
      "epoch": 0.45355618349065113,
      "flos": 23440367352960.0,
      "grad_norm": 1.6790523680169442,
      "language_loss": 0.79651469,
      "learning_rate": 2.3954317659158094e-06,
      "loss": 0.81863713,
      "num_input_tokens_seen": 81187350,
      "step": 3772,
      "time_per_iteration": 2.465524435043335
    },
    {
      "auxiliary_loss_clip": 0.01080716,
      "auxiliary_loss_mlp": 0.01000878,
      "balance_loss_clip": 1.01787233,
      "balance_loss_mlp": 0.99976921,
      "epoch": 0.4536764263812902,
      "flos": 66903161448960.0,
      "grad_norm": 0.8945363325813748,
      "language_loss": 0.56934851,
      "learning_rate": 2.394668141686667e-06,
      "loss": 0.59016442,
      "num_input_tokens_seen": 81249315,
      "step": 3773,
      "time_per_iteration": 3.051337718963623
    },
    {
      "auxiliary_loss_clip": 0.0116444,
      "auxiliary_loss_mlp": 0.01028266,
      "balance_loss_clip": 1.0500102,
      "balance_loss_mlp": 1.0203923,
      "epoch": 0.4537966692719293,
      "flos": 42739766254080.0,
      "grad_norm": 2.0690370994660694,
      "language_loss": 0.69371283,
      "learning_rate": 2.3939044575866813e-06,
      "loss": 0.71563989,
      "num_input_tokens_seen": 81272065,
      "step": 3774,
      "time_per_iteration": 2.6522011756896973
    },
    {
      "auxiliary_loss_clip": 0.01150021,
      "auxiliary_loss_mlp": 0.0076308,
      "balance_loss_clip": 1.04903984,
      "balance_loss_mlp": 1.0004549,
      "epoch": 0.4539169121625684,
      "flos": 35549480517120.0,
      "grad_norm": 2.1064599192742692,
      "language_loss": 0.75257522,
      "learning_rate": 2.3931407137317024e-06,
      "loss": 0.77170622,
      "num_input_tokens_seen": 81292220,
      "step": 3775,
      "time_per_iteration": 2.6045687198638916
    },
    {
      "auxiliary_loss_clip": 0.01138393,
      "auxiliary_loss_mlp": 0.01032629,
      "balance_loss_clip": 1.04696679,
      "balance_loss_mlp": 1.0239501,
      "epoch": 0.45403715505320746,
      "flos": 18514716341760.0,
      "grad_norm": 1.6599063037356312,
      "language_loss": 0.85206449,
      "learning_rate": 2.3923769102375907e-06,
      "loss": 0.87377477,
      "num_input_tokens_seen": 81311085,
      "step": 3776,
      "time_per_iteration": 3.357940912246704
    },
    {
      "auxiliary_loss_clip": 0.01141734,
      "auxiliary_loss_mlp": 0.01034878,
      "balance_loss_clip": 1.04974818,
      "balance_loss_mlp": 1.0260148,
      "epoch": 0.4541573979438466,
      "flos": 25045825639680.0,
      "grad_norm": 2.5178914633459057,
      "language_loss": 0.78348935,
      "learning_rate": 2.391613047220213e-06,
      "loss": 0.80525553,
      "num_input_tokens_seen": 81330985,
      "step": 3777,
      "time_per_iteration": 2.5558977127075195
    },
    {
      "auxiliary_loss_clip": 0.01133966,
      "auxiliary_loss_mlp": 0.01026218,
      "balance_loss_clip": 1.04943693,
      "balance_loss_mlp": 1.01796913,
      "epoch": 0.4542776408344857,
      "flos": 18332397884160.0,
      "grad_norm": 2.0353926887114646,
      "language_loss": 0.79293275,
      "learning_rate": 2.390849124795447e-06,
      "loss": 0.81453454,
      "num_input_tokens_seen": 81346985,
      "step": 3778,
      "time_per_iteration": 2.5255467891693115
    },
    {
      "auxiliary_loss_clip": 0.01185832,
      "auxiliary_loss_mlp": 0.01027391,
      "balance_loss_clip": 1.05525661,
      "balance_loss_mlp": 1.01935673,
      "epoch": 0.45439788372512474,
      "flos": 20701173116160.0,
      "grad_norm": 2.0985837238693246,
      "language_loss": 0.84202588,
      "learning_rate": 2.3900851430791804e-06,
      "loss": 0.86415815,
      "num_input_tokens_seen": 81365005,
      "step": 3779,
      "time_per_iteration": 3.2632970809936523
    },
    {
      "auxiliary_loss_clip": 0.01187104,
      "auxiliary_loss_mlp": 0.01032497,
      "balance_loss_clip": 1.05380249,
      "balance_loss_mlp": 1.0232048,
      "epoch": 0.45451812661576385,
      "flos": 22309432663680.0,
      "grad_norm": 2.3253835641828347,
      "language_loss": 0.84480727,
      "learning_rate": 2.389321102187307e-06,
      "loss": 0.86700326,
      "num_input_tokens_seen": 81383785,
      "step": 3780,
      "time_per_iteration": 2.4269118309020996
    },
    {
      "auxiliary_loss_clip": 0.0115723,
      "auxiliary_loss_mlp": 0.00763807,
      "balance_loss_clip": 1.05281854,
      "balance_loss_mlp": 1.00042903,
      "epoch": 0.4546383695064029,
      "flos": 21763303303680.0,
      "grad_norm": 1.8859639538980257,
      "language_loss": 0.8174938,
      "learning_rate": 2.3885570022357326e-06,
      "loss": 0.83670413,
      "num_input_tokens_seen": 81402915,
      "step": 3781,
      "time_per_iteration": 3.2148799896240234
    },
    {
      "auxiliary_loss_clip": 0.01051458,
      "auxiliary_loss_mlp": 0.01005437,
      "balance_loss_clip": 1.01503432,
      "balance_loss_mlp": 1.00407243,
      "epoch": 0.454758612397042,
      "flos": 64242755694720.0,
      "grad_norm": 0.8152794847847367,
      "language_loss": 0.60924393,
      "learning_rate": 2.38779284334037e-06,
      "loss": 0.6298129,
      "num_input_tokens_seen": 81467890,
      "step": 3782,
      "time_per_iteration": 3.1294891834259033
    },
    {
      "auxiliary_loss_clip": 0.01115241,
      "auxiliary_loss_mlp": 0.01030779,
      "balance_loss_clip": 1.04538035,
      "balance_loss_mlp": 1.02230966,
      "epoch": 0.4548788552876811,
      "flos": 27304175485440.0,
      "grad_norm": 1.92047963167719,
      "language_loss": 0.78968054,
      "learning_rate": 2.387028625617141e-06,
      "loss": 0.81114072,
      "num_input_tokens_seen": 81487105,
      "step": 3783,
      "time_per_iteration": 2.592634677886963
    },
    {
      "auxiliary_loss_clip": 0.01142262,
      "auxiliary_loss_mlp": 0.01028999,
      "balance_loss_clip": 1.04887736,
      "balance_loss_mlp": 1.02086866,
      "epoch": 0.4549990981783202,
      "flos": 22857142222080.0,
      "grad_norm": 1.8124798147360146,
      "language_loss": 0.84771532,
      "learning_rate": 2.3862643491819766e-06,
      "loss": 0.86942792,
      "num_input_tokens_seen": 81505670,
      "step": 3784,
      "time_per_iteration": 3.275575876235962
    },
    {
      "auxiliary_loss_clip": 0.01165969,
      "auxiliary_loss_mlp": 0.01028451,
      "balance_loss_clip": 1.04948103,
      "balance_loss_mlp": 1.02048779,
      "epoch": 0.4551193410689593,
      "flos": 23258587599360.0,
      "grad_norm": 1.7297506885774971,
      "language_loss": 0.84381652,
      "learning_rate": 2.3855000141508186e-06,
      "loss": 0.86576068,
      "num_input_tokens_seen": 81525825,
      "step": 3785,
      "time_per_iteration": 2.4763951301574707
    },
    {
      "auxiliary_loss_clip": 0.01161859,
      "auxiliary_loss_mlp": 0.01031976,
      "balance_loss_clip": 1.05582595,
      "balance_loss_mlp": 1.02308285,
      "epoch": 0.4552395839595984,
      "flos": 20777519473920.0,
      "grad_norm": 2.094454743344971,
      "language_loss": 0.83995044,
      "learning_rate": 2.3847356206396143e-06,
      "loss": 0.86188877,
      "num_input_tokens_seen": 81543135,
      "step": 3786,
      "time_per_iteration": 2.498401165008545
    },
    {
      "auxiliary_loss_clip": 0.0118489,
      "auxiliary_loss_mlp": 0.01027389,
      "balance_loss_clip": 1.0552392,
      "balance_loss_mlp": 1.01892543,
      "epoch": 0.45535982685023746,
      "flos": 23257510191360.0,
      "grad_norm": 1.5030745684404934,
      "language_loss": 0.78684652,
      "learning_rate": 2.3839711687643227e-06,
      "loss": 0.80896932,
      "num_input_tokens_seen": 81564360,
      "step": 3787,
      "time_per_iteration": 2.4950311183929443
    },
    {
      "auxiliary_loss_clip": 0.01171028,
      "auxiliary_loss_mlp": 0.01030485,
      "balance_loss_clip": 1.05355573,
      "balance_loss_mlp": 1.02124071,
      "epoch": 0.45548006974087657,
      "flos": 19646117907840.0,
      "grad_norm": 1.9704479939940027,
      "language_loss": 0.73939848,
      "learning_rate": 2.38320665864091e-06,
      "loss": 0.76141357,
      "num_input_tokens_seen": 81583710,
      "step": 3788,
      "time_per_iteration": 2.4467294216156006
    },
    {
      "auxiliary_loss_clip": 0.01114179,
      "auxiliary_loss_mlp": 0.01027913,
      "balance_loss_clip": 1.04512429,
      "balance_loss_mlp": 1.01945496,
      "epoch": 0.4556003126315157,
      "flos": 20047778766720.0,
      "grad_norm": 1.7535612715074247,
      "language_loss": 0.82016379,
      "learning_rate": 2.3824420903853516e-06,
      "loss": 0.84158474,
      "num_input_tokens_seen": 81602175,
      "step": 3789,
      "time_per_iteration": 2.6008527278900146
    },
    {
      "auxiliary_loss_clip": 0.01170612,
      "auxiliary_loss_mlp": 0.01028043,
      "balance_loss_clip": 1.0553441,
      "balance_loss_mlp": 1.01972854,
      "epoch": 0.45572055552215474,
      "flos": 22959738443520.0,
      "grad_norm": 2.1186014518042464,
      "language_loss": 0.81559336,
      "learning_rate": 2.3816774641136324e-06,
      "loss": 0.83757997,
      "num_input_tokens_seen": 81619430,
      "step": 3790,
      "time_per_iteration": 2.4550976753234863
    },
    {
      "auxiliary_loss_clip": 0.01168393,
      "auxiliary_loss_mlp": 0.00763052,
      "balance_loss_clip": 1.05338836,
      "balance_loss_mlp": 1.00040102,
      "epoch": 0.45584079841279385,
      "flos": 33109925535360.0,
      "grad_norm": 1.8800406877513136,
      "language_loss": 0.71113884,
      "learning_rate": 2.380912779941745e-06,
      "loss": 0.73045325,
      "num_input_tokens_seen": 81642550,
      "step": 3791,
      "time_per_iteration": 2.550929546356201
    },
    {
      "auxiliary_loss_clip": 0.01171577,
      "auxiliary_loss_mlp": 0.01037613,
      "balance_loss_clip": 1.05014384,
      "balance_loss_mlp": 1.0277189,
      "epoch": 0.45596104130343296,
      "flos": 27272179445760.0,
      "grad_norm": 1.8494476814431424,
      "language_loss": 0.82886106,
      "learning_rate": 2.3801480379856918e-06,
      "loss": 0.85095298,
      "num_input_tokens_seen": 81664260,
      "step": 3792,
      "time_per_iteration": 2.5273876190185547
    },
    {
      "auxiliary_loss_clip": 0.01158897,
      "auxiliary_loss_mlp": 0.01033136,
      "balance_loss_clip": 1.05326509,
      "balance_loss_mlp": 1.02513969,
      "epoch": 0.456081284194072,
      "flos": 21579799697280.0,
      "grad_norm": 1.7286153991026363,
      "language_loss": 0.83240891,
      "learning_rate": 2.379383238361484e-06,
      "loss": 0.85432923,
      "num_input_tokens_seen": 81683620,
      "step": 3793,
      "time_per_iteration": 2.487133741378784
    },
    {
      "auxiliary_loss_clip": 0.01166915,
      "auxiliary_loss_mlp": 0.01029734,
      "balance_loss_clip": 1.05071235,
      "balance_loss_mlp": 1.02147329,
      "epoch": 0.4562015270847111,
      "flos": 35918822113920.0,
      "grad_norm": 2.4842672936168753,
      "language_loss": 0.79355621,
      "learning_rate": 2.3786183811851407e-06,
      "loss": 0.81552267,
      "num_input_tokens_seen": 81704325,
      "step": 3794,
      "time_per_iteration": 2.5594379901885986
    },
    {
      "auxiliary_loss_clip": 0.01186374,
      "auxiliary_loss_mlp": 0.01028778,
      "balance_loss_clip": 1.05726063,
      "balance_loss_mlp": 1.02073193,
      "epoch": 0.45632176997535023,
      "flos": 13589783602560.0,
      "grad_norm": 1.6789033066313022,
      "language_loss": 0.80015802,
      "learning_rate": 2.3778534665726892e-06,
      "loss": 0.82230949,
      "num_input_tokens_seen": 81721155,
      "step": 3795,
      "time_per_iteration": 2.379777669906616
    },
    {
      "auxiliary_loss_clip": 0.01159261,
      "auxiliary_loss_mlp": 0.0103411,
      "balance_loss_clip": 1.05166936,
      "balance_loss_mlp": 1.0261054,
      "epoch": 0.4564420128659893,
      "flos": 32635401937920.0,
      "grad_norm": 2.786734573177213,
      "language_loss": 0.72400242,
      "learning_rate": 2.377088494640168e-06,
      "loss": 0.74593616,
      "num_input_tokens_seen": 81742905,
      "step": 3796,
      "time_per_iteration": 2.5297863483428955
    },
    {
      "auxiliary_loss_clip": 0.01164883,
      "auxiliary_loss_mlp": 0.0103014,
      "balance_loss_clip": 1.05331361,
      "balance_loss_mlp": 1.02206349,
      "epoch": 0.4565622557566284,
      "flos": 20377690208640.0,
      "grad_norm": 4.242723131690308,
      "language_loss": 0.78136063,
      "learning_rate": 2.3763234655036216e-06,
      "loss": 0.80331087,
      "num_input_tokens_seen": 81762105,
      "step": 3797,
      "time_per_iteration": 2.4528684616088867
    },
    {
      "auxiliary_loss_clip": 0.01137969,
      "auxiliary_loss_mlp": 0.01028854,
      "balance_loss_clip": 1.04526961,
      "balance_loss_mlp": 1.02058685,
      "epoch": 0.45668249864726745,
      "flos": 25374372364800.0,
      "grad_norm": 2.0709775984358423,
      "language_loss": 0.86847389,
      "learning_rate": 2.3755583792791046e-06,
      "loss": 0.89014214,
      "num_input_tokens_seen": 81781975,
      "step": 3798,
      "time_per_iteration": 2.539468288421631
    },
    {
      "auxiliary_loss_clip": 0.01168977,
      "auxiliary_loss_mlp": 0.01024162,
      "balance_loss_clip": 1.05152678,
      "balance_loss_mlp": 1.01635385,
      "epoch": 0.45680274153790656,
      "flos": 15559806977280.0,
      "grad_norm": 2.146732562241273,
      "language_loss": 0.74601626,
      "learning_rate": 2.3747932360826803e-06,
      "loss": 0.76794761,
      "num_input_tokens_seen": 81798905,
      "step": 3799,
      "time_per_iteration": 2.445308208465576
    },
    {
      "auxiliary_loss_clip": 0.01169416,
      "auxiliary_loss_mlp": 0.01029373,
      "balance_loss_clip": 1.05378258,
      "balance_loss_mlp": 1.02067089,
      "epoch": 0.4569229844285457,
      "flos": 19792884879360.0,
      "grad_norm": 3.0735226956985437,
      "language_loss": 0.8221522,
      "learning_rate": 2.3740280360304205e-06,
      "loss": 0.84414005,
      "num_input_tokens_seen": 81816630,
      "step": 3800,
      "time_per_iteration": 2.4315550327301025
    },
    {
      "auxiliary_loss_clip": 0.01140648,
      "auxiliary_loss_mlp": 0.01029107,
      "balance_loss_clip": 1.05232584,
      "balance_loss_mlp": 1.02033925,
      "epoch": 0.45704322731918473,
      "flos": 24093941270400.0,
      "grad_norm": 1.665226475130565,
      "language_loss": 0.68208665,
      "learning_rate": 2.3732627792384038e-06,
      "loss": 0.70378417,
      "num_input_tokens_seen": 81837700,
      "step": 3801,
      "time_per_iteration": 2.583601713180542
    },
    {
      "auxiliary_loss_clip": 0.01184084,
      "auxiliary_loss_mlp": 0.01027911,
      "balance_loss_clip": 1.05372906,
      "balance_loss_mlp": 1.01939917,
      "epoch": 0.45716347020982384,
      "flos": 31317803245440.0,
      "grad_norm": 3.0154721142397674,
      "language_loss": 0.75556082,
      "learning_rate": 2.3724974658227207e-06,
      "loss": 0.77768075,
      "num_input_tokens_seen": 81858490,
      "step": 3802,
      "time_per_iteration": 2.497620105743408
    },
    {
      "auxiliary_loss_clip": 0.01154197,
      "auxiliary_loss_mlp": 0.00763119,
      "balance_loss_clip": 1.05299437,
      "balance_loss_mlp": 1.0003581,
      "epoch": 0.45728371310046295,
      "flos": 26501392471680.0,
      "grad_norm": 2.28540710772405,
      "language_loss": 0.71281683,
      "learning_rate": 2.3717320958994687e-06,
      "loss": 0.73199004,
      "num_input_tokens_seen": 81876050,
      "step": 3803,
      "time_per_iteration": 3.362089157104492
    },
    {
      "auxiliary_loss_clip": 0.01137497,
      "auxiliary_loss_mlp": 0.01026956,
      "balance_loss_clip": 1.04297101,
      "balance_loss_mlp": 1.0193181,
      "epoch": 0.457403955991102,
      "flos": 17929408222080.0,
      "grad_norm": 2.0194345487736043,
      "language_loss": 0.70159721,
      "learning_rate": 2.3709666695847534e-06,
      "loss": 0.72324181,
      "num_input_tokens_seen": 81894230,
      "step": 3804,
      "time_per_iteration": 2.48596453666687
    },
    {
      "auxiliary_loss_clip": 0.01118354,
      "auxiliary_loss_mlp": 0.01028125,
      "balance_loss_clip": 1.04534924,
      "balance_loss_mlp": 1.02048957,
      "epoch": 0.4575241988817411,
      "flos": 42230660837760.0,
      "grad_norm": 1.5583618484445732,
      "language_loss": 0.69771802,
      "learning_rate": 2.370201186994689e-06,
      "loss": 0.71918273,
      "num_input_tokens_seen": 81917915,
      "step": 3805,
      "time_per_iteration": 2.735992670059204
    },
    {
      "auxiliary_loss_clip": 0.01146528,
      "auxiliary_loss_mlp": 0.01026787,
      "balance_loss_clip": 1.05204535,
      "balance_loss_mlp": 1.01884794,
      "epoch": 0.45764444177238023,
      "flos": 30117309868800.0,
      "grad_norm": 2.056254989275629,
      "language_loss": 0.69927001,
      "learning_rate": 2.369435648245399e-06,
      "loss": 0.72100317,
      "num_input_tokens_seen": 81938130,
      "step": 3806,
      "time_per_iteration": 3.4126553535461426
    },
    {
      "auxiliary_loss_clip": 0.01155552,
      "auxiliary_loss_mlp": 0.01038305,
      "balance_loss_clip": 1.05150104,
      "balance_loss_mlp": 1.02960253,
      "epoch": 0.4577646846630193,
      "flos": 24060293205120.0,
      "grad_norm": 1.679726526005014,
      "language_loss": 0.84967977,
      "learning_rate": 2.368670053453015e-06,
      "loss": 0.87161839,
      "num_input_tokens_seen": 81959820,
      "step": 3807,
      "time_per_iteration": 2.535785436630249
    },
    {
      "auxiliary_loss_clip": 0.01178461,
      "auxiliary_loss_mlp": 0.0102995,
      "balance_loss_clip": 1.05724096,
      "balance_loss_mlp": 1.02100909,
      "epoch": 0.4578849275536584,
      "flos": 17418578952960.0,
      "grad_norm": 2.164443176570588,
      "language_loss": 0.7445218,
      "learning_rate": 2.3679044027336757e-06,
      "loss": 0.76660591,
      "num_input_tokens_seen": 81975710,
      "step": 3808,
      "time_per_iteration": 3.157886505126953
    },
    {
      "auxiliary_loss_clip": 0.01183969,
      "auxiliary_loss_mlp": 0.01028808,
      "balance_loss_clip": 1.05366921,
      "balance_loss_mlp": 1.01993835,
      "epoch": 0.4580051704442975,
      "flos": 13510169107200.0,
      "grad_norm": 4.034110780753551,
      "language_loss": 0.69298583,
      "learning_rate": 2.3671386962035326e-06,
      "loss": 0.71511364,
      "num_input_tokens_seen": 81993180,
      "step": 3809,
      "time_per_iteration": 2.4340832233428955
    },
    {
      "auxiliary_loss_clip": 0.01169907,
      "auxiliary_loss_mlp": 0.01030642,
      "balance_loss_clip": 1.05239487,
      "balance_loss_mlp": 1.02232683,
      "epoch": 0.45812541333493656,
      "flos": 18037606965120.0,
      "grad_norm": 1.9833936227468862,
      "language_loss": 0.68344605,
      "learning_rate": 2.3663729339787405e-06,
      "loss": 0.70545149,
      "num_input_tokens_seen": 82010115,
      "step": 3810,
      "time_per_iteration": 2.450411558151245
    },
    {
      "auxiliary_loss_clip": 0.0118401,
      "auxiliary_loss_mlp": 0.01027722,
      "balance_loss_clip": 1.05341601,
      "balance_loss_mlp": 1.01924682,
      "epoch": 0.45824565622557567,
      "flos": 20222196232320.0,
      "grad_norm": 2.343059868702884,
      "language_loss": 0.7365219,
      "learning_rate": 2.365607116175466e-06,
      "loss": 0.75863922,
      "num_input_tokens_seen": 82025540,
      "step": 3811,
      "time_per_iteration": 3.1849732398986816
    },
    {
      "auxiliary_loss_clip": 0.01181244,
      "auxiliary_loss_mlp": 0.01025431,
      "balance_loss_clip": 1.05337,
      "balance_loss_mlp": 1.01740277,
      "epoch": 0.4583658991162148,
      "flos": 19864885691520.0,
      "grad_norm": 2.6216325182817033,
      "language_loss": 0.67094958,
      "learning_rate": 2.3648412429098825e-06,
      "loss": 0.69301629,
      "num_input_tokens_seen": 82043890,
      "step": 3812,
      "time_per_iteration": 2.469998359680176
    },
    {
      "auxiliary_loss_clip": 0.0113707,
      "auxiliary_loss_mlp": 0.01034498,
      "balance_loss_clip": 1.0495019,
      "balance_loss_mlp": 1.02555776,
      "epoch": 0.45848614200685384,
      "flos": 21029935322880.0,
      "grad_norm": 2.107825383099602,
      "language_loss": 0.82321841,
      "learning_rate": 2.364075314298172e-06,
      "loss": 0.84493411,
      "num_input_tokens_seen": 82061345,
      "step": 3813,
      "time_per_iteration": 2.5379202365875244
    },
    {
      "auxiliary_loss_clip": 0.0117526,
      "auxiliary_loss_mlp": 0.00763366,
      "balance_loss_clip": 1.05519104,
      "balance_loss_mlp": 1.0002799,
      "epoch": 0.45860638489749295,
      "flos": 21069293650560.0,
      "grad_norm": 2.1540155947863795,
      "language_loss": 0.70433772,
      "learning_rate": 2.3633093304565267e-06,
      "loss": 0.72372401,
      "num_input_tokens_seen": 82080400,
      "step": 3814,
      "time_per_iteration": 2.4544529914855957
    },
    {
      "auxiliary_loss_clip": 0.01188547,
      "auxiliary_loss_mlp": 0.01028813,
      "balance_loss_clip": 1.05673897,
      "balance_loss_mlp": 1.02019429,
      "epoch": 0.458726627788132,
      "flos": 26833889692800.0,
      "grad_norm": 1.9495933503349816,
      "language_loss": 0.63341087,
      "learning_rate": 2.3625432915011443e-06,
      "loss": 0.65558445,
      "num_input_tokens_seen": 82102310,
      "step": 3815,
      "time_per_iteration": 2.4853720664978027
    },
    {
      "auxiliary_loss_clip": 0.01150028,
      "auxiliary_loss_mlp": 0.01034011,
      "balance_loss_clip": 1.05091357,
      "balance_loss_mlp": 1.02560687,
      "epoch": 0.4588468706787711,
      "flos": 24097927680000.0,
      "grad_norm": 1.6675263893726129,
      "language_loss": 0.65215021,
      "learning_rate": 2.3617771975482334e-06,
      "loss": 0.67399061,
      "num_input_tokens_seen": 82121140,
      "step": 3816,
      "time_per_iteration": 2.496014356613159
    },
    {
      "auxiliary_loss_clip": 0.01120809,
      "auxiliary_loss_mlp": 0.0102628,
      "balance_loss_clip": 1.04783297,
      "balance_loss_mlp": 1.01863861,
      "epoch": 0.4589671135694102,
      "flos": 17889331622400.0,
      "grad_norm": 1.658072832036926,
      "language_loss": 0.74629033,
      "learning_rate": 2.3610110487140083e-06,
      "loss": 0.76776111,
      "num_input_tokens_seen": 82139575,
      "step": 3817,
      "time_per_iteration": 2.5430617332458496
    },
    {
      "auxiliary_loss_clip": 0.01156699,
      "auxiliary_loss_mlp": 0.01033461,
      "balance_loss_clip": 1.0539124,
      "balance_loss_mlp": 1.02509832,
      "epoch": 0.4590873564600493,
      "flos": 25626967781760.0,
      "grad_norm": 1.794561486391742,
      "language_loss": 0.8051284,
      "learning_rate": 2.360244845114695e-06,
      "loss": 0.82702994,
      "num_input_tokens_seen": 82159195,
      "step": 3818,
      "time_per_iteration": 2.528165578842163
    },
    {
      "auxiliary_loss_clip": 0.01150289,
      "auxiliary_loss_mlp": 0.01028064,
      "balance_loss_clip": 1.05344081,
      "balance_loss_mlp": 1.01942766,
      "epoch": 0.4592075993506884,
      "flos": 18514788168960.0,
      "grad_norm": 2.0803525924065585,
      "language_loss": 0.68684983,
      "learning_rate": 2.3594785868665245e-06,
      "loss": 0.7086333,
      "num_input_tokens_seen": 82175500,
      "step": 3819,
      "time_per_iteration": 2.473207712173462
    },
    {
      "auxiliary_loss_clip": 0.01143613,
      "auxiliary_loss_mlp": 0.00763172,
      "balance_loss_clip": 1.05111611,
      "balance_loss_mlp": 1.00028408,
      "epoch": 0.4593278422413275,
      "flos": 20631111638400.0,
      "grad_norm": 2.3835872762282255,
      "language_loss": 0.80651808,
      "learning_rate": 2.3587122740857386e-06,
      "loss": 0.82558596,
      "num_input_tokens_seen": 82192600,
      "step": 3820,
      "time_per_iteration": 2.5180118083953857
    },
    {
      "auxiliary_loss_clip": 0.01165594,
      "auxiliary_loss_mlp": 0.0102739,
      "balance_loss_clip": 1.0503968,
      "balance_loss_mlp": 1.01959991,
      "epoch": 0.45944808513196655,
      "flos": 21358517961600.0,
      "grad_norm": 1.7674697716118215,
      "language_loss": 0.77810109,
      "learning_rate": 2.357945906888586e-06,
      "loss": 0.80003089,
      "num_input_tokens_seen": 82212040,
      "step": 3821,
      "time_per_iteration": 2.4755611419677734
    },
    {
      "auxiliary_loss_clip": 0.01172685,
      "auxiliary_loss_mlp": 0.01032657,
      "balance_loss_clip": 1.05465913,
      "balance_loss_mlp": 1.02368617,
      "epoch": 0.45956832802260567,
      "flos": 21427789340160.0,
      "grad_norm": 6.857630860590178,
      "language_loss": 0.79621851,
      "learning_rate": 2.357179485391324e-06,
      "loss": 0.81827199,
      "num_input_tokens_seen": 82229895,
      "step": 3822,
      "time_per_iteration": 2.4633724689483643
    },
    {
      "auxiliary_loss_clip": 0.01184048,
      "auxiliary_loss_mlp": 0.0102639,
      "balance_loss_clip": 1.05670071,
      "balance_loss_mlp": 1.01808143,
      "epoch": 0.4596885709132448,
      "flos": 22382654538240.0,
      "grad_norm": 3.3939572568218463,
      "language_loss": 0.86541921,
      "learning_rate": 2.3564130097102173e-06,
      "loss": 0.88752359,
      "num_input_tokens_seen": 82249550,
      "step": 3823,
      "time_per_iteration": 2.431030750274658
    },
    {
      "auxiliary_loss_clip": 0.01148772,
      "auxiliary_loss_mlp": 0.01026805,
      "balance_loss_clip": 1.05376208,
      "balance_loss_mlp": 1.01805544,
      "epoch": 0.45980881380388383,
      "flos": 28981957806720.0,
      "grad_norm": 1.753479658580137,
      "language_loss": 0.75157654,
      "learning_rate": 2.355646479961541e-06,
      "loss": 0.77333236,
      "num_input_tokens_seen": 82268860,
      "step": 3824,
      "time_per_iteration": 2.6220083236694336
    },
    {
      "auxiliary_loss_clip": 0.01182269,
      "auxiliary_loss_mlp": 0.01026609,
      "balance_loss_clip": 1.05386233,
      "balance_loss_mlp": 1.01755476,
      "epoch": 0.45992905669452294,
      "flos": 33396599980800.0,
      "grad_norm": 2.1645082455652007,
      "language_loss": 0.71528888,
      "learning_rate": 2.354879896261576e-06,
      "loss": 0.73737764,
      "num_input_tokens_seen": 82289070,
      "step": 3825,
      "time_per_iteration": 2.5155882835388184
    },
    {
      "auxiliary_loss_clip": 0.01138515,
      "auxiliary_loss_mlp": 0.01030489,
      "balance_loss_clip": 1.05234957,
      "balance_loss_mlp": 1.02254975,
      "epoch": 0.46004929958516205,
      "flos": 36318184502400.0,
      "grad_norm": 1.8018786347648352,
      "language_loss": 0.56658053,
      "learning_rate": 2.3541132587266133e-06,
      "loss": 0.58827055,
      "num_input_tokens_seen": 82311790,
      "step": 3826,
      "time_per_iteration": 2.650313377380371
    },
    {
      "auxiliary_loss_clip": 0.01147093,
      "auxiliary_loss_mlp": 0.01026609,
      "balance_loss_clip": 1.0502224,
      "balance_loss_mlp": 1.01831234,
      "epoch": 0.4601695424758011,
      "flos": 17238451224960.0,
      "grad_norm": 1.9524804197617511,
      "language_loss": 0.69235218,
      "learning_rate": 2.3533465674729515e-06,
      "loss": 0.71408916,
      "num_input_tokens_seen": 82329020,
      "step": 3827,
      "time_per_iteration": 2.5046355724334717
    },
    {
      "auxiliary_loss_clip": 0.01184545,
      "auxiliary_loss_mlp": 0.01030169,
      "balance_loss_clip": 1.05534017,
      "balance_loss_mlp": 1.0213778,
      "epoch": 0.4602897853664402,
      "flos": 15888425529600.0,
      "grad_norm": 1.9207453384285458,
      "language_loss": 0.72819066,
      "learning_rate": 2.352579822616895e-06,
      "loss": 0.75033784,
      "num_input_tokens_seen": 82346455,
      "step": 3828,
      "time_per_iteration": 2.4058139324188232
    },
    {
      "auxiliary_loss_clip": 0.01157537,
      "auxiliary_loss_mlp": 0.01026007,
      "balance_loss_clip": 1.05190682,
      "balance_loss_mlp": 1.01799881,
      "epoch": 0.4604100282570793,
      "flos": 25412617370880.0,
      "grad_norm": 1.68594507157666,
      "language_loss": 0.77542496,
      "learning_rate": 2.351813024274761e-06,
      "loss": 0.7972604,
      "num_input_tokens_seen": 82367810,
      "step": 3829,
      "time_per_iteration": 3.362781286239624
    },
    {
      "auxiliary_loss_clip": 0.01145407,
      "auxiliary_loss_mlp": 0.01031358,
      "balance_loss_clip": 1.05089378,
      "balance_loss_mlp": 1.02305484,
      "epoch": 0.4605302711477184,
      "flos": 27630711048960.0,
      "grad_norm": 1.7564613473035389,
      "language_loss": 0.73894072,
      "learning_rate": 2.3510461725628693e-06,
      "loss": 0.76070833,
      "num_input_tokens_seen": 82388275,
      "step": 3830,
      "time_per_iteration": 2.5806941986083984
    },
    {
      "auxiliary_loss_clip": 0.01143263,
      "auxiliary_loss_mlp": 0.01028616,
      "balance_loss_clip": 1.05027628,
      "balance_loss_mlp": 1.02035475,
      "epoch": 0.4606505140383575,
      "flos": 23839657914240.0,
      "grad_norm": 1.7831035193181826,
      "language_loss": 0.71025854,
      "learning_rate": 2.350279267597554e-06,
      "loss": 0.73197734,
      "num_input_tokens_seen": 82408915,
      "step": 3831,
      "time_per_iteration": 2.596791982650757
    },
    {
      "auxiliary_loss_clip": 0.01171302,
      "auxiliary_loss_mlp": 0.01030528,
      "balance_loss_clip": 1.05531764,
      "balance_loss_mlp": 1.02201009,
      "epoch": 0.46077075692899655,
      "flos": 16107013745280.0,
      "grad_norm": 2.1336997328389207,
      "language_loss": 0.82908005,
      "learning_rate": 2.3495123094951515e-06,
      "loss": 0.8510983,
      "num_input_tokens_seen": 82427260,
      "step": 3832,
      "time_per_iteration": 2.4435081481933594
    },
    {
      "auxiliary_loss_clip": 0.01148236,
      "auxiliary_loss_mlp": 0.01023938,
      "balance_loss_clip": 1.05136442,
      "balance_loss_mlp": 1.01548648,
      "epoch": 0.46089099981963566,
      "flos": 48798147634560.0,
      "grad_norm": 1.9922786381061162,
      "language_loss": 0.75751519,
      "learning_rate": 2.34874529837201e-06,
      "loss": 0.77923691,
      "num_input_tokens_seen": 82450805,
      "step": 3833,
      "time_per_iteration": 3.5840744972229004
    },
    {
      "auxiliary_loss_clip": 0.01107561,
      "auxiliary_loss_mlp": 0.01022866,
      "balance_loss_clip": 1.04419804,
      "balance_loss_mlp": 1.01472449,
      "epoch": 0.46101124271027477,
      "flos": 19099234362240.0,
      "grad_norm": 2.171825253013662,
      "language_loss": 0.79380012,
      "learning_rate": 2.347978234344483e-06,
      "loss": 0.81510437,
      "num_input_tokens_seen": 82467010,
      "step": 3834,
      "time_per_iteration": 2.5585708618164062
    },
    {
      "auxiliary_loss_clip": 0.01173899,
      "auxiliary_loss_mlp": 0.0103647,
      "balance_loss_clip": 1.0553906,
      "balance_loss_mlp": 1.02728498,
      "epoch": 0.4611314856009138,
      "flos": 39347931853440.0,
      "grad_norm": 1.8323993137472896,
      "language_loss": 0.69013137,
      "learning_rate": 2.347211117528935e-06,
      "loss": 0.71223503,
      "num_input_tokens_seen": 82489310,
      "step": 3835,
      "time_per_iteration": 3.407909393310547
    },
    {
      "auxiliary_loss_clip": 0.01150674,
      "auxiliary_loss_mlp": 0.01030206,
      "balance_loss_clip": 1.05464113,
      "balance_loss_mlp": 1.02174199,
      "epoch": 0.46125172849155294,
      "flos": 20810772489600.0,
      "grad_norm": 1.5469928731392724,
      "language_loss": 0.71623951,
      "learning_rate": 2.3464439480417374e-06,
      "loss": 0.73804832,
      "num_input_tokens_seen": 82508830,
      "step": 3836,
      "time_per_iteration": 2.569108724594116
    },
    {
      "auxiliary_loss_clip": 0.01171436,
      "auxiliary_loss_mlp": 0.01032622,
      "balance_loss_clip": 1.0537746,
      "balance_loss_mlp": 1.0240624,
      "epoch": 0.46137197138219205,
      "flos": 17930808852480.0,
      "grad_norm": 2.350554078955025,
      "language_loss": 0.77454567,
      "learning_rate": 2.3456767259992676e-06,
      "loss": 0.79658628,
      "num_input_tokens_seen": 82526475,
      "step": 3837,
      "time_per_iteration": 2.458272933959961
    },
    {
      "auxiliary_loss_clip": 0.0118257,
      "auxiliary_loss_mlp": 0.00763591,
      "balance_loss_clip": 1.05277205,
      "balance_loss_mlp": 1.00021482,
      "epoch": 0.4614922142728311,
      "flos": 16836610798080.0,
      "grad_norm": 2.6381895017292294,
      "language_loss": 0.88564372,
      "learning_rate": 2.3449094515179135e-06,
      "loss": 0.90510535,
      "num_input_tokens_seen": 82543935,
      "step": 3838,
      "time_per_iteration": 3.176616907119751
    },
    {
      "auxiliary_loss_clip": 0.01160292,
      "auxiliary_loss_mlp": 0.01027995,
      "balance_loss_clip": 1.05135155,
      "balance_loss_mlp": 1.01922071,
      "epoch": 0.4616124571634702,
      "flos": 26614906427520.0,
      "grad_norm": 1.6624241672844116,
      "language_loss": 0.81702513,
      "learning_rate": 2.34414212471407e-06,
      "loss": 0.83890796,
      "num_input_tokens_seen": 82563730,
      "step": 3839,
      "time_per_iteration": 2.5248005390167236
    },
    {
      "auxiliary_loss_clip": 0.01176128,
      "auxiliary_loss_mlp": 0.01026621,
      "balance_loss_clip": 1.05352831,
      "balance_loss_mlp": 1.01817524,
      "epoch": 0.4617327000541093,
      "flos": 20340127560960.0,
      "grad_norm": 1.939585706846273,
      "language_loss": 0.72500026,
      "learning_rate": 2.3433747457041394e-06,
      "loss": 0.74702775,
      "num_input_tokens_seen": 82582435,
      "step": 3840,
      "time_per_iteration": 2.439389228820801
    },
    {
      "auxiliary_loss_clip": 0.01143336,
      "auxiliary_loss_mlp": 0.01030357,
      "balance_loss_clip": 1.05171943,
      "balance_loss_mlp": 1.02132726,
      "epoch": 0.4618529429447484,
      "flos": 29570749545600.0,
      "grad_norm": 1.9176713613126162,
      "language_loss": 0.84948707,
      "learning_rate": 2.342607314604533e-06,
      "loss": 0.87122405,
      "num_input_tokens_seen": 82602185,
      "step": 3841,
      "time_per_iteration": 2.584968090057373
    },
    {
      "auxiliary_loss_clip": 0.01172325,
      "auxiliary_loss_mlp": 0.01031801,
      "balance_loss_clip": 1.05704713,
      "balance_loss_mlp": 1.02339697,
      "epoch": 0.4619731858353875,
      "flos": 19787030962560.0,
      "grad_norm": 1.8793781865567267,
      "language_loss": 0.83882046,
      "learning_rate": 2.3418398315316694e-06,
      "loss": 0.86086166,
      "num_input_tokens_seen": 82620005,
      "step": 3842,
      "time_per_iteration": 2.4580259323120117
    },
    {
      "auxiliary_loss_clip": 0.01185784,
      "auxiliary_loss_mlp": 0.01039139,
      "balance_loss_clip": 1.05766439,
      "balance_loss_mlp": 1.03060961,
      "epoch": 0.4620934287260266,
      "flos": 18951138587520.0,
      "grad_norm": 2.2831325224672594,
      "language_loss": 0.77869642,
      "learning_rate": 2.3410722966019755e-06,
      "loss": 0.8009457,
      "num_input_tokens_seen": 82635120,
      "step": 3843,
      "time_per_iteration": 2.396442174911499
    },
    {
      "auxiliary_loss_clip": 0.01168172,
      "auxiliary_loss_mlp": 0.0102786,
      "balance_loss_clip": 1.0526011,
      "balance_loss_mlp": 1.01926446,
      "epoch": 0.46221367161666566,
      "flos": 37341674634240.0,
      "grad_norm": 1.85605789797141,
      "language_loss": 0.65926111,
      "learning_rate": 2.3403047099318848e-06,
      "loss": 0.68122143,
      "num_input_tokens_seen": 82659190,
      "step": 3844,
      "time_per_iteration": 2.5998647212982178
    },
    {
      "auxiliary_loss_clip": 0.01122707,
      "auxiliary_loss_mlp": 0.01024872,
      "balance_loss_clip": 1.0470562,
      "balance_loss_mlp": 1.01669991,
      "epoch": 0.46233391450730477,
      "flos": 14428549065600.0,
      "grad_norm": 2.099728382360556,
      "language_loss": 0.75131309,
      "learning_rate": 2.3395370716378405e-06,
      "loss": 0.77278888,
      "num_input_tokens_seen": 82676635,
      "step": 3845,
      "time_per_iteration": 2.537075996398926
    },
    {
      "auxiliary_loss_clip": 0.01173542,
      "auxiliary_loss_mlp": 0.01032499,
      "balance_loss_clip": 1.05349278,
      "balance_loss_mlp": 1.02434552,
      "epoch": 0.4624541573979438,
      "flos": 22493044010880.0,
      "grad_norm": 2.342088528436429,
      "language_loss": 0.72600794,
      "learning_rate": 2.338769381836292e-06,
      "loss": 0.74806833,
      "num_input_tokens_seen": 82696245,
      "step": 3846,
      "time_per_iteration": 2.4618353843688965
    },
    {
      "auxiliary_loss_clip": 0.01140005,
      "auxiliary_loss_mlp": 0.01034251,
      "balance_loss_clip": 1.05322385,
      "balance_loss_mlp": 1.02590084,
      "epoch": 0.46257440028858293,
      "flos": 14465070218880.0,
      "grad_norm": 2.3751685649516583,
      "language_loss": 0.73069572,
      "learning_rate": 2.3380016406436984e-06,
      "loss": 0.75243831,
      "num_input_tokens_seen": 82713725,
      "step": 3847,
      "time_per_iteration": 2.506568193435669
    },
    {
      "auxiliary_loss_clip": 0.01129059,
      "auxiliary_loss_mlp": 0.01034719,
      "balance_loss_clip": 1.05363083,
      "balance_loss_mlp": 1.02552223,
      "epoch": 0.46269464317922204,
      "flos": 23332204523520.0,
      "grad_norm": 2.1321927862599144,
      "language_loss": 0.8112278,
      "learning_rate": 2.337233848176524e-06,
      "loss": 0.8328656,
      "num_input_tokens_seen": 82731495,
      "step": 3848,
      "time_per_iteration": 2.5549581050872803
    },
    {
      "auxiliary_loss_clip": 0.01121183,
      "auxiliary_loss_mlp": 0.01028705,
      "balance_loss_clip": 1.04880321,
      "balance_loss_mlp": 1.02044988,
      "epoch": 0.4628148860698611,
      "flos": 18552027594240.0,
      "grad_norm": 2.0328690964437954,
      "language_loss": 0.8365525,
      "learning_rate": 2.3364660045512435e-06,
      "loss": 0.85805136,
      "num_input_tokens_seen": 82750255,
      "step": 3849,
      "time_per_iteration": 2.545180082321167
    },
    {
      "auxiliary_loss_clip": 0.01062477,
      "auxiliary_loss_mlp": 0.01003105,
      "balance_loss_clip": 1.02090359,
      "balance_loss_mlp": 1.002038,
      "epoch": 0.4629351289605002,
      "flos": 70667569670400.0,
      "grad_norm": 1.0325058665732432,
      "language_loss": 0.58241451,
      "learning_rate": 2.335698109884337e-06,
      "loss": 0.60307026,
      "num_input_tokens_seen": 82815460,
      "step": 3850,
      "time_per_iteration": 3.229626178741455
    },
    {
      "auxiliary_loss_clip": 0.01050997,
      "auxiliary_loss_mlp": 0.01003792,
      "balance_loss_clip": 1.03229666,
      "balance_loss_mlp": 1.00217104,
      "epoch": 0.4630553718511393,
      "flos": 59687200465920.0,
      "grad_norm": 0.7892037359808743,
      "language_loss": 0.59842908,
      "learning_rate": 2.334930164292294e-06,
      "loss": 0.61897695,
      "num_input_tokens_seen": 82878010,
      "step": 3851,
      "time_per_iteration": 3.260227918624878
    },
    {
      "auxiliary_loss_clip": 0.01120033,
      "auxiliary_loss_mlp": 0.01028735,
      "balance_loss_clip": 1.04607761,
      "balance_loss_mlp": 1.02115333,
      "epoch": 0.4631756147417784,
      "flos": 15960605909760.0,
      "grad_norm": 2.0209326958860667,
      "language_loss": 0.79903626,
      "learning_rate": 2.334162167891612e-06,
      "loss": 0.82052398,
      "num_input_tokens_seen": 82895275,
      "step": 3852,
      "time_per_iteration": 2.555870294570923
    },
    {
      "auxiliary_loss_clip": 0.01157046,
      "auxiliary_loss_mlp": 0.0103239,
      "balance_loss_clip": 1.05050707,
      "balance_loss_mlp": 1.02332997,
      "epoch": 0.4632958576324175,
      "flos": 16472907636480.0,
      "grad_norm": 2.4383968191116066,
      "language_loss": 0.75190365,
      "learning_rate": 2.333394120798795e-06,
      "loss": 0.77379799,
      "num_input_tokens_seen": 82914010,
      "step": 3853,
      "time_per_iteration": 2.510403871536255
    },
    {
      "auxiliary_loss_clip": 0.01153787,
      "auxiliary_loss_mlp": 0.01023884,
      "balance_loss_clip": 1.0492934,
      "balance_loss_mlp": 1.01502132,
      "epoch": 0.4634161005230566,
      "flos": 22346492520960.0,
      "grad_norm": 2.247237940543874,
      "language_loss": 0.71964312,
      "learning_rate": 2.3326260231303545e-06,
      "loss": 0.74141979,
      "num_input_tokens_seen": 82932610,
      "step": 3854,
      "time_per_iteration": 2.5033962726593018
    },
    {
      "auxiliary_loss_clip": 0.01182588,
      "auxiliary_loss_mlp": 0.01025743,
      "balance_loss_clip": 1.05686522,
      "balance_loss_mlp": 1.01766622,
      "epoch": 0.46353634341369565,
      "flos": 15742233175680.0,
      "grad_norm": 1.6235630507052292,
      "language_loss": 0.86317807,
      "learning_rate": 2.331857875002811e-06,
      "loss": 0.88526136,
      "num_input_tokens_seen": 82951210,
      "step": 3855,
      "time_per_iteration": 2.417605400085449
    },
    {
      "auxiliary_loss_clip": 0.01157342,
      "auxiliary_loss_mlp": 0.01032131,
      "balance_loss_clip": 1.05550969,
      "balance_loss_mlp": 1.02400088,
      "epoch": 0.46365658630433476,
      "flos": 28329820433280.0,
      "grad_norm": 1.6870892582817414,
      "language_loss": 0.75995284,
      "learning_rate": 2.3310896765326916e-06,
      "loss": 0.7818476,
      "num_input_tokens_seen": 82972210,
      "step": 3856,
      "time_per_iteration": 3.3949105739593506
    },
    {
      "auxiliary_loss_clip": 0.01138665,
      "auxiliary_loss_mlp": 0.01033063,
      "balance_loss_clip": 1.05045128,
      "balance_loss_mlp": 1.02427781,
      "epoch": 0.46377682919497387,
      "flos": 24608074590720.0,
      "grad_norm": 1.5815510143294145,
      "language_loss": 0.84264368,
      "learning_rate": 2.330321427836531e-06,
      "loss": 0.86436093,
      "num_input_tokens_seen": 82994080,
      "step": 3857,
      "time_per_iteration": 2.5600061416625977
    },
    {
      "auxiliary_loss_clip": 0.01165121,
      "auxiliary_loss_mlp": 0.01026022,
      "balance_loss_clip": 1.05233979,
      "balance_loss_mlp": 1.01702785,
      "epoch": 0.4638970720856129,
      "flos": 19060953442560.0,
      "grad_norm": 1.605440585062139,
      "language_loss": 0.82842326,
      "learning_rate": 2.3295531290308733e-06,
      "loss": 0.8503347,
      "num_input_tokens_seen": 83012230,
      "step": 3858,
      "time_per_iteration": 2.44889760017395
    },
    {
      "auxiliary_loss_clip": 0.01185408,
      "auxiliary_loss_mlp": 0.00763224,
      "balance_loss_clip": 1.05544567,
      "balance_loss_mlp": 1.00024676,
      "epoch": 0.46401731497625204,
      "flos": 18471012468480.0,
      "grad_norm": 2.515361157905502,
      "language_loss": 0.75979018,
      "learning_rate": 2.3287847802322678e-06,
      "loss": 0.77927655,
      "num_input_tokens_seen": 83027800,
      "step": 3859,
      "time_per_iteration": 2.388307571411133
    },
    {
      "auxiliary_loss_clip": 0.01164749,
      "auxiliary_loss_mlp": 0.01026611,
      "balance_loss_clip": 1.05549955,
      "balance_loss_mlp": 1.01785541,
      "epoch": 0.4641375578668911,
      "flos": 26067053214720.0,
      "grad_norm": 1.730979245222694,
      "language_loss": 0.83645052,
      "learning_rate": 2.3280163815572723e-06,
      "loss": 0.85836411,
      "num_input_tokens_seen": 83048395,
      "step": 3860,
      "time_per_iteration": 3.386744499206543
    },
    {
      "auxiliary_loss_clip": 0.01146104,
      "auxiliary_loss_mlp": 0.01026458,
      "balance_loss_clip": 1.04920149,
      "balance_loss_mlp": 1.01835442,
      "epoch": 0.4642578007575302,
      "flos": 19570382081280.0,
      "grad_norm": 2.2315802115092533,
      "language_loss": 0.76647878,
      "learning_rate": 2.3272479331224522e-06,
      "loss": 0.78820431,
      "num_input_tokens_seen": 83065825,
      "step": 3861,
      "time_per_iteration": 2.4834401607513428
    },
    {
      "auxiliary_loss_clip": 0.01185677,
      "auxiliary_loss_mlp": 0.01025621,
      "balance_loss_clip": 1.05557144,
      "balance_loss_mlp": 1.01763952,
      "epoch": 0.4643780436481693,
      "flos": 28186249772160.0,
      "grad_norm": 1.659882677175064,
      "language_loss": 0.77978754,
      "learning_rate": 2.3264794350443817e-06,
      "loss": 0.80190051,
      "num_input_tokens_seen": 83087920,
      "step": 3862,
      "time_per_iteration": 3.3304646015167236
    },
    {
      "auxiliary_loss_clip": 0.01170686,
      "auxiliary_loss_mlp": 0.01024434,
      "balance_loss_clip": 1.05011213,
      "balance_loss_mlp": 1.01621449,
      "epoch": 0.46449828653880837,
      "flos": 25375270204800.0,
      "grad_norm": 1.8237793170602534,
      "language_loss": 0.78528023,
      "learning_rate": 2.3257108874396396e-06,
      "loss": 0.80723143,
      "num_input_tokens_seen": 83109015,
      "step": 3863,
      "time_per_iteration": 2.516768217086792
    },
    {
      "auxiliary_loss_clip": 0.01155008,
      "auxiliary_loss_mlp": 0.01037081,
      "balance_loss_clip": 1.0512743,
      "balance_loss_mlp": 1.02834868,
      "epoch": 0.4646185294294475,
      "flos": 16034330574720.0,
      "grad_norm": 1.9480766458141583,
      "language_loss": 0.73751605,
      "learning_rate": 2.3249422904248152e-06,
      "loss": 0.75943691,
      "num_input_tokens_seen": 83127450,
      "step": 3864,
      "time_per_iteration": 3.2613096237182617
    },
    {
      "auxiliary_loss_clip": 0.01171798,
      "auxiliary_loss_mlp": 0.01028525,
      "balance_loss_clip": 1.0532186,
      "balance_loss_mlp": 1.02071977,
      "epoch": 0.4647387723200866,
      "flos": 26363101109760.0,
      "grad_norm": 1.4045701801822494,
      "language_loss": 0.86915839,
      "learning_rate": 2.324173644116504e-06,
      "loss": 0.89116168,
      "num_input_tokens_seen": 83150300,
      "step": 3865,
      "time_per_iteration": 2.5376906394958496
    },
    {
      "auxiliary_loss_clip": 0.01165689,
      "auxiliary_loss_mlp": 0.01026539,
      "balance_loss_clip": 1.0545218,
      "balance_loss_mlp": 1.01884961,
      "epoch": 0.46485901521072565,
      "flos": 27160209774720.0,
      "grad_norm": 2.0221487893706342,
      "language_loss": 0.81538665,
      "learning_rate": 2.3234049486313087e-06,
      "loss": 0.83730888,
      "num_input_tokens_seen": 83171750,
      "step": 3866,
      "time_per_iteration": 2.511503219604492
    },
    {
      "auxiliary_loss_clip": 0.01166536,
      "auxiliary_loss_mlp": 0.01026238,
      "balance_loss_clip": 1.05282724,
      "balance_loss_mlp": 1.01875186,
      "epoch": 0.46497925810136476,
      "flos": 24279851088000.0,
      "grad_norm": 2.5105874711915845,
      "language_loss": 0.75770646,
      "learning_rate": 2.322636204085839e-06,
      "loss": 0.77963424,
      "num_input_tokens_seen": 83191820,
      "step": 3867,
      "time_per_iteration": 2.4915366172790527
    },
    {
      "auxiliary_loss_clip": 0.01144801,
      "auxiliary_loss_mlp": 0.01032367,
      "balance_loss_clip": 1.04699802,
      "balance_loss_mlp": 1.02395701,
      "epoch": 0.46509950099200387,
      "flos": 16253134272000.0,
      "grad_norm": 2.1503541626242586,
      "language_loss": 0.78623295,
      "learning_rate": 2.3218674105967143e-06,
      "loss": 0.80800462,
      "num_input_tokens_seen": 83210085,
      "step": 3868,
      "time_per_iteration": 2.469186544418335
    },
    {
      "auxiliary_loss_clip": 0.01148773,
      "auxiliary_loss_mlp": 0.01029401,
      "balance_loss_clip": 1.04975271,
      "balance_loss_mlp": 1.02129507,
      "epoch": 0.4652197438826429,
      "flos": 23442270773760.0,
      "grad_norm": 1.5589792835783813,
      "language_loss": 0.8346957,
      "learning_rate": 2.3210985682805593e-06,
      "loss": 0.85647738,
      "num_input_tokens_seen": 83231865,
      "step": 3869,
      "time_per_iteration": 2.558034658432007
    },
    {
      "auxiliary_loss_clip": 0.01185475,
      "auxiliary_loss_mlp": 0.01026578,
      "balance_loss_clip": 1.05710042,
      "balance_loss_mlp": 1.01839995,
      "epoch": 0.46533998677328203,
      "flos": 16216397637120.0,
      "grad_norm": 2.2559173279609803,
      "language_loss": 0.68302858,
      "learning_rate": 2.320329677254007e-06,
      "loss": 0.70514911,
      "num_input_tokens_seen": 83249195,
      "step": 3870,
      "time_per_iteration": 2.3962395191192627
    },
    {
      "auxiliary_loss_clip": 0.0118221,
      "auxiliary_loss_mlp": 0.01026913,
      "balance_loss_clip": 1.05451286,
      "balance_loss_mlp": 1.01862216,
      "epoch": 0.46546022966392114,
      "flos": 21141869080320.0,
      "grad_norm": 2.7025901157905556,
      "language_loss": 0.72160763,
      "learning_rate": 2.319560737633697e-06,
      "loss": 0.74369889,
      "num_input_tokens_seen": 83267915,
      "step": 3871,
      "time_per_iteration": 2.4524593353271484
    },
    {
      "auxiliary_loss_clip": 0.011467,
      "auxiliary_loss_mlp": 0.01028435,
      "balance_loss_clip": 1.04927289,
      "balance_loss_mlp": 1.01963747,
      "epoch": 0.4655804725545602,
      "flos": 41171942442240.0,
      "grad_norm": 1.7049153411179157,
      "language_loss": 0.6811558,
      "learning_rate": 2.3187917495362775e-06,
      "loss": 0.70290715,
      "num_input_tokens_seen": 83292325,
      "step": 3872,
      "time_per_iteration": 2.7060701847076416
    },
    {
      "auxiliary_loss_clip": 0.01126992,
      "auxiliary_loss_mlp": 0.01035594,
      "balance_loss_clip": 1.05036211,
      "balance_loss_mlp": 1.0270648,
      "epoch": 0.4657007154451993,
      "flos": 19570956698880.0,
      "grad_norm": 2.5185733140767805,
      "language_loss": 0.77119905,
      "learning_rate": 2.318022713078403e-06,
      "loss": 0.79282498,
      "num_input_tokens_seen": 83306905,
      "step": 3873,
      "time_per_iteration": 2.528513193130493
    },
    {
      "auxiliary_loss_clip": 0.01154149,
      "auxiliary_loss_mlp": 0.01032861,
      "balance_loss_clip": 1.05237174,
      "balance_loss_mlp": 1.02481437,
      "epoch": 0.4658209583358384,
      "flos": 15517826956800.0,
      "grad_norm": 2.1639419508203273,
      "language_loss": 0.84712434,
      "learning_rate": 2.3172536283767354e-06,
      "loss": 0.86899447,
      "num_input_tokens_seen": 83320665,
      "step": 3874,
      "time_per_iteration": 2.446023464202881
    },
    {
      "auxiliary_loss_clip": 0.01138724,
      "auxiliary_loss_mlp": 0.01025478,
      "balance_loss_clip": 1.05147064,
      "balance_loss_mlp": 1.01665688,
      "epoch": 0.4659412012264775,
      "flos": 14903180403840.0,
      "grad_norm": 2.42409469856878,
      "language_loss": 0.80850017,
      "learning_rate": 2.3164844955479447e-06,
      "loss": 0.83014214,
      "num_input_tokens_seen": 83336475,
      "step": 3875,
      "time_per_iteration": 2.4981679916381836
    },
    {
      "auxiliary_loss_clip": 0.01136046,
      "auxiliary_loss_mlp": 0.01026874,
      "balance_loss_clip": 1.04999542,
      "balance_loss_mlp": 1.01835036,
      "epoch": 0.4660614441171166,
      "flos": 24425612478720.0,
      "grad_norm": 1.6047061721604563,
      "language_loss": 0.70705074,
      "learning_rate": 2.3157153147087082e-06,
      "loss": 0.7286799,
      "num_input_tokens_seen": 83358365,
      "step": 3876,
      "time_per_iteration": 2.606342315673828
    },
    {
      "auxiliary_loss_clip": 0.01135429,
      "auxiliary_loss_mlp": 0.01027477,
      "balance_loss_clip": 1.0527494,
      "balance_loss_mlp": 1.01957297,
      "epoch": 0.46618168700775564,
      "flos": 22091095843200.0,
      "grad_norm": 1.8187304219263447,
      "language_loss": 0.83066249,
      "learning_rate": 2.314946085975709e-06,
      "loss": 0.85229158,
      "num_input_tokens_seen": 83377345,
      "step": 3877,
      "time_per_iteration": 2.5547399520874023
    },
    {
      "auxiliary_loss_clip": 0.01134007,
      "auxiliary_loss_mlp": 0.01029665,
      "balance_loss_clip": 1.0530827,
      "balance_loss_mlp": 1.02146959,
      "epoch": 0.46630192989839475,
      "flos": 26176975810560.0,
      "grad_norm": 1.6979508638888605,
      "language_loss": 0.82295507,
      "learning_rate": 2.3141768094656393e-06,
      "loss": 0.84459186,
      "num_input_tokens_seen": 83395920,
      "step": 3878,
      "time_per_iteration": 2.5586094856262207
    },
    {
      "auxiliary_loss_clip": 0.01105843,
      "auxiliary_loss_mlp": 0.0102888,
      "balance_loss_clip": 1.04454708,
      "balance_loss_mlp": 1.02097595,
      "epoch": 0.46642217278903386,
      "flos": 11509622150400.0,
      "grad_norm": 2.3558260735724907,
      "language_loss": 0.82952344,
      "learning_rate": 2.3134074852951966e-06,
      "loss": 0.85087067,
      "num_input_tokens_seen": 83412510,
      "step": 3879,
      "time_per_iteration": 2.580610513687134
    },
    {
      "auxiliary_loss_clip": 0.01119884,
      "auxiliary_loss_mlp": 0.01030945,
      "balance_loss_clip": 1.04363728,
      "balance_loss_mlp": 1.02316689,
      "epoch": 0.4665424156796729,
      "flos": 32306819299200.0,
      "grad_norm": 1.6116375456185172,
      "language_loss": 0.77827567,
      "learning_rate": 2.312638113581088e-06,
      "loss": 0.79978389,
      "num_input_tokens_seen": 83432995,
      "step": 3880,
      "time_per_iteration": 2.635580062866211
    },
    {
      "auxiliary_loss_clip": 0.01168217,
      "auxiliary_loss_mlp": 0.01029725,
      "balance_loss_clip": 1.05089295,
      "balance_loss_mlp": 1.02157664,
      "epoch": 0.46666265857031203,
      "flos": 18436179254400.0,
      "grad_norm": 2.573096329465983,
      "language_loss": 0.77883887,
      "learning_rate": 2.311868694440027e-06,
      "loss": 0.80081832,
      "num_input_tokens_seen": 83447415,
      "step": 3881,
      "time_per_iteration": 2.4474406242370605
    },
    {
      "auxiliary_loss_clip": 0.01086467,
      "auxiliary_loss_mlp": 0.01004618,
      "balance_loss_clip": 1.02231169,
      "balance_loss_mlp": 1.00350344,
      "epoch": 0.46678290146095114,
      "flos": 68438989221120.0,
      "grad_norm": 0.732041733792607,
      "language_loss": 0.62523288,
      "learning_rate": 2.3110992279887323e-06,
      "loss": 0.64614373,
      "num_input_tokens_seen": 83519340,
      "step": 3882,
      "time_per_iteration": 3.1565401554107666
    },
    {
      "auxiliary_loss_clip": 0.01147862,
      "auxiliary_loss_mlp": 0.01028464,
      "balance_loss_clip": 1.05331504,
      "balance_loss_mlp": 1.02008343,
      "epoch": 0.4669031443515902,
      "flos": 17712507945600.0,
      "grad_norm": 2.2361374047629696,
      "language_loss": 0.84642488,
      "learning_rate": 2.310329714343932e-06,
      "loss": 0.86818814,
      "num_input_tokens_seen": 83535490,
      "step": 3883,
      "time_per_iteration": 3.3573288917541504
    },
    {
      "auxiliary_loss_clip": 0.01150141,
      "auxiliary_loss_mlp": 0.01023739,
      "balance_loss_clip": 1.05067158,
      "balance_loss_mlp": 1.01568651,
      "epoch": 0.4670233872422293,
      "flos": 23947748916480.0,
      "grad_norm": 2.0114157679523506,
      "language_loss": 0.8227917,
      "learning_rate": 2.309560153622361e-06,
      "loss": 0.84453058,
      "num_input_tokens_seen": 83552400,
      "step": 3884,
      "time_per_iteration": 2.5054590702056885
    },
    {
      "auxiliary_loss_clip": 0.01140206,
      "auxiliary_loss_mlp": 0.01029606,
      "balance_loss_clip": 1.05041337,
      "balance_loss_mlp": 1.02079654,
      "epoch": 0.4671436301328684,
      "flos": 28111268131200.0,
      "grad_norm": 1.907342007010759,
      "language_loss": 0.74632037,
      "learning_rate": 2.3087905459407602e-06,
      "loss": 0.76801848,
      "num_input_tokens_seen": 83571340,
      "step": 3885,
      "time_per_iteration": 2.583822250366211
    },
    {
      "auxiliary_loss_clip": 0.01074645,
      "auxiliary_loss_mlp": 0.010012,
      "balance_loss_clip": 1.01950741,
      "balance_loss_mlp": 1.00006795,
      "epoch": 0.46726387302350747,
      "flos": 69369684566400.0,
      "grad_norm": 0.7899103192596285,
      "language_loss": 0.62985563,
      "learning_rate": 2.3080208914158795e-06,
      "loss": 0.65061414,
      "num_input_tokens_seen": 83634340,
      "step": 3886,
      "time_per_iteration": 3.9647345542907715
    },
    {
      "auxiliary_loss_clip": 0.01153786,
      "auxiliary_loss_mlp": 0.01026643,
      "balance_loss_clip": 1.05349183,
      "balance_loss_mlp": 1.01876879,
      "epoch": 0.4673841159141466,
      "flos": 25519666878720.0,
      "grad_norm": 2.180772374069027,
      "language_loss": 0.72144479,
      "learning_rate": 2.3072511901644753e-06,
      "loss": 0.74324906,
      "num_input_tokens_seen": 83653410,
      "step": 3887,
      "time_per_iteration": 2.528275489807129
    },
    {
      "auxiliary_loss_clip": 0.01183669,
      "auxiliary_loss_mlp": 0.01024972,
      "balance_loss_clip": 1.05625856,
      "balance_loss_mlp": 1.01689017,
      "epoch": 0.4675043588047857,
      "flos": 24499265316480.0,
      "grad_norm": 1.7702952062515918,
      "language_loss": 0.80995989,
      "learning_rate": 2.306481442303309e-06,
      "loss": 0.83204627,
      "num_input_tokens_seen": 83672985,
      "step": 3888,
      "time_per_iteration": 3.247695207595825
    },
    {
      "auxiliary_loss_clip": 0.01170851,
      "auxiliary_loss_mlp": 0.0102845,
      "balance_loss_clip": 1.05205393,
      "balance_loss_mlp": 1.01982522,
      "epoch": 0.46762460169542475,
      "flos": 20960771685120.0,
      "grad_norm": 1.885880303298299,
      "language_loss": 0.73227996,
      "learning_rate": 2.3057116479491515e-06,
      "loss": 0.75427294,
      "num_input_tokens_seen": 83692395,
      "step": 3889,
      "time_per_iteration": 2.4668185710906982
    },
    {
      "auxiliary_loss_clip": 0.01163405,
      "auxiliary_loss_mlp": 0.01029331,
      "balance_loss_clip": 1.04871678,
      "balance_loss_mlp": 1.0213145,
      "epoch": 0.46774484458606386,
      "flos": 19171666137600.0,
      "grad_norm": 1.8307758839218344,
      "language_loss": 0.76158923,
      "learning_rate": 2.30494180721878e-06,
      "loss": 0.78351659,
      "num_input_tokens_seen": 83709735,
      "step": 3890,
      "time_per_iteration": 2.437392234802246
    },
    {
      "auxiliary_loss_clip": 0.01166558,
      "auxiliary_loss_mlp": 0.01032485,
      "balance_loss_clip": 1.0514648,
      "balance_loss_mlp": 1.02476072,
      "epoch": 0.4678650874767029,
      "flos": 17967689141760.0,
      "grad_norm": 1.984678040888482,
      "language_loss": 0.89754528,
      "learning_rate": 2.3041719202289794e-06,
      "loss": 0.91953576,
      "num_input_tokens_seen": 83725910,
      "step": 3891,
      "time_per_iteration": 3.1704788208007812
    },
    {
      "auxiliary_loss_clip": 0.01168691,
      "auxiliary_loss_mlp": 0.0103224,
      "balance_loss_clip": 1.05288756,
      "balance_loss_mlp": 1.02413988,
      "epoch": 0.467985330367342,
      "flos": 21360816432000.0,
      "grad_norm": 1.664477723877608,
      "language_loss": 0.80685574,
      "learning_rate": 2.30340198709654e-06,
      "loss": 0.82886505,
      "num_input_tokens_seen": 83745745,
      "step": 3892,
      "time_per_iteration": 2.475205659866333
    },
    {
      "auxiliary_loss_clip": 0.01157089,
      "auxiliary_loss_mlp": 0.01030517,
      "balance_loss_clip": 1.04808879,
      "balance_loss_mlp": 1.02253628,
      "epoch": 0.46810557325798113,
      "flos": 20521835487360.0,
      "grad_norm": 2.178028118512589,
      "language_loss": 0.74731648,
      "learning_rate": 2.3026320079382605e-06,
      "loss": 0.76919258,
      "num_input_tokens_seen": 83762680,
      "step": 3893,
      "time_per_iteration": 2.478456974029541
    },
    {
      "auxiliary_loss_clip": 0.01181422,
      "auxiliary_loss_mlp": 0.01024631,
      "balance_loss_clip": 1.05420351,
      "balance_loss_mlp": 1.01641786,
      "epoch": 0.4682258161486202,
      "flos": 30117848572800.0,
      "grad_norm": 1.9910731750170454,
      "language_loss": 0.76259762,
      "learning_rate": 2.3018619828709454e-06,
      "loss": 0.78465819,
      "num_input_tokens_seen": 83784220,
      "step": 3894,
      "time_per_iteration": 2.4849154949188232
    },
    {
      "auxiliary_loss_clip": 0.01165536,
      "auxiliary_loss_mlp": 0.00762786,
      "balance_loss_clip": 1.05523419,
      "balance_loss_mlp": 1.00016975,
      "epoch": 0.4683460590392593,
      "flos": 25293357239040.0,
      "grad_norm": 1.941504932459736,
      "language_loss": 0.8214823,
      "learning_rate": 2.3010919120114084e-06,
      "loss": 0.84076548,
      "num_input_tokens_seen": 83800750,
      "step": 3895,
      "time_per_iteration": 2.471978187561035
    },
    {
      "auxiliary_loss_clip": 0.01162205,
      "auxiliary_loss_mlp": 0.01032129,
      "balance_loss_clip": 1.04592657,
      "balance_loss_mlp": 1.02428472,
      "epoch": 0.4684663019298984,
      "flos": 15368330551680.0,
      "grad_norm": 2.936894641281481,
      "language_loss": 0.66227627,
      "learning_rate": 2.3003217954764672e-06,
      "loss": 0.6842196,
      "num_input_tokens_seen": 83815455,
      "step": 3896,
      "time_per_iteration": 2.399125814437866
    },
    {
      "auxiliary_loss_clip": 0.01168341,
      "auxiliary_loss_mlp": 0.01026117,
      "balance_loss_clip": 1.04838705,
      "balance_loss_mlp": 1.01792777,
      "epoch": 0.46858654482053747,
      "flos": 27778842737280.0,
      "grad_norm": 2.183958799472053,
      "language_loss": 0.7964617,
      "learning_rate": 2.299551633382949e-06,
      "loss": 0.81840628,
      "num_input_tokens_seen": 83835765,
      "step": 3897,
      "time_per_iteration": 2.498365879058838
    },
    {
      "auxiliary_loss_clip": 0.01146293,
      "auxiliary_loss_mlp": 0.01029172,
      "balance_loss_clip": 1.04847121,
      "balance_loss_mlp": 1.02112532,
      "epoch": 0.4687067877111766,
      "flos": 18040623707520.0,
      "grad_norm": 1.8873209137383269,
      "language_loss": 0.85583794,
      "learning_rate": 2.2987814258476854e-06,
      "loss": 0.87759262,
      "num_input_tokens_seen": 83853565,
      "step": 3898,
      "time_per_iteration": 2.459045886993408
    },
    {
      "auxiliary_loss_clip": 0.01126318,
      "auxiliary_loss_mlp": 0.01025096,
      "balance_loss_clip": 1.04328096,
      "balance_loss_mlp": 1.01698685,
      "epoch": 0.4688270306018157,
      "flos": 16977380198400.0,
      "grad_norm": 2.3232820564551893,
      "language_loss": 0.67809123,
      "learning_rate": 2.2980111729875177e-06,
      "loss": 0.69960535,
      "num_input_tokens_seen": 83869815,
      "step": 3899,
      "time_per_iteration": 2.5148820877075195
    },
    {
      "auxiliary_loss_clip": 0.01149602,
      "auxiliary_loss_mlp": 0.01033599,
      "balance_loss_clip": 1.05162537,
      "balance_loss_mlp": 1.02554631,
      "epoch": 0.46894727349245474,
      "flos": 17821640442240.0,
      "grad_norm": 1.7038890059976342,
      "language_loss": 0.82639122,
      "learning_rate": 2.2972408749192917e-06,
      "loss": 0.84822327,
      "num_input_tokens_seen": 83887545,
      "step": 3900,
      "time_per_iteration": 2.454667568206787
    },
    {
      "auxiliary_loss_clip": 0.01167148,
      "auxiliary_loss_mlp": 0.00762295,
      "balance_loss_clip": 1.05345964,
      "balance_loss_mlp": 1.00015974,
      "epoch": 0.46906751638309385,
      "flos": 21471349559040.0,
      "grad_norm": 1.9022616262999756,
      "language_loss": 0.66984117,
      "learning_rate": 2.296470531759861e-06,
      "loss": 0.68913567,
      "num_input_tokens_seen": 83905645,
      "step": 3901,
      "time_per_iteration": 2.5472047328948975
    },
    {
      "auxiliary_loss_clip": 0.01134553,
      "auxiliary_loss_mlp": 0.01027173,
      "balance_loss_clip": 1.04632318,
      "balance_loss_mlp": 1.01845884,
      "epoch": 0.46918775927373296,
      "flos": 20337829090560.0,
      "grad_norm": 2.104826364359253,
      "language_loss": 0.79586041,
      "learning_rate": 2.2957001436260866e-06,
      "loss": 0.81747758,
      "num_input_tokens_seen": 83922705,
      "step": 3902,
      "time_per_iteration": 2.50502610206604
    },
    {
      "auxiliary_loss_clip": 0.01149945,
      "auxiliary_loss_mlp": 0.01030722,
      "balance_loss_clip": 1.04967451,
      "balance_loss_mlp": 1.02277637,
      "epoch": 0.469308002164372,
      "flos": 18403249461120.0,
      "grad_norm": 1.6900208873205684,
      "language_loss": 0.73074389,
      "learning_rate": 2.294929710634836e-06,
      "loss": 0.75255048,
      "num_input_tokens_seen": 83940795,
      "step": 3903,
      "time_per_iteration": 2.4626681804656982
    },
    {
      "auxiliary_loss_clip": 0.01163985,
      "auxiliary_loss_mlp": 0.01032766,
      "balance_loss_clip": 1.04815865,
      "balance_loss_mlp": 1.02452862,
      "epoch": 0.46942824505501113,
      "flos": 37962067363200.0,
      "grad_norm": 1.7988188607603042,
      "language_loss": 0.6076479,
      "learning_rate": 2.2941592329029823e-06,
      "loss": 0.62961543,
      "num_input_tokens_seen": 83961900,
      "step": 3904,
      "time_per_iteration": 2.6006581783294678
    },
    {
      "auxiliary_loss_clip": 0.01164936,
      "auxiliary_loss_mlp": 0.01031881,
      "balance_loss_clip": 1.05158997,
      "balance_loss_mlp": 1.02277911,
      "epoch": 0.46954848794565024,
      "flos": 21872507627520.0,
      "grad_norm": 2.013836162482629,
      "language_loss": 0.79043615,
      "learning_rate": 2.2933887105474067e-06,
      "loss": 0.81240427,
      "num_input_tokens_seen": 83980075,
      "step": 3905,
      "time_per_iteration": 2.468151807785034
    },
    {
      "auxiliary_loss_clip": 0.01165186,
      "auxiliary_loss_mlp": 0.01030691,
      "balance_loss_clip": 1.05300975,
      "balance_loss_mlp": 1.02302909,
      "epoch": 0.4696687308362893,
      "flos": 22016545165440.0,
      "grad_norm": 1.7684303629614024,
      "language_loss": 0.8133927,
      "learning_rate": 2.2926181436849974e-06,
      "loss": 0.83535147,
      "num_input_tokens_seen": 83999430,
      "step": 3906,
      "time_per_iteration": 2.4747769832611084
    },
    {
      "auxiliary_loss_clip": 0.01166751,
      "auxiliary_loss_mlp": 0.01033423,
      "balance_loss_clip": 1.05305803,
      "balance_loss_mlp": 1.02475095,
      "epoch": 0.4697889737269284,
      "flos": 21613663244160.0,
      "grad_norm": 1.7270848221075503,
      "language_loss": 0.72615135,
      "learning_rate": 2.2918475324326478e-06,
      "loss": 0.74815309,
      "num_input_tokens_seen": 84019150,
      "step": 3907,
      "time_per_iteration": 2.4720046520233154
    },
    {
      "auxiliary_loss_clip": 0.01172444,
      "auxiliary_loss_mlp": 0.00762847,
      "balance_loss_clip": 1.05482447,
      "balance_loss_mlp": 1.00018561,
      "epoch": 0.46990921661756746,
      "flos": 25228323665280.0,
      "grad_norm": 2.414298547109381,
      "language_loss": 0.9137364,
      "learning_rate": 2.2910768769072603e-06,
      "loss": 0.93308932,
      "num_input_tokens_seen": 84037930,
      "step": 3908,
      "time_per_iteration": 2.494840145111084
    },
    {
      "auxiliary_loss_clip": 0.011628,
      "auxiliary_loss_mlp": 0.01032886,
      "balance_loss_clip": 1.05139899,
      "balance_loss_mlp": 1.02514374,
      "epoch": 0.47002945950820657,
      "flos": 13844031045120.0,
      "grad_norm": 1.8640808578300783,
      "language_loss": 0.75583982,
      "learning_rate": 2.2903061772257417e-06,
      "loss": 0.77779669,
      "num_input_tokens_seen": 84055915,
      "step": 3909,
      "time_per_iteration": 2.4327032566070557
    },
    {
      "auxiliary_loss_clip": 0.01166474,
      "auxiliary_loss_mlp": 0.01030212,
      "balance_loss_clip": 1.05212092,
      "balance_loss_mlp": 1.02214789,
      "epoch": 0.4701497023988457,
      "flos": 26247001374720.0,
      "grad_norm": 1.4465889587670249,
      "language_loss": 0.78676677,
      "learning_rate": 2.289535433505007e-06,
      "loss": 0.80873364,
      "num_input_tokens_seen": 84077270,
      "step": 3910,
      "time_per_iteration": 3.34236478805542
    },
    {
      "auxiliary_loss_clip": 0.01154239,
      "auxiliary_loss_mlp": 0.01027731,
      "balance_loss_clip": 1.04852319,
      "balance_loss_mlp": 1.01959467,
      "epoch": 0.47026994528948474,
      "flos": 25629517647360.0,
      "grad_norm": 3.3325974196098884,
      "language_loss": 0.63712442,
      "learning_rate": 2.2887646458619767e-06,
      "loss": 0.65894413,
      "num_input_tokens_seen": 84098635,
      "step": 3911,
      "time_per_iteration": 2.5121309757232666
    },
    {
      "auxiliary_loss_clip": 0.0114579,
      "auxiliary_loss_mlp": 0.01033416,
      "balance_loss_clip": 1.04975271,
      "balance_loss_mlp": 1.02493477,
      "epoch": 0.47039018818012385,
      "flos": 20554406144640.0,
      "grad_norm": 1.8168180917927572,
      "language_loss": 0.76332533,
      "learning_rate": 2.2879938144135797e-06,
      "loss": 0.78511739,
      "num_input_tokens_seen": 84114740,
      "step": 3912,
      "time_per_iteration": 3.391508102416992
    },
    {
      "auxiliary_loss_clip": 0.01137282,
      "auxiliary_loss_mlp": 0.00762165,
      "balance_loss_clip": 1.04772663,
      "balance_loss_mlp": 1.0001483,
      "epoch": 0.47051043107076296,
      "flos": 21577249831680.0,
      "grad_norm": 1.566237535722384,
      "language_loss": 0.74709392,
      "learning_rate": 2.2872229392767496e-06,
      "loss": 0.76608837,
      "num_input_tokens_seen": 84134845,
      "step": 3913,
      "time_per_iteration": 2.554697036743164
    },
    {
      "auxiliary_loss_clip": 0.01173013,
      "auxiliary_loss_mlp": 0.01030438,
      "balance_loss_clip": 1.05492735,
      "balance_loss_mlp": 1.02301705,
      "epoch": 0.470630673961402,
      "flos": 18953185662720.0,
      "grad_norm": 1.6615889273849738,
      "language_loss": 0.74864447,
      "learning_rate": 2.286452020568428e-06,
      "loss": 0.770679,
      "num_input_tokens_seen": 84152920,
      "step": 3914,
      "time_per_iteration": 2.4670512676239014
    },
    {
      "auxiliary_loss_clip": 0.01187811,
      "auxiliary_loss_mlp": 0.01029877,
      "balance_loss_clip": 1.05464709,
      "balance_loss_mlp": 1.02084088,
      "epoch": 0.4707509168520411,
      "flos": 19938969492480.0,
      "grad_norm": 1.7767280007631059,
      "language_loss": 0.73110145,
      "learning_rate": 2.2856810584055637e-06,
      "loss": 0.75327832,
      "num_input_tokens_seen": 84170455,
      "step": 3915,
      "time_per_iteration": 3.1501214504241943
    },
    {
      "auxiliary_loss_clip": 0.01169901,
      "auxiliary_loss_mlp": 0.01026995,
      "balance_loss_clip": 1.05301285,
      "balance_loss_mlp": 1.01920438,
      "epoch": 0.47087115974268023,
      "flos": 40118754741120.0,
      "grad_norm": 1.4843282628754895,
      "language_loss": 0.67796385,
      "learning_rate": 2.2849100529051085e-06,
      "loss": 0.69993281,
      "num_input_tokens_seen": 84197390,
      "step": 3916,
      "time_per_iteration": 2.6483914852142334
    },
    {
      "auxiliary_loss_clip": 0.01181751,
      "auxiliary_loss_mlp": 0.01028849,
      "balance_loss_clip": 1.05486059,
      "balance_loss_mlp": 1.0206002,
      "epoch": 0.4709914026333193,
      "flos": 13552723745280.0,
      "grad_norm": 2.6199059526797646,
      "language_loss": 0.80272996,
      "learning_rate": 2.284139004184026e-06,
      "loss": 0.8248359,
      "num_input_tokens_seen": 84214620,
      "step": 3917,
      "time_per_iteration": 3.078840732574463
    },
    {
      "auxiliary_loss_clip": 0.01184876,
      "auxiliary_loss_mlp": 0.01025919,
      "balance_loss_clip": 1.05471694,
      "balance_loss_mlp": 1.01772952,
      "epoch": 0.4711116455239584,
      "flos": 19974628719360.0,
      "grad_norm": 1.9085208601447747,
      "language_loss": 0.74340433,
      "learning_rate": 2.2833679123592814e-06,
      "loss": 0.76551229,
      "num_input_tokens_seen": 84231880,
      "step": 3918,
      "time_per_iteration": 2.4177355766296387
    },
    {
      "auxiliary_loss_clip": 0.01152112,
      "auxiliary_loss_mlp": 0.01029914,
      "balance_loss_clip": 1.05053329,
      "balance_loss_mlp": 1.02140808,
      "epoch": 0.4712318884145975,
      "flos": 32124824064000.0,
      "grad_norm": 1.8913302106108993,
      "language_loss": 0.63578331,
      "learning_rate": 2.2825967775478508e-06,
      "loss": 0.65760362,
      "num_input_tokens_seen": 84252980,
      "step": 3919,
      "time_per_iteration": 2.5593221187591553
    },
    {
      "auxiliary_loss_clip": 0.01181308,
      "auxiliary_loss_mlp": 0.01029988,
      "balance_loss_clip": 1.05168343,
      "balance_loss_mlp": 1.0218457,
      "epoch": 0.47135213130523657,
      "flos": 20047850593920.0,
      "grad_norm": 1.9486405646148768,
      "language_loss": 0.83488429,
      "learning_rate": 2.2818255998667135e-06,
      "loss": 0.85699725,
      "num_input_tokens_seen": 84271490,
      "step": 3920,
      "time_per_iteration": 2.407737970352173
    },
    {
      "auxiliary_loss_clip": 0.01168589,
      "auxiliary_loss_mlp": 0.01023757,
      "balance_loss_clip": 1.05431485,
      "balance_loss_mlp": 1.01621699,
      "epoch": 0.4714723741958757,
      "flos": 19426990988160.0,
      "grad_norm": 1.67337167104691,
      "language_loss": 0.79162836,
      "learning_rate": 2.2810543794328566e-06,
      "loss": 0.81355184,
      "num_input_tokens_seen": 84290525,
      "step": 3921,
      "time_per_iteration": 2.436403751373291
    },
    {
      "auxiliary_loss_clip": 0.01171074,
      "auxiliary_loss_mlp": 0.010297,
      "balance_loss_clip": 1.0525527,
      "balance_loss_mlp": 1.02193356,
      "epoch": 0.4715926170865148,
      "flos": 20373883367040.0,
      "grad_norm": 1.6731921078568863,
      "language_loss": 0.82470763,
      "learning_rate": 2.2802831163632735e-06,
      "loss": 0.84671533,
      "num_input_tokens_seen": 84309245,
      "step": 3922,
      "time_per_iteration": 2.5063672065734863
    },
    {
      "auxiliary_loss_clip": 0.01114736,
      "auxiliary_loss_mlp": 0.01029042,
      "balance_loss_clip": 1.04751801,
      "balance_loss_mlp": 1.0205543,
      "epoch": 0.47171285997715384,
      "flos": 22672884430080.0,
      "grad_norm": 1.656064427367903,
      "language_loss": 0.74663508,
      "learning_rate": 2.279511810774965e-06,
      "loss": 0.76807296,
      "num_input_tokens_seen": 84330775,
      "step": 3923,
      "time_per_iteration": 2.5902345180511475
    },
    {
      "auxiliary_loss_clip": 0.01184181,
      "auxiliary_loss_mlp": 0.01027104,
      "balance_loss_clip": 1.05463362,
      "balance_loss_mlp": 1.01911116,
      "epoch": 0.47183310286779295,
      "flos": 21105419754240.0,
      "grad_norm": 1.9027523324417548,
      "language_loss": 0.71974021,
      "learning_rate": 2.2787404627849364e-06,
      "loss": 0.74185306,
      "num_input_tokens_seen": 84349985,
      "step": 3924,
      "time_per_iteration": 2.416029691696167
    },
    {
      "auxiliary_loss_clip": 0.01151481,
      "auxiliary_loss_mlp": 0.01027267,
      "balance_loss_clip": 1.04859531,
      "balance_loss_mlp": 1.01920867,
      "epoch": 0.471953345758432,
      "flos": 21726566668800.0,
      "grad_norm": 1.9920244565664502,
      "language_loss": 0.79076815,
      "learning_rate": 2.277969072510202e-06,
      "loss": 0.81255561,
      "num_input_tokens_seen": 84368965,
      "step": 3925,
      "time_per_iteration": 2.4915759563446045
    },
    {
      "auxiliary_loss_clip": 0.01154154,
      "auxiliary_loss_mlp": 0.01026082,
      "balance_loss_clip": 1.05168402,
      "balance_loss_mlp": 1.01827955,
      "epoch": 0.4720735886490711,
      "flos": 19861078849920.0,
      "grad_norm": 1.6432953144551201,
      "language_loss": 0.81646621,
      "learning_rate": 2.2771976400677803e-06,
      "loss": 0.83826864,
      "num_input_tokens_seen": 84387795,
      "step": 3926,
      "time_per_iteration": 2.491424083709717
    },
    {
      "auxiliary_loss_clip": 0.01115938,
      "auxiliary_loss_mlp": 0.01024031,
      "balance_loss_clip": 1.04512691,
      "balance_loss_mlp": 1.0161097,
      "epoch": 0.47219383153971023,
      "flos": 19171809792000.0,
      "grad_norm": 1.757102901117818,
      "language_loss": 0.78484219,
      "learning_rate": 2.2764261655746965e-06,
      "loss": 0.80624187,
      "num_input_tokens_seen": 84405290,
      "step": 3927,
      "time_per_iteration": 2.5622406005859375
    },
    {
      "auxiliary_loss_clip": 0.01137155,
      "auxiliary_loss_mlp": 0.01023608,
      "balance_loss_clip": 1.04924273,
      "balance_loss_mlp": 1.01570392,
      "epoch": 0.4723140744303493,
      "flos": 23224005780480.0,
      "grad_norm": 1.6525497716001576,
      "language_loss": 0.75936806,
      "learning_rate": 2.2756546491479832e-06,
      "loss": 0.78097564,
      "num_input_tokens_seen": 84426205,
      "step": 3928,
      "time_per_iteration": 2.594552993774414
    },
    {
      "auxiliary_loss_clip": 0.01183145,
      "auxiliary_loss_mlp": 0.00762574,
      "balance_loss_clip": 1.05210543,
      "balance_loss_mlp": 1.00017548,
      "epoch": 0.4724343173209884,
      "flos": 18223265387520.0,
      "grad_norm": 4.575438564586592,
      "language_loss": 0.80469501,
      "learning_rate": 2.274883090904679e-06,
      "loss": 0.82415217,
      "num_input_tokens_seen": 84443970,
      "step": 3929,
      "time_per_iteration": 2.482083320617676
    },
    {
      "auxiliary_loss_clip": 0.01185622,
      "auxiliary_loss_mlp": 0.01028069,
      "balance_loss_clip": 1.05644011,
      "balance_loss_mlp": 1.02007627,
      "epoch": 0.4725545602116275,
      "flos": 21251037490560.0,
      "grad_norm": 2.3492833442457774,
      "language_loss": 0.68133175,
      "learning_rate": 2.2741114909618283e-06,
      "loss": 0.70346868,
      "num_input_tokens_seen": 84459865,
      "step": 3930,
      "time_per_iteration": 2.428098678588867
    },
    {
      "auxiliary_loss_clip": 0.01141188,
      "auxiliary_loss_mlp": 0.01024984,
      "balance_loss_clip": 1.04971039,
      "balance_loss_mlp": 1.01655602,
      "epoch": 0.47267480310226656,
      "flos": 21434002392960.0,
      "grad_norm": 1.9010653593579911,
      "language_loss": 0.71740603,
      "learning_rate": 2.2733398494364828e-06,
      "loss": 0.73906779,
      "num_input_tokens_seen": 84479110,
      "step": 3931,
      "time_per_iteration": 2.6437652111053467
    },
    {
      "auxiliary_loss_clip": 0.01151119,
      "auxiliary_loss_mlp": 0.01027011,
      "balance_loss_clip": 1.05338371,
      "balance_loss_mlp": 1.01914883,
      "epoch": 0.47279504599290567,
      "flos": 18770508069120.0,
      "grad_norm": 1.8301577529125357,
      "language_loss": 0.84467971,
      "learning_rate": 2.272568166445699e-06,
      "loss": 0.86646104,
      "num_input_tokens_seen": 84497675,
      "step": 3932,
      "time_per_iteration": 2.4781692028045654
    },
    {
      "auxiliary_loss_clip": 0.01167848,
      "auxiliary_loss_mlp": 0.01021889,
      "balance_loss_clip": 1.05021477,
      "balance_loss_mlp": 1.01361609,
      "epoch": 0.4729152888835448,
      "flos": 21105742976640.0,
      "grad_norm": 1.9076541834854692,
      "language_loss": 0.64131486,
      "learning_rate": 2.271796442106541e-06,
      "loss": 0.6632123,
      "num_input_tokens_seen": 84517030,
      "step": 3933,
      "time_per_iteration": 2.4738810062408447
    },
    {
      "auxiliary_loss_clip": 0.01048427,
      "auxiliary_loss_mlp": 0.01004574,
      "balance_loss_clip": 1.01725721,
      "balance_loss_mlp": 1.00341153,
      "epoch": 0.47303553177418384,
      "flos": 70201877840640.0,
      "grad_norm": 0.805979241650793,
      "language_loss": 0.56495887,
      "learning_rate": 2.271024676536079e-06,
      "loss": 0.58548892,
      "num_input_tokens_seen": 84577290,
      "step": 3934,
      "time_per_iteration": 3.096956491470337
    },
    {
      "auxiliary_loss_clip": 0.01161425,
      "auxiliary_loss_mlp": 0.01029103,
      "balance_loss_clip": 1.05564463,
      "balance_loss_mlp": 1.02038836,
      "epoch": 0.47315577466482295,
      "flos": 22455122227200.0,
      "grad_norm": 2.1328800265111583,
      "language_loss": 0.73396826,
      "learning_rate": 2.2702528698513894e-06,
      "loss": 0.7558735,
      "num_input_tokens_seen": 84598415,
      "step": 3935,
      "time_per_iteration": 2.5184085369110107
    },
    {
      "auxiliary_loss_clip": 0.01153546,
      "auxiliary_loss_mlp": 0.01028118,
      "balance_loss_clip": 1.0472182,
      "balance_loss_mlp": 1.0199579,
      "epoch": 0.47327601755546206,
      "flos": 24352857480960.0,
      "grad_norm": 2.103317361248548,
      "language_loss": 0.78858531,
      "learning_rate": 2.269481022169554e-06,
      "loss": 0.81040192,
      "num_input_tokens_seen": 84617010,
      "step": 3936,
      "time_per_iteration": 2.5378787517547607
    },
    {
      "auxiliary_loss_clip": 0.01159156,
      "auxiliary_loss_mlp": 0.01026561,
      "balance_loss_clip": 1.04911327,
      "balance_loss_mlp": 1.01791883,
      "epoch": 0.4733962604461011,
      "flos": 22926772736640.0,
      "grad_norm": 1.537479449737051,
      "language_loss": 0.80446088,
      "learning_rate": 2.2687091336076614e-06,
      "loss": 0.82631803,
      "num_input_tokens_seen": 84636350,
      "step": 3937,
      "time_per_iteration": 3.367605686187744
    },
    {
      "auxiliary_loss_clip": 0.01167021,
      "auxiliary_loss_mlp": 0.01034119,
      "balance_loss_clip": 1.05243397,
      "balance_loss_mlp": 1.0260694,
      "epoch": 0.4735165033367402,
      "flos": 18327369980160.0,
      "grad_norm": 1.7593878940941947,
      "language_loss": 0.79858375,
      "learning_rate": 2.267937204282807e-06,
      "loss": 0.82059515,
      "num_input_tokens_seen": 84653490,
      "step": 3938,
      "time_per_iteration": 2.447352886199951
    },
    {
      "auxiliary_loss_clip": 0.01175472,
      "auxiliary_loss_mlp": 0.01026577,
      "balance_loss_clip": 1.05418169,
      "balance_loss_mlp": 1.01808929,
      "epoch": 0.4736367462273793,
      "flos": 23037018554880.0,
      "grad_norm": 1.8830531469526242,
      "language_loss": 0.79087877,
      "learning_rate": 2.2671652343120926e-06,
      "loss": 0.81289923,
      "num_input_tokens_seen": 84673965,
      "step": 3939,
      "time_per_iteration": 3.3205392360687256
    },
    {
      "auxiliary_loss_clip": 0.01182226,
      "auxiliary_loss_mlp": 0.01026962,
      "balance_loss_clip": 1.05430079,
      "balance_loss_mlp": 1.01934767,
      "epoch": 0.4737569891180184,
      "flos": 25374336451200.0,
      "grad_norm": 1.695863902800287,
      "language_loss": 0.80493605,
      "learning_rate": 2.2663932238126236e-06,
      "loss": 0.82702792,
      "num_input_tokens_seen": 84692525,
      "step": 3940,
      "time_per_iteration": 2.4844329357147217
    },
    {
      "auxiliary_loss_clip": 0.01168258,
      "auxiliary_loss_mlp": 0.01024126,
      "balance_loss_clip": 1.05049801,
      "balance_loss_mlp": 1.01565027,
      "epoch": 0.4738772320086575,
      "flos": 25849326925440.0,
      "grad_norm": 1.4025689413624673,
      "language_loss": 0.80538106,
      "learning_rate": 2.265621172901515e-06,
      "loss": 0.82730484,
      "num_input_tokens_seen": 84715640,
      "step": 3941,
      "time_per_iteration": 3.304386615753174
    },
    {
      "auxiliary_loss_clip": 0.01186283,
      "auxiliary_loss_mlp": 0.01032524,
      "balance_loss_clip": 1.05710828,
      "balance_loss_mlp": 1.02434611,
      "epoch": 0.47399747489929656,
      "flos": 27564420499200.0,
      "grad_norm": 2.213206429423058,
      "language_loss": 0.71631354,
      "learning_rate": 2.2648490816958854e-06,
      "loss": 0.73850167,
      "num_input_tokens_seen": 84736635,
      "step": 3942,
      "time_per_iteration": 2.50063157081604
    },
    {
      "auxiliary_loss_clip": 0.01168264,
      "auxiliary_loss_mlp": 0.01031198,
      "balance_loss_clip": 1.05015123,
      "balance_loss_mlp": 1.02203131,
      "epoch": 0.47411771778993567,
      "flos": 24863650836480.0,
      "grad_norm": 2.296572266514965,
      "language_loss": 0.73123455,
      "learning_rate": 2.264076950312861e-06,
      "loss": 0.75322914,
      "num_input_tokens_seen": 84755445,
      "step": 3943,
      "time_per_iteration": 2.4793343544006348
    },
    {
      "auxiliary_loss_clip": 0.01159549,
      "auxiliary_loss_mlp": 0.01033708,
      "balance_loss_clip": 1.05114079,
      "balance_loss_mlp": 1.02481544,
      "epoch": 0.4742379606805748,
      "flos": 22748009725440.0,
      "grad_norm": 1.9875040261026784,
      "language_loss": 0.82636118,
      "learning_rate": 2.2633047788695727e-06,
      "loss": 0.84829378,
      "num_input_tokens_seen": 84775750,
      "step": 3944,
      "time_per_iteration": 3.277961254119873
    },
    {
      "auxiliary_loss_clip": 0.01153429,
      "auxiliary_loss_mlp": 0.01028802,
      "balance_loss_clip": 1.05159068,
      "balance_loss_mlp": 1.02140808,
      "epoch": 0.47435820357121383,
      "flos": 19681130689920.0,
      "grad_norm": 1.8106678933938967,
      "language_loss": 0.64074922,
      "learning_rate": 2.262532567483159e-06,
      "loss": 0.66257155,
      "num_input_tokens_seen": 84794310,
      "step": 3945,
      "time_per_iteration": 2.4848594665527344
    },
    {
      "auxiliary_loss_clip": 0.01187705,
      "auxiliary_loss_mlp": 0.00762947,
      "balance_loss_clip": 1.05689502,
      "balance_loss_mlp": 1.00022018,
      "epoch": 0.47447844646185294,
      "flos": 25228718714880.0,
      "grad_norm": 1.953219185358619,
      "language_loss": 0.8039096,
      "learning_rate": 2.2617603162707635e-06,
      "loss": 0.82341611,
      "num_input_tokens_seen": 84814720,
      "step": 3946,
      "time_per_iteration": 2.459381341934204
    },
    {
      "auxiliary_loss_clip": 0.01182574,
      "auxiliary_loss_mlp": 0.0102756,
      "balance_loss_clip": 1.05416536,
      "balance_loss_mlp": 1.01962066,
      "epoch": 0.47459868935249205,
      "flos": 24570619683840.0,
      "grad_norm": 1.7823515575382127,
      "language_loss": 0.82610595,
      "learning_rate": 2.2609880253495363e-06,
      "loss": 0.84820735,
      "num_input_tokens_seen": 84834355,
      "step": 3947,
      "time_per_iteration": 2.469965696334839
    },
    {
      "auxiliary_loss_clip": 0.01149464,
      "auxiliary_loss_mlp": 0.01032572,
      "balance_loss_clip": 1.04830933,
      "balance_loss_mlp": 1.02450728,
      "epoch": 0.4747189322431311,
      "flos": 20558500295040.0,
      "grad_norm": 1.9899514402168406,
      "language_loss": 0.86516255,
      "learning_rate": 2.260215694836633e-06,
      "loss": 0.88698292,
      "num_input_tokens_seen": 84853530,
      "step": 3948,
      "time_per_iteration": 2.5163052082061768
    },
    {
      "auxiliary_loss_clip": 0.01127495,
      "auxiliary_loss_mlp": 0.00762548,
      "balance_loss_clip": 1.04538691,
      "balance_loss_mlp": 1.00017416,
      "epoch": 0.4748391751337702,
      "flos": 25995231970560.0,
      "grad_norm": 2.1232131002539476,
      "language_loss": 0.65029842,
      "learning_rate": 2.2594433248492157e-06,
      "loss": 0.66919887,
      "num_input_tokens_seen": 84872505,
      "step": 3949,
      "time_per_iteration": 2.6297905445098877
    },
    {
      "auxiliary_loss_clip": 0.01172745,
      "auxiliary_loss_mlp": 0.01031224,
      "balance_loss_clip": 1.05143785,
      "balance_loss_mlp": 1.02367806,
      "epoch": 0.47495941802440933,
      "flos": 22821052032000.0,
      "grad_norm": 1.7018345763394311,
      "language_loss": 0.80039042,
      "learning_rate": 2.2586709155044527e-06,
      "loss": 0.82243013,
      "num_input_tokens_seen": 84893105,
      "step": 3950,
      "time_per_iteration": 2.485506057739258
    },
    {
      "auxiliary_loss_clip": 0.01184137,
      "auxiliary_loss_mlp": 0.0102735,
      "balance_loss_clip": 1.05480194,
      "balance_loss_mlp": 1.01877904,
      "epoch": 0.4750796609150484,
      "flos": 27891782075520.0,
      "grad_norm": 1.535165517223909,
      "language_loss": 0.75993085,
      "learning_rate": 2.2578984669195167e-06,
      "loss": 0.78204578,
      "num_input_tokens_seen": 84914070,
      "step": 3951,
      "time_per_iteration": 2.4928951263427734
    },
    {
      "auxiliary_loss_clip": 0.01164282,
      "auxiliary_loss_mlp": 0.01027623,
      "balance_loss_clip": 1.04787517,
      "balance_loss_mlp": 1.01997006,
      "epoch": 0.4751999038056875,
      "flos": 35660085471360.0,
      "grad_norm": 1.7078249981280178,
      "language_loss": 0.68011773,
      "learning_rate": 2.2571259792115887e-06,
      "loss": 0.7020368,
      "num_input_tokens_seen": 84935290,
      "step": 3952,
      "time_per_iteration": 2.6143741607666016
    },
    {
      "auxiliary_loss_clip": 0.01162092,
      "auxiliary_loss_mlp": 0.01027994,
      "balance_loss_clip": 1.04994595,
      "balance_loss_mlp": 1.02075469,
      "epoch": 0.4753201466963266,
      "flos": 22090880361600.0,
      "grad_norm": 2.010348380700964,
      "language_loss": 0.79343057,
      "learning_rate": 2.2563534524978544e-06,
      "loss": 0.81533134,
      "num_input_tokens_seen": 84952760,
      "step": 3953,
      "time_per_iteration": 2.4709224700927734
    },
    {
      "auxiliary_loss_clip": 0.0113746,
      "auxiliary_loss_mlp": 0.01025891,
      "balance_loss_clip": 1.05321193,
      "balance_loss_mlp": 1.01849985,
      "epoch": 0.47544038958696566,
      "flos": 30190854965760.0,
      "grad_norm": 3.4154588355327014,
      "language_loss": 0.70507705,
      "learning_rate": 2.2555808868955052e-06,
      "loss": 0.72671056,
      "num_input_tokens_seen": 84974890,
      "step": 3954,
      "time_per_iteration": 2.616994619369507
    },
    {
      "auxiliary_loss_clip": 0.01125724,
      "auxiliary_loss_mlp": 0.01027405,
      "balance_loss_clip": 1.04717135,
      "balance_loss_mlp": 1.01856577,
      "epoch": 0.47556063247760477,
      "flos": 23472219738240.0,
      "grad_norm": 2.242850593187867,
      "language_loss": 0.73732722,
      "learning_rate": 2.254808282521738e-06,
      "loss": 0.7588585,
      "num_input_tokens_seen": 84993640,
      "step": 3955,
      "time_per_iteration": 2.5972847938537598
    },
    {
      "auxiliary_loss_clip": 0.01142271,
      "auxiliary_loss_mlp": 0.00762986,
      "balance_loss_clip": 1.04797673,
      "balance_loss_mlp": 1.00015473,
      "epoch": 0.4756808753682438,
      "flos": 25155209531520.0,
      "grad_norm": 1.688456891365973,
      "language_loss": 0.80921721,
      "learning_rate": 2.2540356394937573e-06,
      "loss": 0.82826972,
      "num_input_tokens_seen": 85012340,
      "step": 3956,
      "time_per_iteration": 2.560209274291992
    },
    {
      "auxiliary_loss_clip": 0.01145335,
      "auxiliary_loss_mlp": 0.01026819,
      "balance_loss_clip": 1.04867911,
      "balance_loss_mlp": 1.01842046,
      "epoch": 0.47580111825888294,
      "flos": 15669729573120.0,
      "grad_norm": 2.5840713815426923,
      "language_loss": 0.83982933,
      "learning_rate": 2.253262957928772e-06,
      "loss": 0.86155081,
      "num_input_tokens_seen": 85029225,
      "step": 3957,
      "time_per_iteration": 2.498357057571411
    },
    {
      "auxiliary_loss_clip": 0.01146383,
      "auxiliary_loss_mlp": 0.01027263,
      "balance_loss_clip": 1.0465858,
      "balance_loss_mlp": 1.01910937,
      "epoch": 0.47592136114952205,
      "flos": 17636556637440.0,
      "grad_norm": 1.6481076950228488,
      "language_loss": 0.71979707,
      "learning_rate": 2.2524902379439976e-06,
      "loss": 0.74153352,
      "num_input_tokens_seen": 85047895,
      "step": 3958,
      "time_per_iteration": 2.4685184955596924
    },
    {
      "auxiliary_loss_clip": 0.01036937,
      "auxiliary_loss_mlp": 0.01015124,
      "balance_loss_clip": 1.03191209,
      "balance_loss_mlp": 1.01313341,
      "epoch": 0.4760416040401611,
      "flos": 61417159292160.0,
      "grad_norm": 0.7416526265400596,
      "language_loss": 0.63709635,
      "learning_rate": 2.251717479656655e-06,
      "loss": 0.65761691,
      "num_input_tokens_seen": 85112690,
      "step": 3959,
      "time_per_iteration": 3.199294328689575
    },
    {
      "auxiliary_loss_clip": 0.01183416,
      "auxiliary_loss_mlp": 0.01030478,
      "balance_loss_clip": 1.05317748,
      "balance_loss_mlp": 1.02187753,
      "epoch": 0.4761618469308002,
      "flos": 18405871153920.0,
      "grad_norm": 2.5720710813834837,
      "language_loss": 0.76329547,
      "learning_rate": 2.2509446831839704e-06,
      "loss": 0.78543437,
      "num_input_tokens_seen": 85132130,
      "step": 3960,
      "time_per_iteration": 2.4176218509674072
    },
    {
      "auxiliary_loss_clip": 0.01156009,
      "auxiliary_loss_mlp": 0.01033162,
      "balance_loss_clip": 1.04874122,
      "balance_loss_mlp": 1.02462065,
      "epoch": 0.4762820898214393,
      "flos": 18040911016320.0,
      "grad_norm": 2.3312981433200153,
      "language_loss": 0.82654268,
      "learning_rate": 2.250171848643177e-06,
      "loss": 0.84843439,
      "num_input_tokens_seen": 85149420,
      "step": 3961,
      "time_per_iteration": 2.463630199432373
    },
    {
      "auxiliary_loss_clip": 0.01150583,
      "auxiliary_loss_mlp": 0.01033925,
      "balance_loss_clip": 1.05071545,
      "balance_loss_mlp": 1.02666855,
      "epoch": 0.4764023327120784,
      "flos": 19318253541120.0,
      "grad_norm": 1.7757762507366792,
      "language_loss": 0.86164963,
      "learning_rate": 2.249398976151513e-06,
      "loss": 0.88349473,
      "num_input_tokens_seen": 85166970,
      "step": 3962,
      "time_per_iteration": 2.5673716068267822
    },
    {
      "auxiliary_loss_clip": 0.0118074,
      "auxiliary_loss_mlp": 0.01031746,
      "balance_loss_clip": 1.05287385,
      "balance_loss_mlp": 1.02404189,
      "epoch": 0.4765225756027175,
      "flos": 22747255539840.0,
      "grad_norm": 2.269308122831583,
      "language_loss": 0.78704858,
      "learning_rate": 2.248626065826223e-06,
      "loss": 0.80917341,
      "num_input_tokens_seen": 85185175,
      "step": 3963,
      "time_per_iteration": 3.270092248916626
    },
    {
      "auxiliary_loss_clip": 0.01085144,
      "auxiliary_loss_mlp": 0.01003451,
      "balance_loss_clip": 1.02123535,
      "balance_loss_mlp": 1.00227058,
      "epoch": 0.4766428184933566,
      "flos": 65933392106880.0,
      "grad_norm": 0.7628748756633362,
      "language_loss": 0.62553471,
      "learning_rate": 2.2478531177845564e-06,
      "loss": 0.64642066,
      "num_input_tokens_seen": 85246170,
      "step": 3964,
      "time_per_iteration": 2.9979188442230225
    },
    {
      "auxiliary_loss_clip": 0.01154793,
      "auxiliary_loss_mlp": 0.01023169,
      "balance_loss_clip": 1.04986763,
      "balance_loss_mlp": 1.01571274,
      "epoch": 0.47676306138399566,
      "flos": 24136495908480.0,
      "grad_norm": 2.2676695959364475,
      "language_loss": 0.84941137,
      "learning_rate": 2.247080132143769e-06,
      "loss": 0.87119102,
      "num_input_tokens_seen": 85268525,
      "step": 3965,
      "time_per_iteration": 2.5312819480895996
    },
    {
      "auxiliary_loss_clip": 0.01137259,
      "auxiliary_loss_mlp": 0.01029202,
      "balance_loss_clip": 1.04306388,
      "balance_loss_mlp": 1.0206604,
      "epoch": 0.47688330427463477,
      "flos": 12604322995200.0,
      "grad_norm": 1.946599524660357,
      "language_loss": 0.69220626,
      "learning_rate": 2.246307109021121e-06,
      "loss": 0.71387088,
      "num_input_tokens_seen": 85285930,
      "step": 3966,
      "time_per_iteration": 3.349525213241577
    },
    {
      "auxiliary_loss_clip": 0.01150135,
      "auxiliary_loss_mlp": 0.0102937,
      "balance_loss_clip": 1.04738307,
      "balance_loss_mlp": 1.02122211,
      "epoch": 0.4770035471652739,
      "flos": 21390585828480.0,
      "grad_norm": 1.8210990834071763,
      "language_loss": 0.82020861,
      "learning_rate": 2.2455340485338817e-06,
      "loss": 0.8420037,
      "num_input_tokens_seen": 85303565,
      "step": 3967,
      "time_per_iteration": 2.5074305534362793
    },
    {
      "auxiliary_loss_clip": 0.01167821,
      "auxiliary_loss_mlp": 0.01027496,
      "balance_loss_clip": 1.05049324,
      "balance_loss_mlp": 1.019449,
      "epoch": 0.47712379005591293,
      "flos": 25156251025920.0,
      "grad_norm": 2.2546258739980893,
      "language_loss": 0.67999876,
      "learning_rate": 2.244760950799322e-06,
      "loss": 0.70195192,
      "num_input_tokens_seen": 85321835,
      "step": 3968,
      "time_per_iteration": 3.3435447216033936
    },
    {
      "auxiliary_loss_clip": 0.01124292,
      "auxiliary_loss_mlp": 0.01029178,
      "balance_loss_clip": 1.04623389,
      "balance_loss_mlp": 1.02103591,
      "epoch": 0.47724403294655204,
      "flos": 22054323294720.0,
      "grad_norm": 1.7402937587528589,
      "language_loss": 0.72403592,
      "learning_rate": 2.2439878159347203e-06,
      "loss": 0.74557066,
      "num_input_tokens_seen": 85341260,
      "step": 3969,
      "time_per_iteration": 2.5409762859344482
    },
    {
      "auxiliary_loss_clip": 0.01084001,
      "auxiliary_loss_mlp": 0.01001324,
      "balance_loss_clip": 1.01954353,
      "balance_loss_mlp": 1.00013757,
      "epoch": 0.4773642758371911,
      "flos": 70229387658240.0,
      "grad_norm": 0.7336429076439863,
      "language_loss": 0.55572265,
      "learning_rate": 2.2432146440573616e-06,
      "loss": 0.57657593,
      "num_input_tokens_seen": 85407220,
      "step": 3970,
      "time_per_iteration": 3.126549243927002
    },
    {
      "auxiliary_loss_clip": 0.0115302,
      "auxiliary_loss_mlp": 0.01027214,
      "balance_loss_clip": 1.05038774,
      "balance_loss_mlp": 1.01932216,
      "epoch": 0.4774845187278302,
      "flos": 23548602009600.0,
      "grad_norm": 1.8888335085994998,
      "language_loss": 0.66403818,
      "learning_rate": 2.242441435284534e-06,
      "loss": 0.68584049,
      "num_input_tokens_seen": 85426095,
      "step": 3971,
      "time_per_iteration": 3.262711524963379
    },
    {
      "auxiliary_loss_clip": 0.01171728,
      "auxiliary_loss_mlp": 0.01032068,
      "balance_loss_clip": 1.05407298,
      "balance_loss_mlp": 1.02297819,
      "epoch": 0.4776047616184693,
      "flos": 23075371301760.0,
      "grad_norm": 2.1976847190672597,
      "language_loss": 0.85757387,
      "learning_rate": 2.2416681897335337e-06,
      "loss": 0.87961185,
      "num_input_tokens_seen": 85444245,
      "step": 3972,
      "time_per_iteration": 2.455953598022461
    },
    {
      "auxiliary_loss_clip": 0.0112803,
      "auxiliary_loss_mlp": 0.01034323,
      "balance_loss_clip": 1.05024052,
      "balance_loss_mlp": 1.02555537,
      "epoch": 0.4777250045091084,
      "flos": 31898119374720.0,
      "grad_norm": 1.776389658067525,
      "language_loss": 0.67022085,
      "learning_rate": 2.240894907521661e-06,
      "loss": 0.69184434,
      "num_input_tokens_seen": 85463325,
      "step": 3973,
      "time_per_iteration": 2.6557347774505615
    },
    {
      "auxiliary_loss_clip": 0.01151655,
      "auxiliary_loss_mlp": 0.01027543,
      "balance_loss_clip": 1.04840493,
      "balance_loss_mlp": 1.01963377,
      "epoch": 0.4778452473997475,
      "flos": 24278163148800.0,
      "grad_norm": 1.761736449021304,
      "language_loss": 0.64108646,
      "learning_rate": 2.240121588766223e-06,
      "loss": 0.66287845,
      "num_input_tokens_seen": 85483375,
      "step": 3974,
      "time_per_iteration": 2.5389761924743652
    },
    {
      "auxiliary_loss_clip": 0.01146781,
      "auxiliary_loss_mlp": 0.01027696,
      "balance_loss_clip": 1.04763746,
      "balance_loss_mlp": 1.01957822,
      "epoch": 0.4779654902903866,
      "flos": 31575031516800.0,
      "grad_norm": 1.632525483683521,
      "language_loss": 0.71173179,
      "learning_rate": 2.239348233584531e-06,
      "loss": 0.73347658,
      "num_input_tokens_seen": 85504230,
      "step": 3975,
      "time_per_iteration": 2.5691754817962646
    },
    {
      "auxiliary_loss_clip": 0.01168319,
      "auxiliary_loss_mlp": 0.01034031,
      "balance_loss_clip": 1.05092502,
      "balance_loss_mlp": 1.02532268,
      "epoch": 0.47808573318102565,
      "flos": 19500428344320.0,
      "grad_norm": 1.9025432427835756,
      "language_loss": 0.81269556,
      "learning_rate": 2.2385748420939013e-06,
      "loss": 0.83471906,
      "num_input_tokens_seen": 85523425,
      "step": 3976,
      "time_per_iteration": 2.4381492137908936
    },
    {
      "auxiliary_loss_clip": 0.0118344,
      "auxiliary_loss_mlp": 0.01029721,
      "balance_loss_clip": 1.05749369,
      "balance_loss_mlp": 1.02167475,
      "epoch": 0.47820597607166476,
      "flos": 22601135013120.0,
      "grad_norm": 1.8277971858326207,
      "language_loss": 0.7250663,
      "learning_rate": 2.2378014144116583e-06,
      "loss": 0.74719793,
      "num_input_tokens_seen": 85542235,
      "step": 3977,
      "time_per_iteration": 2.418173313140869
    },
    {
      "auxiliary_loss_clip": 0.01186584,
      "auxiliary_loss_mlp": 0.01034421,
      "balance_loss_clip": 1.0548079,
      "balance_loss_mlp": 1.02673841,
      "epoch": 0.4783262189623039,
      "flos": 23003011353600.0,
      "grad_norm": 1.858051524971096,
      "language_loss": 0.79761928,
      "learning_rate": 2.23702795065513e-06,
      "loss": 0.81982934,
      "num_input_tokens_seen": 85561815,
      "step": 3978,
      "time_per_iteration": 2.4411394596099854
    },
    {
      "auxiliary_loss_clip": 0.01074899,
      "auxiliary_loss_mlp": 0.01001818,
      "balance_loss_clip": 1.02020633,
      "balance_loss_mlp": 1.00062561,
      "epoch": 0.47844646185294293,
      "flos": 49772801226240.0,
      "grad_norm": 0.9939924251529546,
      "language_loss": 0.67503238,
      "learning_rate": 2.2362544509416493e-06,
      "loss": 0.69579953,
      "num_input_tokens_seen": 85613930,
      "step": 3979,
      "time_per_iteration": 2.9299569129943848
    },
    {
      "auxiliary_loss_clip": 0.01144554,
      "auxiliary_loss_mlp": 0.01034241,
      "balance_loss_clip": 1.04686737,
      "balance_loss_mlp": 1.02612305,
      "epoch": 0.47856670474358204,
      "flos": 20229558520320.0,
      "grad_norm": 2.1673715058317016,
      "language_loss": 0.82476485,
      "learning_rate": 2.2354809153885572e-06,
      "loss": 0.84655285,
      "num_input_tokens_seen": 85631000,
      "step": 3980,
      "time_per_iteration": 2.48453688621521
    },
    {
      "auxiliary_loss_clip": 0.01166358,
      "auxiliary_loss_mlp": 0.0103113,
      "balance_loss_clip": 1.04986358,
      "balance_loss_mlp": 1.02284467,
      "epoch": 0.47868694763422115,
      "flos": 20990936131200.0,
      "grad_norm": 1.6820024842968728,
      "language_loss": 0.83080387,
      "learning_rate": 2.234707344113197e-06,
      "loss": 0.85277867,
      "num_input_tokens_seen": 85649095,
      "step": 3981,
      "time_per_iteration": 2.435882329940796
    },
    {
      "auxiliary_loss_clip": 0.01177533,
      "auxiliary_loss_mlp": 0.01025285,
      "balance_loss_clip": 1.05177999,
      "balance_loss_mlp": 1.01770902,
      "epoch": 0.4788071905248602,
      "flos": 19026551191680.0,
      "grad_norm": 2.2898127098641003,
      "language_loss": 0.77545059,
      "learning_rate": 2.233933737232919e-06,
      "loss": 0.7974788,
      "num_input_tokens_seen": 85666875,
      "step": 3982,
      "time_per_iteration": 2.4090487957000732
    },
    {
      "auxiliary_loss_clip": 0.01116574,
      "auxiliary_loss_mlp": 0.00762428,
      "balance_loss_clip": 1.04493928,
      "balance_loss_mlp": 1.00005317,
      "epoch": 0.4789274334154993,
      "flos": 23002221254400.0,
      "grad_norm": 1.8953805438065205,
      "language_loss": 0.78105259,
      "learning_rate": 2.2331600948650793e-06,
      "loss": 0.7998426,
      "num_input_tokens_seen": 85687020,
      "step": 3983,
      "time_per_iteration": 2.545619487762451
    },
    {
      "auxiliary_loss_clip": 0.01129713,
      "auxiliary_loss_mlp": 0.00763618,
      "balance_loss_clip": 1.04945779,
      "balance_loss_mlp": 1.00006223,
      "epoch": 0.4790476763061384,
      "flos": 23075586783360.0,
      "grad_norm": 1.4386954195740431,
      "language_loss": 0.80384129,
      "learning_rate": 2.2323864171270386e-06,
      "loss": 0.82277459,
      "num_input_tokens_seen": 85708290,
      "step": 3984,
      "time_per_iteration": 2.549447536468506
    },
    {
      "auxiliary_loss_clip": 0.01141761,
      "auxiliary_loss_mlp": 0.01029303,
      "balance_loss_clip": 1.04670978,
      "balance_loss_mlp": 1.0204339,
      "epoch": 0.4791679191967775,
      "flos": 21179288073600.0,
      "grad_norm": 1.7739503564224635,
      "language_loss": 0.72467792,
      "learning_rate": 2.231612704136164e-06,
      "loss": 0.74638855,
      "num_input_tokens_seen": 85728660,
      "step": 3985,
      "time_per_iteration": 2.5680487155914307
    },
    {
      "auxiliary_loss_clip": 0.01163455,
      "auxiliary_loss_mlp": 0.01029451,
      "balance_loss_clip": 1.04987931,
      "balance_loss_mlp": 1.02105272,
      "epoch": 0.4792881620874166,
      "flos": 22301495758080.0,
      "grad_norm": 2.430686904060973,
      "language_loss": 0.74837899,
      "learning_rate": 2.2308389560098253e-06,
      "loss": 0.77030802,
      "num_input_tokens_seen": 85745035,
      "step": 3986,
      "time_per_iteration": 2.4403202533721924
    },
    {
      "auxiliary_loss_clip": 0.01146213,
      "auxiliary_loss_mlp": 0.01028561,
      "balance_loss_clip": 1.05261624,
      "balance_loss_mlp": 1.01976955,
      "epoch": 0.47940840497805565,
      "flos": 17420877423360.0,
      "grad_norm": 3.8204621774113923,
      "language_loss": 0.77177101,
      "learning_rate": 2.2300651728654008e-06,
      "loss": 0.79351878,
      "num_input_tokens_seen": 85760295,
      "step": 3987,
      "time_per_iteration": 2.511549949645996
    },
    {
      "auxiliary_loss_clip": 0.01065503,
      "auxiliary_loss_mlp": 0.00753057,
      "balance_loss_clip": 1.0169965,
      "balance_loss_mlp": 0.9999547,
      "epoch": 0.47952864786869476,
      "flos": 65358175708800.0,
      "grad_norm": 0.7327198027798575,
      "language_loss": 0.60208541,
      "learning_rate": 2.229291354820272e-06,
      "loss": 0.62027103,
      "num_input_tokens_seen": 85821305,
      "step": 3988,
      "time_per_iteration": 3.0782713890075684
    },
    {
      "auxiliary_loss_clip": 0.01164959,
      "auxiliary_loss_mlp": 0.0103044,
      "balance_loss_clip": 1.04941332,
      "balance_loss_mlp": 1.02195215,
      "epoch": 0.47964889075933387,
      "flos": 16799802336000.0,
      "grad_norm": 1.9265582463770996,
      "language_loss": 0.76056862,
      "learning_rate": 2.228517501991828e-06,
      "loss": 0.78252256,
      "num_input_tokens_seen": 85840105,
      "step": 3989,
      "time_per_iteration": 2.4439468383789062
    },
    {
      "auxiliary_loss_clip": 0.01057014,
      "auxiliary_loss_mlp": 0.01003261,
      "balance_loss_clip": 1.01804996,
      "balance_loss_mlp": 1.0021286,
      "epoch": 0.4797691336499729,
      "flos": 70079244808320.0,
      "grad_norm": 0.8269018489069816,
      "language_loss": 0.6105842,
      "learning_rate": 2.22774361449746e-06,
      "loss": 0.6311869,
      "num_input_tokens_seen": 85896585,
      "step": 3990,
      "time_per_iteration": 3.928361654281616
    },
    {
      "auxiliary_loss_clip": 0.0111311,
      "auxiliary_loss_mlp": 0.01027864,
      "balance_loss_clip": 1.04853439,
      "balance_loss_mlp": 1.01973951,
      "epoch": 0.47988937654061203,
      "flos": 18953329317120.0,
      "grad_norm": 2.865706420917079,
      "language_loss": 0.70823878,
      "learning_rate": 2.2269696924545668e-06,
      "loss": 0.72964853,
      "num_input_tokens_seen": 85914415,
      "step": 3991,
      "time_per_iteration": 2.58437442779541
    },
    {
      "auxiliary_loss_clip": 0.01142927,
      "auxiliary_loss_mlp": 0.01029191,
      "balance_loss_clip": 1.05286992,
      "balance_loss_mlp": 1.02127528,
      "epoch": 0.48000961943125114,
      "flos": 14461981649280.0,
      "grad_norm": 2.3733513305385134,
      "language_loss": 0.78378046,
      "learning_rate": 2.2261957359805523e-06,
      "loss": 0.80550164,
      "num_input_tokens_seen": 85931650,
      "step": 3992,
      "time_per_iteration": 2.5114989280700684
    },
    {
      "auxiliary_loss_clip": 0.01182526,
      "auxiliary_loss_mlp": 0.01023449,
      "balance_loss_clip": 1.05332077,
      "balance_loss_mlp": 1.01557255,
      "epoch": 0.4801298623218902,
      "flos": 27051149105280.0,
      "grad_norm": 2.7940978832993286,
      "language_loss": 0.74187064,
      "learning_rate": 2.225421745192823e-06,
      "loss": 0.76393032,
      "num_input_tokens_seen": 85951805,
      "step": 3993,
      "time_per_iteration": 3.326186180114746
    },
    {
      "auxiliary_loss_clip": 0.01166701,
      "auxiliary_loss_mlp": 0.01029029,
      "balance_loss_clip": 1.05206537,
      "balance_loss_mlp": 1.02068472,
      "epoch": 0.4802501052125293,
      "flos": 26355236031360.0,
      "grad_norm": 2.1713753448950857,
      "language_loss": 0.78271371,
      "learning_rate": 2.2246477202087955e-06,
      "loss": 0.80467105,
      "num_input_tokens_seen": 85972485,
      "step": 3994,
      "time_per_iteration": 2.505716562271118
    },
    {
      "auxiliary_loss_clip": 0.01155824,
      "auxiliary_loss_mlp": 0.01031157,
      "balance_loss_clip": 1.05045068,
      "balance_loss_mlp": 1.02380204,
      "epoch": 0.4803703481031684,
      "flos": 20993916960000.0,
      "grad_norm": 1.5563073175802402,
      "language_loss": 0.8277812,
      "learning_rate": 2.223873661145887e-06,
      "loss": 0.84965098,
      "num_input_tokens_seen": 85992540,
      "step": 3995,
      "time_per_iteration": 3.3509137630462646
    },
    {
      "auxiliary_loss_clip": 0.01155156,
      "auxiliary_loss_mlp": 0.00762901,
      "balance_loss_clip": 1.05538583,
      "balance_loss_mlp": 1.00001454,
      "epoch": 0.4804905909938075,
      "flos": 20703722981760.0,
      "grad_norm": 1.5493262984513207,
      "language_loss": 0.71310496,
      "learning_rate": 2.2230995681215226e-06,
      "loss": 0.73228556,
      "num_input_tokens_seen": 86012065,
      "step": 3996,
      "time_per_iteration": 2.5059707164764404
    },
    {
      "auxiliary_loss_clip": 0.01138284,
      "auxiliary_loss_mlp": 0.0102645,
      "balance_loss_clip": 1.05118155,
      "balance_loss_mlp": 1.01886284,
      "epoch": 0.4806108338844466,
      "flos": 16654831044480.0,
      "grad_norm": 2.3737159898476636,
      "language_loss": 0.77964944,
      "learning_rate": 2.2223254412531305e-06,
      "loss": 0.80129671,
      "num_input_tokens_seen": 86029435,
      "step": 3997,
      "time_per_iteration": 3.2671427726745605
    },
    {
      "auxiliary_loss_clip": 0.01138616,
      "auxiliary_loss_mlp": 0.01025982,
      "balance_loss_clip": 1.04543066,
      "balance_loss_mlp": 1.01820016,
      "epoch": 0.4807310767750857,
      "flos": 20011329440640.0,
      "grad_norm": 1.8537446652298992,
      "language_loss": 0.82521695,
      "learning_rate": 2.221551280658146e-06,
      "loss": 0.84686291,
      "num_input_tokens_seen": 86048495,
      "step": 3998,
      "time_per_iteration": 2.47847056388855
    },
    {
      "auxiliary_loss_clip": 0.01122415,
      "auxiliary_loss_mlp": 0.01025565,
      "balance_loss_clip": 1.0481379,
      "balance_loss_mlp": 1.01785791,
      "epoch": 0.48085131966572475,
      "flos": 23185257984000.0,
      "grad_norm": 1.6348183843005455,
      "language_loss": 0.7434231,
      "learning_rate": 2.2207770864540085e-06,
      "loss": 0.76490295,
      "num_input_tokens_seen": 86067470,
      "step": 3999,
      "time_per_iteration": 2.564797878265381
    },
    {
      "auxiliary_loss_clip": 0.01146218,
      "auxiliary_loss_mlp": 0.01029356,
      "balance_loss_clip": 1.0497179,
      "balance_loss_mlp": 1.02116036,
      "epoch": 0.48097156255636386,
      "flos": 20558643949440.0,
      "grad_norm": 2.026055341939582,
      "language_loss": 0.73038745,
      "learning_rate": 2.220002858758162e-06,
      "loss": 0.75214314,
      "num_input_tokens_seen": 86085460,
      "step": 4000,
      "time_per_iteration": 2.4935972690582275
    },
    {
      "auxiliary_loss_clip": 0.01071435,
      "auxiliary_loss_mlp": 0.0100087,
      "balance_loss_clip": 1.01699233,
      "balance_loss_mlp": 0.99974364,
      "epoch": 0.481091805447003,
      "flos": 70511608817280.0,
      "grad_norm": 0.9024897626198788,
      "language_loss": 0.60884935,
      "learning_rate": 2.2192285976880573e-06,
      "loss": 0.62957239,
      "num_input_tokens_seen": 86149715,
      "step": 4001,
      "time_per_iteration": 3.0582528114318848
    },
    {
      "auxiliary_loss_clip": 0.01143968,
      "auxiliary_loss_mlp": 0.00762087,
      "balance_loss_clip": 1.04902232,
      "balance_loss_mlp": 1.00003994,
      "epoch": 0.48121204833764203,
      "flos": 36428214839040.0,
      "grad_norm": 1.5436701532378354,
      "language_loss": 0.80716383,
      "learning_rate": 2.2184543033611485e-06,
      "loss": 0.82622439,
      "num_input_tokens_seen": 86170795,
      "step": 4002,
      "time_per_iteration": 2.66536283493042
    },
    {
      "auxiliary_loss_clip": 0.01169806,
      "auxiliary_loss_mlp": 0.01030919,
      "balance_loss_clip": 1.05230081,
      "balance_loss_mlp": 1.02324176,
      "epoch": 0.48133229122828114,
      "flos": 27490264871040.0,
      "grad_norm": 2.255906294504358,
      "language_loss": 0.8202607,
      "learning_rate": 2.2176799758948957e-06,
      "loss": 0.84226793,
      "num_input_tokens_seen": 86190955,
      "step": 4003,
      "time_per_iteration": 2.516401529312134
    },
    {
      "auxiliary_loss_clip": 0.01149143,
      "auxiliary_loss_mlp": 0.01031889,
      "balance_loss_clip": 1.05103505,
      "balance_loss_mlp": 1.02409852,
      "epoch": 0.4814525341189202,
      "flos": 43072802179200.0,
      "grad_norm": 1.8202217279820037,
      "language_loss": 0.73234135,
      "learning_rate": 2.2169056154067635e-06,
      "loss": 0.7541517,
      "num_input_tokens_seen": 86214875,
      "step": 4004,
      "time_per_iteration": 2.7078208923339844
    },
    {
      "auxiliary_loss_clip": 0.01171497,
      "auxiliary_loss_mlp": 0.00762869,
      "balance_loss_clip": 1.05524826,
      "balance_loss_mlp": 1.00001323,
      "epoch": 0.4815727770095593,
      "flos": 24236901400320.0,
      "grad_norm": 1.8220416222679847,
      "language_loss": 0.82599258,
      "learning_rate": 2.216131222014222e-06,
      "loss": 0.84533632,
      "num_input_tokens_seen": 86232950,
      "step": 4005,
      "time_per_iteration": 2.5012128353118896
    },
    {
      "auxiliary_loss_clip": 0.01135469,
      "auxiliary_loss_mlp": 0.01032463,
      "balance_loss_clip": 1.0492537,
      "balance_loss_mlp": 1.02397561,
      "epoch": 0.4816930199001984,
      "flos": 18113630100480.0,
      "grad_norm": 2.1702239270510395,
      "language_loss": 0.79992437,
      "learning_rate": 2.2153567958347455e-06,
      "loss": 0.82160366,
      "num_input_tokens_seen": 86249160,
      "step": 4006,
      "time_per_iteration": 2.4960286617279053
    },
    {
      "auxiliary_loss_clip": 0.0115401,
      "auxiliary_loss_mlp": 0.0102627,
      "balance_loss_clip": 1.05311728,
      "balance_loss_mlp": 1.01799643,
      "epoch": 0.48181326279083747,
      "flos": 17274720983040.0,
      "grad_norm": 2.036974211627786,
      "language_loss": 0.80058861,
      "learning_rate": 2.214582336985815e-06,
      "loss": 0.82239139,
      "num_input_tokens_seen": 86267060,
      "step": 4007,
      "time_per_iteration": 2.470130205154419
    },
    {
      "auxiliary_loss_clip": 0.01144592,
      "auxiliary_loss_mlp": 0.01029109,
      "balance_loss_clip": 1.04900789,
      "balance_loss_mlp": 1.02032948,
      "epoch": 0.4819335056814766,
      "flos": 14903252231040.0,
      "grad_norm": 2.197255688775241,
      "language_loss": 0.66410673,
      "learning_rate": 2.2138078455849142e-06,
      "loss": 0.68584371,
      "num_input_tokens_seen": 86285055,
      "step": 4008,
      "time_per_iteration": 2.46886944770813
    },
    {
      "auxiliary_loss_clip": 0.01175495,
      "auxiliary_loss_mlp": 0.01029232,
      "balance_loss_clip": 1.05457318,
      "balance_loss_mlp": 1.02145362,
      "epoch": 0.4820537485721157,
      "flos": 19244888012160.0,
      "grad_norm": 2.0553792454781026,
      "language_loss": 0.78387129,
      "learning_rate": 2.2130333217495334e-06,
      "loss": 0.80591857,
      "num_input_tokens_seen": 86304225,
      "step": 4009,
      "time_per_iteration": 2.4619975090026855
    },
    {
      "auxiliary_loss_clip": 0.01150172,
      "auxiliary_loss_mlp": 0.01027808,
      "balance_loss_clip": 1.05039942,
      "balance_loss_mlp": 1.01979709,
      "epoch": 0.48217399146275475,
      "flos": 16033791870720.0,
      "grad_norm": 3.1064920092466664,
      "language_loss": 0.67348146,
      "learning_rate": 2.2122587655971665e-06,
      "loss": 0.69526124,
      "num_input_tokens_seen": 86319170,
      "step": 4010,
      "time_per_iteration": 2.5164031982421875
    },
    {
      "auxiliary_loss_clip": 0.01154258,
      "auxiliary_loss_mlp": 0.01028917,
      "balance_loss_clip": 1.05053461,
      "balance_loss_mlp": 1.02138257,
      "epoch": 0.48229423435339386,
      "flos": 24134197438080.0,
      "grad_norm": 1.6605789856705067,
      "language_loss": 0.6407125,
      "learning_rate": 2.211484177245314e-06,
      "loss": 0.66254425,
      "num_input_tokens_seen": 86338760,
      "step": 4011,
      "time_per_iteration": 2.512725353240967
    },
    {
      "auxiliary_loss_clip": 0.01185934,
      "auxiliary_loss_mlp": 0.01032464,
      "balance_loss_clip": 1.05541587,
      "balance_loss_mlp": 1.02424407,
      "epoch": 0.48241447724403297,
      "flos": 23805435231360.0,
      "grad_norm": 1.9295960021644893,
      "language_loss": 0.71834379,
      "learning_rate": 2.21070955681148e-06,
      "loss": 0.74052781,
      "num_input_tokens_seen": 86357865,
      "step": 4012,
      "time_per_iteration": 2.449613332748413
    },
    {
      "auxiliary_loss_clip": 0.01132562,
      "auxiliary_loss_mlp": 0.0102813,
      "balance_loss_clip": 1.04869962,
      "balance_loss_mlp": 1.02030432,
      "epoch": 0.482534720134672,
      "flos": 23110312256640.0,
      "grad_norm": 1.5838853825164083,
      "language_loss": 0.77872318,
      "learning_rate": 2.209934904413174e-06,
      "loss": 0.8003301,
      "num_input_tokens_seen": 86379470,
      "step": 4013,
      "time_per_iteration": 2.5532708168029785
    },
    {
      "auxiliary_loss_clip": 0.01104976,
      "auxiliary_loss_mlp": 0.0102861,
      "balance_loss_clip": 1.03781271,
      "balance_loss_mlp": 1.01998472,
      "epoch": 0.48265496302531113,
      "flos": 20923819568640.0,
      "grad_norm": 1.8990438432396708,
      "language_loss": 0.71644902,
      "learning_rate": 2.2091602201679095e-06,
      "loss": 0.73778486,
      "num_input_tokens_seen": 86399080,
      "step": 4014,
      "time_per_iteration": 2.612545967102051
    },
    {
      "auxiliary_loss_clip": 0.01146133,
      "auxiliary_loss_mlp": 0.01025539,
      "balance_loss_clip": 1.05163813,
      "balance_loss_mlp": 1.01771331,
      "epoch": 0.48277520591595025,
      "flos": 15231152511360.0,
      "grad_norm": 2.346585112815046,
      "language_loss": 0.83471477,
      "learning_rate": 2.208385504193206e-06,
      "loss": 0.85643142,
      "num_input_tokens_seen": 86416580,
      "step": 4015,
      "time_per_iteration": 2.585113763809204
    },
    {
      "auxiliary_loss_clip": 0.0118359,
      "auxiliary_loss_mlp": 0.01020524,
      "balance_loss_clip": 1.0540421,
      "balance_loss_mlp": 1.01265633,
      "epoch": 0.4828954488065893,
      "flos": 17858664385920.0,
      "grad_norm": 1.8973272530851735,
      "language_loss": 0.81183368,
      "learning_rate": 2.2076107566065873e-06,
      "loss": 0.83387482,
      "num_input_tokens_seen": 86434365,
      "step": 4016,
      "time_per_iteration": 2.4063713550567627
    },
    {
      "auxiliary_loss_clip": 0.01174544,
      "auxiliary_loss_mlp": 0.01031973,
      "balance_loss_clip": 1.05557704,
      "balance_loss_mlp": 1.02471304,
      "epoch": 0.4830156916972284,
      "flos": 32087405070720.0,
      "grad_norm": 2.1320415127706593,
      "language_loss": 0.75556278,
      "learning_rate": 2.2068359775255816e-06,
      "loss": 0.77762794,
      "num_input_tokens_seen": 86452675,
      "step": 4017,
      "time_per_iteration": 3.383261203765869
    },
    {
      "auxiliary_loss_clip": 0.01121257,
      "auxiliary_loss_mlp": 0.01030258,
      "balance_loss_clip": 1.04743946,
      "balance_loss_mlp": 1.02215123,
      "epoch": 0.48313593458786747,
      "flos": 21871717528320.0,
      "grad_norm": 2.789894383423264,
      "language_loss": 0.77914602,
      "learning_rate": 2.206061167067723e-06,
      "loss": 0.80066115,
      "num_input_tokens_seen": 86470785,
      "step": 4018,
      "time_per_iteration": 2.5599255561828613
    },
    {
      "auxiliary_loss_clip": 0.01137514,
      "auxiliary_loss_mlp": 0.01026552,
      "balance_loss_clip": 1.04612303,
      "balance_loss_mlp": 1.01775455,
      "epoch": 0.4832561774785066,
      "flos": 22601206840320.0,
      "grad_norm": 1.9769122285139347,
      "language_loss": 0.79428267,
      "learning_rate": 2.205286325350549e-06,
      "loss": 0.81592333,
      "num_input_tokens_seen": 86489850,
      "step": 4019,
      "time_per_iteration": 3.403686761856079
    },
    {
      "auxiliary_loss_clip": 0.01126783,
      "auxiliary_loss_mlp": 0.01028771,
      "balance_loss_clip": 1.04873157,
      "balance_loss_mlp": 1.0213325,
      "epoch": 0.4833764203691457,
      "flos": 13437342282240.0,
      "grad_norm": 4.350399558320729,
      "language_loss": 0.72452468,
      "learning_rate": 2.204511452491603e-06,
      "loss": 0.74608022,
      "num_input_tokens_seen": 86506475,
      "step": 4020,
      "time_per_iteration": 2.5112051963806152
    },
    {
      "auxiliary_loss_clip": 0.01182201,
      "auxiliary_loss_mlp": 0.0103085,
      "balance_loss_clip": 1.05754459,
      "balance_loss_mlp": 1.02304745,
      "epoch": 0.48349666325978474,
      "flos": 44128036955520.0,
      "grad_norm": 1.7958611125963242,
      "language_loss": 0.74959695,
      "learning_rate": 2.2037365486084316e-06,
      "loss": 0.77172744,
      "num_input_tokens_seen": 86529715,
      "step": 4021,
      "time_per_iteration": 3.460017442703247
    },
    {
      "auxiliary_loss_clip": 0.01150004,
      "auxiliary_loss_mlp": 0.01030309,
      "balance_loss_clip": 1.04870439,
      "balance_loss_mlp": 1.02190423,
      "epoch": 0.48361690615042385,
      "flos": 26028377245440.0,
      "grad_norm": 2.422226814400363,
      "language_loss": 0.78060877,
      "learning_rate": 2.2029616138185886e-06,
      "loss": 0.80241197,
      "num_input_tokens_seen": 86548715,
      "step": 4022,
      "time_per_iteration": 2.5632619857788086
    },
    {
      "auxiliary_loss_clip": 0.01143809,
      "auxiliary_loss_mlp": 0.01030575,
      "balance_loss_clip": 1.05586326,
      "balance_loss_mlp": 1.02272809,
      "epoch": 0.48373714904106296,
      "flos": 22273306560000.0,
      "grad_norm": 1.6361565300408787,
      "language_loss": 0.82443738,
      "learning_rate": 2.202186648239629e-06,
      "loss": 0.84618121,
      "num_input_tokens_seen": 86568650,
      "step": 4023,
      "time_per_iteration": 2.533316135406494
    },
    {
      "auxiliary_loss_clip": 0.01169299,
      "auxiliary_loss_mlp": 0.01027875,
      "balance_loss_clip": 1.05483937,
      "balance_loss_mlp": 1.02019751,
      "epoch": 0.483857391931702,
      "flos": 28292293699200.0,
      "grad_norm": 1.7159135744511294,
      "language_loss": 0.71730506,
      "learning_rate": 2.201411651989117e-06,
      "loss": 0.73927683,
      "num_input_tokens_seen": 86590630,
      "step": 4024,
      "time_per_iteration": 3.3014490604400635
    },
    {
      "auxiliary_loss_clip": 0.01157861,
      "auxiliary_loss_mlp": 0.00762423,
      "balance_loss_clip": 1.05555797,
      "balance_loss_mlp": 1.00001144,
      "epoch": 0.48397763482234113,
      "flos": 27418048577280.0,
      "grad_norm": 1.9426368383318275,
      "language_loss": 0.7784096,
      "learning_rate": 2.2006366251846167e-06,
      "loss": 0.79761243,
      "num_input_tokens_seen": 86611270,
      "step": 4025,
      "time_per_iteration": 2.5484061241149902
    },
    {
      "auxiliary_loss_clip": 0.01156138,
      "auxiliary_loss_mlp": 0.0103026,
      "balance_loss_clip": 1.05429506,
      "balance_loss_mlp": 1.02299428,
      "epoch": 0.48409787771298024,
      "flos": 16797252470400.0,
      "grad_norm": 1.7070869882399438,
      "language_loss": 0.75565135,
      "learning_rate": 2.1998615679436997e-06,
      "loss": 0.77751541,
      "num_input_tokens_seen": 86628810,
      "step": 4026,
      "time_per_iteration": 2.4667627811431885
    },
    {
      "auxiliary_loss_clip": 0.01165313,
      "auxiliary_loss_mlp": 0.01030019,
      "balance_loss_clip": 1.05495644,
      "balance_loss_mlp": 1.0214715,
      "epoch": 0.4842181206036193,
      "flos": 25083496028160.0,
      "grad_norm": 2.0833561273461214,
      "language_loss": 0.77035522,
      "learning_rate": 2.199086480383942e-06,
      "loss": 0.79230851,
      "num_input_tokens_seen": 86648185,
      "step": 4027,
      "time_per_iteration": 2.5388615131378174
    },
    {
      "auxiliary_loss_clip": 0.01169003,
      "auxiliary_loss_mlp": 0.0103441,
      "balance_loss_clip": 1.05447793,
      "balance_loss_mlp": 1.02546382,
      "epoch": 0.4843383634942584,
      "flos": 30372311496960.0,
      "grad_norm": 3.164689330864456,
      "language_loss": 0.67496449,
      "learning_rate": 2.1983113626229234e-06,
      "loss": 0.6969986,
      "num_input_tokens_seen": 86667435,
      "step": 4028,
      "time_per_iteration": 2.5717720985412598
    },
    {
      "auxiliary_loss_clip": 0.01137424,
      "auxiliary_loss_mlp": 0.0076264,
      "balance_loss_clip": 1.0501976,
      "balance_loss_mlp": 1.00003588,
      "epoch": 0.4844586063848975,
      "flos": 20413564917120.0,
      "grad_norm": 1.791005216862981,
      "language_loss": 0.78418124,
      "learning_rate": 2.1975362147782293e-06,
      "loss": 0.80318189,
      "num_input_tokens_seen": 86686630,
      "step": 4029,
      "time_per_iteration": 2.5636374950408936
    },
    {
      "auxiliary_loss_clip": 0.010733,
      "auxiliary_loss_mlp": 0.01002223,
      "balance_loss_clip": 1.03510737,
      "balance_loss_mlp": 1.00051832,
      "epoch": 0.48457884927553657,
      "flos": 70303722854400.0,
      "grad_norm": 0.693168743604907,
      "language_loss": 0.54143679,
      "learning_rate": 2.196761036967448e-06,
      "loss": 0.56219196,
      "num_input_tokens_seen": 86754595,
      "step": 4030,
      "time_per_iteration": 3.2325029373168945
    },
    {
      "auxiliary_loss_clip": 0.01165098,
      "auxiliary_loss_mlp": 0.0102481,
      "balance_loss_clip": 1.05240369,
      "balance_loss_mlp": 1.01714468,
      "epoch": 0.4846990921661757,
      "flos": 19934516206080.0,
      "grad_norm": 1.615163937148822,
      "language_loss": 0.77574557,
      "learning_rate": 2.1959858293081743e-06,
      "loss": 0.79764467,
      "num_input_tokens_seen": 86773730,
      "step": 4031,
      "time_per_iteration": 2.475933790206909
    },
    {
      "auxiliary_loss_clip": 0.01139956,
      "auxiliary_loss_mlp": 0.0102794,
      "balance_loss_clip": 1.05192852,
      "balance_loss_mlp": 1.01986337,
      "epoch": 0.4848193350568148,
      "flos": 23075945919360.0,
      "grad_norm": 1.6198280526942903,
      "language_loss": 0.76113212,
      "learning_rate": 2.1952105919180056e-06,
      "loss": 0.78281111,
      "num_input_tokens_seen": 86792985,
      "step": 4032,
      "time_per_iteration": 2.5688636302948
    },
    {
      "auxiliary_loss_clip": 0.01156093,
      "auxiliary_loss_mlp": 0.0102073,
      "balance_loss_clip": 1.05360687,
      "balance_loss_mlp": 1.01236784,
      "epoch": 0.48493957794745385,
      "flos": 22455481363200.0,
      "grad_norm": 2.257211362095846,
      "language_loss": 0.67835373,
      "learning_rate": 2.1944353249145456e-06,
      "loss": 0.700122,
      "num_input_tokens_seen": 86812095,
      "step": 4033,
      "time_per_iteration": 2.550600528717041
    },
    {
      "auxiliary_loss_clip": 0.01185952,
      "auxiliary_loss_mlp": 0.01030916,
      "balance_loss_clip": 1.05857873,
      "balance_loss_mlp": 1.02329826,
      "epoch": 0.48505982083809296,
      "flos": 25046112948480.0,
      "grad_norm": 1.453879475665238,
      "language_loss": 0.74531084,
      "learning_rate": 2.193660028415401e-06,
      "loss": 0.76747954,
      "num_input_tokens_seen": 86832875,
      "step": 4034,
      "time_per_iteration": 2.4884185791015625
    },
    {
      "auxiliary_loss_clip": 0.01147805,
      "auxiliary_loss_mlp": 0.01023538,
      "balance_loss_clip": 1.05176902,
      "balance_loss_mlp": 1.01554513,
      "epoch": 0.485180063728732,
      "flos": 26761386090240.0,
      "grad_norm": 2.288873035078123,
      "language_loss": 0.82473969,
      "learning_rate": 2.1928847025381852e-06,
      "loss": 0.84645307,
      "num_input_tokens_seen": 86853480,
      "step": 4035,
      "time_per_iteration": 2.567490577697754
    },
    {
      "auxiliary_loss_clip": 0.01168067,
      "auxiliary_loss_mlp": 0.01027018,
      "balance_loss_clip": 1.05062187,
      "balance_loss_mlp": 1.01819611,
      "epoch": 0.4853003066193711,
      "flos": 24059143969920.0,
      "grad_norm": 1.647676469350279,
      "language_loss": 0.83809036,
      "learning_rate": 2.192109347400512e-06,
      "loss": 0.8600412,
      "num_input_tokens_seen": 86873695,
      "step": 4036,
      "time_per_iteration": 2.5026042461395264
    },
    {
      "auxiliary_loss_clip": 0.01158305,
      "auxiliary_loss_mlp": 0.01032987,
      "balance_loss_clip": 1.05186319,
      "balance_loss_mlp": 1.02411163,
      "epoch": 0.48542054951001024,
      "flos": 23076376882560.0,
      "grad_norm": 1.6434095893706393,
      "language_loss": 0.79051471,
      "learning_rate": 2.191333963120004e-06,
      "loss": 0.81242764,
      "num_input_tokens_seen": 86892675,
      "step": 4037,
      "time_per_iteration": 2.514356851577759
    },
    {
      "auxiliary_loss_clip": 0.01158109,
      "auxiliary_loss_mlp": 0.01032314,
      "balance_loss_clip": 1.05376327,
      "balance_loss_mlp": 1.02360606,
      "epoch": 0.4855407924006493,
      "flos": 25664889565440.0,
      "grad_norm": 2.4253763732347147,
      "language_loss": 0.70460498,
      "learning_rate": 2.190558549814286e-06,
      "loss": 0.72650921,
      "num_input_tokens_seen": 86912835,
      "step": 4038,
      "time_per_iteration": 2.537055492401123
    },
    {
      "auxiliary_loss_clip": 0.01154999,
      "auxiliary_loss_mlp": 0.01026686,
      "balance_loss_clip": 1.05221272,
      "balance_loss_mlp": 1.01881254,
      "epoch": 0.4856610352912884,
      "flos": 23987933256960.0,
      "grad_norm": 1.7292565221305087,
      "language_loss": 0.79308236,
      "learning_rate": 2.1897831076009872e-06,
      "loss": 0.81489921,
      "num_input_tokens_seen": 86932475,
      "step": 4039,
      "time_per_iteration": 2.512850761413574
    },
    {
      "auxiliary_loss_clip": 0.01170373,
      "auxiliary_loss_mlp": 0.01025949,
      "balance_loss_clip": 1.05404365,
      "balance_loss_mlp": 1.01807523,
      "epoch": 0.4857812781819275,
      "flos": 24096814358400.0,
      "grad_norm": 1.613813208040834,
      "language_loss": 0.80333507,
      "learning_rate": 2.1890076365977426e-06,
      "loss": 0.82529831,
      "num_input_tokens_seen": 86952300,
      "step": 4040,
      "time_per_iteration": 2.4895386695861816
    },
    {
      "auxiliary_loss_clip": 0.01056137,
      "auxiliary_loss_mlp": 0.01007984,
      "balance_loss_clip": 1.01821899,
      "balance_loss_mlp": 1.00675011,
      "epoch": 0.48590152107256657,
      "flos": 56266635185280.0,
      "grad_norm": 0.8551880458968137,
      "language_loss": 0.52817249,
      "learning_rate": 2.188232136922189e-06,
      "loss": 0.5488137,
      "num_input_tokens_seen": 87010420,
      "step": 4041,
      "time_per_iteration": 3.0015077590942383
    },
    {
      "auxiliary_loss_clip": 0.01111436,
      "auxiliary_loss_mlp": 0.01030993,
      "balance_loss_clip": 1.04953575,
      "balance_loss_mlp": 1.0224216,
      "epoch": 0.4860217639632057,
      "flos": 20046988667520.0,
      "grad_norm": 1.843162427098352,
      "language_loss": 0.75431663,
      "learning_rate": 2.187456608691971e-06,
      "loss": 0.77574092,
      "num_input_tokens_seen": 87029295,
      "step": 4042,
      "time_per_iteration": 2.5970029830932617
    },
    {
      "auxiliary_loss_clip": 0.01149956,
      "auxiliary_loss_mlp": 0.01032229,
      "balance_loss_clip": 1.05697215,
      "balance_loss_mlp": 1.02371144,
      "epoch": 0.4861420068538448,
      "flos": 17822143232640.0,
      "grad_norm": 1.8753014925101452,
      "language_loss": 0.87677664,
      "learning_rate": 2.1866810520247334e-06,
      "loss": 0.89859855,
      "num_input_tokens_seen": 87048165,
      "step": 4043,
      "time_per_iteration": 2.5197627544403076
    },
    {
      "auxiliary_loss_clip": 0.01174997,
      "auxiliary_loss_mlp": 0.01024559,
      "balance_loss_clip": 1.05332828,
      "balance_loss_mlp": 1.01584458,
      "epoch": 0.48626224974448384,
      "flos": 26250125857920.0,
      "grad_norm": 2.3554491165112736,
      "language_loss": 0.64988488,
      "learning_rate": 2.185905467038129e-06,
      "loss": 0.67188048,
      "num_input_tokens_seen": 87067070,
      "step": 4044,
      "time_per_iteration": 3.413688898086548
    },
    {
      "auxiliary_loss_clip": 0.01183121,
      "auxiliary_loss_mlp": 0.01025501,
      "balance_loss_clip": 1.05732,
      "balance_loss_mlp": 1.01768053,
      "epoch": 0.48638249263512295,
      "flos": 22054502862720.0,
      "grad_norm": 2.0463745882145945,
      "language_loss": 0.77482259,
      "learning_rate": 2.1851298538498127e-06,
      "loss": 0.7969088,
      "num_input_tokens_seen": 87086785,
      "step": 4045,
      "time_per_iteration": 2.4491310119628906
    },
    {
      "auxiliary_loss_clip": 0.01177968,
      "auxiliary_loss_mlp": 0.00763345,
      "balance_loss_clip": 1.05739164,
      "balance_loss_mlp": 1.00016522,
      "epoch": 0.48650273552576206,
      "flos": 25119945354240.0,
      "grad_norm": 1.8469257868097886,
      "language_loss": 0.79962122,
      "learning_rate": 2.184354212577446e-06,
      "loss": 0.81903434,
      "num_input_tokens_seen": 87107090,
      "step": 4046,
      "time_per_iteration": 3.3428757190704346
    },
    {
      "auxiliary_loss_clip": 0.01187616,
      "auxiliary_loss_mlp": 0.01026972,
      "balance_loss_clip": 1.05630851,
      "balance_loss_mlp": 1.01869869,
      "epoch": 0.4866229784164011,
      "flos": 17456931699840.0,
      "grad_norm": 4.218166042504802,
      "language_loss": 0.62811887,
      "learning_rate": 2.1835785433386907e-06,
      "loss": 0.65026474,
      "num_input_tokens_seen": 87125905,
      "step": 4047,
      "time_per_iteration": 3.2227940559387207
    },
    {
      "auxiliary_loss_clip": 0.01132377,
      "auxiliary_loss_mlp": 0.01031397,
      "balance_loss_clip": 1.05056989,
      "balance_loss_mlp": 1.02305818,
      "epoch": 0.48674322130704023,
      "flos": 23331127115520.0,
      "grad_norm": 1.8853798710415162,
      "language_loss": 0.65365708,
      "learning_rate": 2.182802846251216e-06,
      "loss": 0.67529476,
      "num_input_tokens_seen": 87146175,
      "step": 4048,
      "time_per_iteration": 2.538496494293213
    },
    {
      "auxiliary_loss_clip": 0.01147126,
      "auxiliary_loss_mlp": 0.0102621,
      "balance_loss_clip": 1.05112743,
      "balance_loss_mlp": 1.01804423,
      "epoch": 0.4868634641976793,
      "flos": 28804344030720.0,
      "grad_norm": 2.2938470278100827,
      "language_loss": 0.72392857,
      "learning_rate": 2.182027121432696e-06,
      "loss": 0.74566191,
      "num_input_tokens_seen": 87166800,
      "step": 4049,
      "time_per_iteration": 2.5950021743774414
    },
    {
      "auxiliary_loss_clip": 0.01188154,
      "auxiliary_loss_mlp": 0.01030456,
      "balance_loss_clip": 1.05713403,
      "balance_loss_mlp": 1.02131259,
      "epoch": 0.4869837070883184,
      "flos": 19025976574080.0,
      "grad_norm": 1.7345402287166836,
      "language_loss": 0.82433403,
      "learning_rate": 2.1812513690008054e-06,
      "loss": 0.84652013,
      "num_input_tokens_seen": 87185920,
      "step": 4050,
      "time_per_iteration": 2.4374701976776123
    },
    {
      "auxiliary_loss_clip": 0.01178251,
      "auxiliary_loss_mlp": 0.01028711,
      "balance_loss_clip": 1.05663967,
      "balance_loss_mlp": 1.01998472,
      "epoch": 0.4871039499789575,
      "flos": 15121409483520.0,
      "grad_norm": 2.0933243825728116,
      "language_loss": 0.79741073,
      "learning_rate": 2.180475589073227e-06,
      "loss": 0.8194803,
      "num_input_tokens_seen": 87203620,
      "step": 4051,
      "time_per_iteration": 3.2058520317077637
    },
    {
      "auxiliary_loss_clip": 0.0116048,
      "auxiliary_loss_mlp": 0.01024475,
      "balance_loss_clip": 1.05248952,
      "balance_loss_mlp": 1.01639235,
      "epoch": 0.48722419286959656,
      "flos": 26174066808960.0,
      "grad_norm": 1.59038909181476,
      "language_loss": 0.73389429,
      "learning_rate": 2.1796997817676456e-06,
      "loss": 0.75574374,
      "num_input_tokens_seen": 87224630,
      "step": 4052,
      "time_per_iteration": 2.5167155265808105
    },
    {
      "auxiliary_loss_clip": 0.01172781,
      "auxiliary_loss_mlp": 0.00762206,
      "balance_loss_clip": 1.0552392,
      "balance_loss_mlp": 1.00008011,
      "epoch": 0.4873444357602357,
      "flos": 24026142349440.0,
      "grad_norm": 1.5389986628754269,
      "language_loss": 0.67266291,
      "learning_rate": 2.1789239472017494e-06,
      "loss": 0.69201279,
      "num_input_tokens_seen": 87246280,
      "step": 4053,
      "time_per_iteration": 2.5113210678100586
    },
    {
      "auxiliary_loss_clip": 0.0114349,
      "auxiliary_loss_mlp": 0.01033206,
      "balance_loss_clip": 1.05210698,
      "balance_loss_mlp": 1.02468824,
      "epoch": 0.4874646786508748,
      "flos": 22820441500800.0,
      "grad_norm": 1.9762733860961847,
      "language_loss": 0.7297858,
      "learning_rate": 2.1781480854932326e-06,
      "loss": 0.75155276,
      "num_input_tokens_seen": 87266045,
      "step": 4054,
      "time_per_iteration": 2.5399374961853027
    },
    {
      "auxiliary_loss_clip": 0.01126496,
      "auxiliary_loss_mlp": 0.0103157,
      "balance_loss_clip": 1.05019116,
      "balance_loss_mlp": 1.02375579,
      "epoch": 0.48758492154151384,
      "flos": 21287594557440.0,
      "grad_norm": 1.8853190002157212,
      "language_loss": 0.79476541,
      "learning_rate": 2.1773721967597933e-06,
      "loss": 0.81634605,
      "num_input_tokens_seen": 87284495,
      "step": 4055,
      "time_per_iteration": 2.597148895263672
    },
    {
      "auxiliary_loss_clip": 0.01049061,
      "auxiliary_loss_mlp": 0.01001406,
      "balance_loss_clip": 1.01509893,
      "balance_loss_mlp": 1.00019598,
      "epoch": 0.48770516443215295,
      "flos": 62244109180800.0,
      "grad_norm": 0.8504286983227286,
      "language_loss": 0.57372952,
      "learning_rate": 2.1765962811191322e-06,
      "loss": 0.59423423,
      "num_input_tokens_seen": 87338960,
      "step": 4056,
      "time_per_iteration": 2.999025344848633
    },
    {
      "auxiliary_loss_clip": 0.01037147,
      "auxiliary_loss_mlp": 0.0100351,
      "balance_loss_clip": 1.02121603,
      "balance_loss_mlp": 1.00238371,
      "epoch": 0.48782540732279206,
      "flos": 66133451882880.0,
      "grad_norm": 0.8236176291206496,
      "language_loss": 0.62008166,
      "learning_rate": 2.1758203386889566e-06,
      "loss": 0.64048827,
      "num_input_tokens_seen": 87401730,
      "step": 4057,
      "time_per_iteration": 3.179783821105957
    },
    {
      "auxiliary_loss_clip": 0.01140356,
      "auxiliary_loss_mlp": 0.00762801,
      "balance_loss_clip": 1.04939401,
      "balance_loss_mlp": 1.00007892,
      "epoch": 0.4879456502134311,
      "flos": 14607922608000.0,
      "grad_norm": 2.185603282625709,
      "language_loss": 0.84332621,
      "learning_rate": 2.1750443695869746e-06,
      "loss": 0.8623578,
      "num_input_tokens_seen": 87417300,
      "step": 4058,
      "time_per_iteration": 2.5328805446624756
    },
    {
      "auxiliary_loss_clip": 0.01172297,
      "auxiliary_loss_mlp": 0.01029766,
      "balance_loss_clip": 1.05412626,
      "balance_loss_mlp": 1.02167809,
      "epoch": 0.4880658931040702,
      "flos": 19500464257920.0,
      "grad_norm": 2.0124907170253756,
      "language_loss": 0.85885882,
      "learning_rate": 2.174268373930901e-06,
      "loss": 0.88087946,
      "num_input_tokens_seen": 87434815,
      "step": 4059,
      "time_per_iteration": 2.495298385620117
    },
    {
      "auxiliary_loss_clip": 0.01137662,
      "auxiliary_loss_mlp": 0.00763332,
      "balance_loss_clip": 1.05353796,
      "balance_loss_mlp": 1.00006366,
      "epoch": 0.48818613599470934,
      "flos": 16723060928640.0,
      "grad_norm": 2.077818884417373,
      "language_loss": 0.79981762,
      "learning_rate": 2.1734923518384537e-06,
      "loss": 0.81882757,
      "num_input_tokens_seen": 87451420,
      "step": 4060,
      "time_per_iteration": 2.490832805633545
    },
    {
      "auxiliary_loss_clip": 0.01127164,
      "auxiliary_loss_mlp": 0.0103271,
      "balance_loss_clip": 1.05108762,
      "balance_loss_mlp": 1.02489567,
      "epoch": 0.4883063788853484,
      "flos": 26756932803840.0,
      "grad_norm": 1.7850121293091334,
      "language_loss": 0.82266873,
      "learning_rate": 2.1727163034273547e-06,
      "loss": 0.84426749,
      "num_input_tokens_seen": 87469585,
      "step": 4061,
      "time_per_iteration": 2.587679624557495
    },
    {
      "auxiliary_loss_clip": 0.01172929,
      "auxiliary_loss_mlp": 0.01024303,
      "balance_loss_clip": 1.05345392,
      "balance_loss_mlp": 1.01585746,
      "epoch": 0.4884266217759875,
      "flos": 16763388923520.0,
      "grad_norm": 2.4614690520949862,
      "language_loss": 0.79285216,
      "learning_rate": 2.17194022881533e-06,
      "loss": 0.81482452,
      "num_input_tokens_seen": 87485675,
      "step": 4062,
      "time_per_iteration": 2.4265897274017334
    },
    {
      "auxiliary_loss_clip": 0.01158876,
      "auxiliary_loss_mlp": 0.01032421,
      "balance_loss_clip": 1.05140376,
      "balance_loss_mlp": 1.02390909,
      "epoch": 0.4885468646666266,
      "flos": 24207132003840.0,
      "grad_norm": 6.143432545327328,
      "language_loss": 0.67689884,
      "learning_rate": 2.1711641281201092e-06,
      "loss": 0.69881183,
      "num_input_tokens_seen": 87505605,
      "step": 4063,
      "time_per_iteration": 2.533766508102417
    },
    {
      "auxiliary_loss_clip": 0.01168889,
      "auxiliary_loss_mlp": 0.01026654,
      "balance_loss_clip": 1.05443549,
      "balance_loss_mlp": 1.01848829,
      "epoch": 0.48866710755726567,
      "flos": 14610795696000.0,
      "grad_norm": 2.0452451851030835,
      "language_loss": 0.79497457,
      "learning_rate": 2.1703880014594264e-06,
      "loss": 0.81693,
      "num_input_tokens_seen": 87523195,
      "step": 4064,
      "time_per_iteration": 2.443293333053589
    },
    {
      "auxiliary_loss_clip": 0.01124908,
      "auxiliary_loss_mlp": 0.01031485,
      "balance_loss_clip": 1.05432439,
      "balance_loss_mlp": 1.02397454,
      "epoch": 0.4887873504479048,
      "flos": 28804451771520.0,
      "grad_norm": 2.367919336367914,
      "language_loss": 0.73787099,
      "learning_rate": 2.1696118489510182e-06,
      "loss": 0.75943494,
      "num_input_tokens_seen": 87544125,
      "step": 4065,
      "time_per_iteration": 2.6290082931518555
    },
    {
      "auxiliary_loss_clip": 0.01147638,
      "auxiliary_loss_mlp": 0.00763423,
      "balance_loss_clip": 1.0521009,
      "balance_loss_mlp": 1.00004148,
      "epoch": 0.48890759333854383,
      "flos": 22784387224320.0,
      "grad_norm": 1.7240674376886689,
      "language_loss": 0.72531772,
      "learning_rate": 2.1688356707126286e-06,
      "loss": 0.74442828,
      "num_input_tokens_seen": 87563745,
      "step": 4066,
      "time_per_iteration": 2.549661159515381
    },
    {
      "auxiliary_loss_clip": 0.01137688,
      "auxiliary_loss_mlp": 0.01028774,
      "balance_loss_clip": 1.04970789,
      "balance_loss_mlp": 1.02037561,
      "epoch": 0.48902783622918294,
      "flos": 17786088956160.0,
      "grad_norm": 3.0225175160789726,
      "language_loss": 0.70128107,
      "learning_rate": 2.168059466862001e-06,
      "loss": 0.72294563,
      "num_input_tokens_seen": 87581895,
      "step": 4067,
      "time_per_iteration": 2.527517318725586
    },
    {
      "auxiliary_loss_clip": 0.01155495,
      "auxiliary_loss_mlp": 0.01027418,
      "balance_loss_clip": 1.04899478,
      "balance_loss_mlp": 1.0198009,
      "epoch": 0.48914807911982205,
      "flos": 22310294590080.0,
      "grad_norm": 1.9855676841262162,
      "language_loss": 0.8183217,
      "learning_rate": 2.167283237516887e-06,
      "loss": 0.84015083,
      "num_input_tokens_seen": 87600170,
      "step": 4068,
      "time_per_iteration": 2.5178065299987793
    },
    {
      "auxiliary_loss_clip": 0.01159729,
      "auxiliary_loss_mlp": 0.01033758,
      "balance_loss_clip": 1.05256808,
      "balance_loss_mlp": 1.02556229,
      "epoch": 0.4892683220104611,
      "flos": 16363020954240.0,
      "grad_norm": 1.7547126209766728,
      "language_loss": 0.74762404,
      "learning_rate": 2.1665069827950383e-06,
      "loss": 0.76955891,
      "num_input_tokens_seen": 87617455,
      "step": 4069,
      "time_per_iteration": 2.485710859298706
    },
    {
      "auxiliary_loss_clip": 0.01157013,
      "auxiliary_loss_mlp": 0.01026805,
      "balance_loss_clip": 1.05280459,
      "balance_loss_mlp": 1.01913369,
      "epoch": 0.4893885649011002,
      "flos": 15739144606080.0,
      "grad_norm": 2.3863527753022313,
      "language_loss": 0.86746252,
      "learning_rate": 2.1657307028142126e-06,
      "loss": 0.88930064,
      "num_input_tokens_seen": 87634995,
      "step": 4070,
      "time_per_iteration": 3.367102861404419
    },
    {
      "auxiliary_loss_clip": 0.01156453,
      "auxiliary_loss_mlp": 0.01034344,
      "balance_loss_clip": 1.05322647,
      "balance_loss_mlp": 1.02535522,
      "epoch": 0.48950880779173933,
      "flos": 28581984887040.0,
      "grad_norm": 3.149404413788562,
      "language_loss": 0.67488945,
      "learning_rate": 2.164954397692171e-06,
      "loss": 0.69679743,
      "num_input_tokens_seen": 87654420,
      "step": 4071,
      "time_per_iteration": 2.571049690246582
    },
    {
      "auxiliary_loss_clip": 0.0106145,
      "auxiliary_loss_mlp": 0.01003267,
      "balance_loss_clip": 1.01723826,
      "balance_loss_mlp": 1.00205135,
      "epoch": 0.4896290506823784,
      "flos": 66186310746240.0,
      "grad_norm": 1.0803737716027464,
      "language_loss": 0.77365255,
      "learning_rate": 2.164178067546678e-06,
      "loss": 0.79429972,
      "num_input_tokens_seen": 87713585,
      "step": 4072,
      "time_per_iteration": 3.135568857192993
    },
    {
      "auxiliary_loss_clip": 0.01159518,
      "auxiliary_loss_mlp": 0.01030447,
      "balance_loss_clip": 1.04995441,
      "balance_loss_mlp": 1.02207875,
      "epoch": 0.4897492935730175,
      "flos": 12531065207040.0,
      "grad_norm": 1.9436197279462866,
      "language_loss": 0.90988368,
      "learning_rate": 2.163401712495504e-06,
      "loss": 0.93178332,
      "num_input_tokens_seen": 87731280,
      "step": 4073,
      "time_per_iteration": 3.352949380874634
    },
    {
      "auxiliary_loss_clip": 0.01135533,
      "auxiliary_loss_mlp": 0.01036582,
      "balance_loss_clip": 1.05201387,
      "balance_loss_mlp": 1.02802885,
      "epoch": 0.4898695364636566,
      "flos": 23476816679040.0,
      "grad_norm": 1.509612098462676,
      "language_loss": 0.79023528,
      "learning_rate": 2.1626253326564194e-06,
      "loss": 0.81195647,
      "num_input_tokens_seen": 87750230,
      "step": 4074,
      "time_per_iteration": 3.3941211700439453
    },
    {
      "auxiliary_loss_clip": 0.01153414,
      "auxiliary_loss_mlp": 0.01030634,
      "balance_loss_clip": 1.04936397,
      "balance_loss_mlp": 1.02222383,
      "epoch": 0.48998977935429566,
      "flos": 27160209774720.0,
      "grad_norm": 3.837295918962431,
      "language_loss": 0.76680166,
      "learning_rate": 2.161848928147201e-06,
      "loss": 0.78864217,
      "num_input_tokens_seen": 87770500,
      "step": 4075,
      "time_per_iteration": 2.560128927230835
    },
    {
      "auxiliary_loss_clip": 0.01171607,
      "auxiliary_loss_mlp": 0.01026177,
      "balance_loss_clip": 1.05549455,
      "balance_loss_mlp": 1.01795769,
      "epoch": 0.4901100222449348,
      "flos": 20339588856960.0,
      "grad_norm": 1.970392440756129,
      "language_loss": 0.8081584,
      "learning_rate": 2.161072499085629e-06,
      "loss": 0.8301363,
      "num_input_tokens_seen": 87789495,
      "step": 4076,
      "time_per_iteration": 2.478710651397705
    },
    {
      "auxiliary_loss_clip": 0.01147646,
      "auxiliary_loss_mlp": 0.01026849,
      "balance_loss_clip": 1.05219328,
      "balance_loss_mlp": 1.01894557,
      "epoch": 0.4902302651355739,
      "flos": 30446359384320.0,
      "grad_norm": 1.8004402661580634,
      "language_loss": 0.83366829,
      "learning_rate": 2.160296045589487e-06,
      "loss": 0.8554132,
      "num_input_tokens_seen": 87812955,
      "step": 4077,
      "time_per_iteration": 2.6417996883392334
    },
    {
      "auxiliary_loss_clip": 0.01168822,
      "auxiliary_loss_mlp": 0.01024558,
      "balance_loss_clip": 1.05364525,
      "balance_loss_mlp": 1.01617169,
      "epoch": 0.49035050802621294,
      "flos": 19174180089600.0,
      "grad_norm": 2.377398819649,
      "language_loss": 0.69271755,
      "learning_rate": 2.159519567776562e-06,
      "loss": 0.71465135,
      "num_input_tokens_seen": 87832605,
      "step": 4078,
      "time_per_iteration": 3.205803632736206
    },
    {
      "auxiliary_loss_clip": 0.01129338,
      "auxiliary_loss_mlp": 0.01027053,
      "balance_loss_clip": 1.04453564,
      "balance_loss_mlp": 1.01866043,
      "epoch": 0.49047075091685205,
      "flos": 22228489365120.0,
      "grad_norm": 2.6030239595589744,
      "language_loss": 0.70642829,
      "learning_rate": 2.1587430657646463e-06,
      "loss": 0.72799218,
      "num_input_tokens_seen": 87846040,
      "step": 4079,
      "time_per_iteration": 2.547797679901123
    },
    {
      "auxiliary_loss_clip": 0.01154888,
      "auxiliary_loss_mlp": 0.01024846,
      "balance_loss_clip": 1.05300307,
      "balance_loss_mlp": 1.01694822,
      "epoch": 0.4905909938074911,
      "flos": 20156516213760.0,
      "grad_norm": 1.9767102803922114,
      "language_loss": 0.77868295,
      "learning_rate": 2.157966539671533e-06,
      "loss": 0.80048025,
      "num_input_tokens_seen": 87865680,
      "step": 4080,
      "time_per_iteration": 2.5252723693847656
    },
    {
      "auxiliary_loss_clip": 0.0114174,
      "auxiliary_loss_mlp": 0.01025557,
      "balance_loss_clip": 1.04926026,
      "balance_loss_mlp": 1.01799011,
      "epoch": 0.4907112366981302,
      "flos": 17202217380480.0,
      "grad_norm": 2.0091646715340072,
      "language_loss": 0.6726476,
      "learning_rate": 2.157189989615021e-06,
      "loss": 0.69432056,
      "num_input_tokens_seen": 87884270,
      "step": 4081,
      "time_per_iteration": 2.5656538009643555
    },
    {
      "auxiliary_loss_clip": 0.01170981,
      "auxiliary_loss_mlp": 0.00763178,
      "balance_loss_clip": 1.0515815,
      "balance_loss_mlp": 1.00011992,
      "epoch": 0.4908314795887693,
      "flos": 21688968107520.0,
      "grad_norm": 2.211378102152233,
      "language_loss": 0.75354463,
      "learning_rate": 2.156413415712913e-06,
      "loss": 0.77288622,
      "num_input_tokens_seen": 87906320,
      "step": 4082,
      "time_per_iteration": 2.5558478832244873
    },
    {
      "auxiliary_loss_clip": 0.01163424,
      "auxiliary_loss_mlp": 0.00763654,
      "balance_loss_clip": 1.05446339,
      "balance_loss_mlp": 1.00014174,
      "epoch": 0.4909517224794084,
      "flos": 26213676531840.0,
      "grad_norm": 1.6783046842293747,
      "language_loss": 0.78429568,
      "learning_rate": 2.155636818083014e-06,
      "loss": 0.80356646,
      "num_input_tokens_seen": 87927690,
      "step": 4083,
      "time_per_iteration": 2.538339853286743
    },
    {
      "auxiliary_loss_clip": 0.0115417,
      "auxiliary_loss_mlp": 0.01024173,
      "balance_loss_clip": 1.05390143,
      "balance_loss_mlp": 1.01689482,
      "epoch": 0.4910719653700475,
      "flos": 23148377694720.0,
      "grad_norm": 1.8961320654349467,
      "language_loss": 0.84105182,
      "learning_rate": 2.154860196843134e-06,
      "loss": 0.86283529,
      "num_input_tokens_seen": 87946885,
      "step": 4084,
      "time_per_iteration": 2.513349771499634
    },
    {
      "auxiliary_loss_clip": 0.0118527,
      "auxiliary_loss_mlp": 0.01028536,
      "balance_loss_clip": 1.05567265,
      "balance_loss_mlp": 1.02063835,
      "epoch": 0.4911922082606866,
      "flos": 23331845387520.0,
      "grad_norm": 1.6832848412475891,
      "language_loss": 0.76682121,
      "learning_rate": 2.154083552111085e-06,
      "loss": 0.78895926,
      "num_input_tokens_seen": 87966055,
      "step": 4085,
      "time_per_iteration": 2.453343152999878
    },
    {
      "auxiliary_loss_clip": 0.01185093,
      "auxiliary_loss_mlp": 0.01027101,
      "balance_loss_clip": 1.05306721,
      "balance_loss_mlp": 1.01885223,
      "epoch": 0.49131245115132566,
      "flos": 29203239542400.0,
      "grad_norm": 1.943740418708593,
      "language_loss": 0.81703532,
      "learning_rate": 2.1533068840046834e-06,
      "loss": 0.83915728,
      "num_input_tokens_seen": 87986320,
      "step": 4086,
      "time_per_iteration": 2.488363742828369
    },
    {
      "auxiliary_loss_clip": 0.01149195,
      "auxiliary_loss_mlp": 0.00763245,
      "balance_loss_clip": 1.05005085,
      "balance_loss_mlp": 1.00013399,
      "epoch": 0.49143269404196477,
      "flos": 20147465986560.0,
      "grad_norm": 2.328205497385505,
      "language_loss": 0.61776161,
      "learning_rate": 2.152530192641749e-06,
      "loss": 0.636886,
      "num_input_tokens_seen": 88001230,
      "step": 4087,
      "time_per_iteration": 2.4580159187316895
    },
    {
      "auxiliary_loss_clip": 0.01172731,
      "auxiliary_loss_mlp": 0.01032853,
      "balance_loss_clip": 1.05240154,
      "balance_loss_mlp": 1.0245502,
      "epoch": 0.4915529369326039,
      "flos": 24389809597440.0,
      "grad_norm": 1.8741599549456955,
      "language_loss": 0.72298652,
      "learning_rate": 2.1517534781401068e-06,
      "loss": 0.74504244,
      "num_input_tokens_seen": 88019110,
      "step": 4088,
      "time_per_iteration": 2.491710901260376
    },
    {
      "auxiliary_loss_clip": 0.01169037,
      "auxiliary_loss_mlp": 0.01024429,
      "balance_loss_clip": 1.0533483,
      "balance_loss_mlp": 1.01623356,
      "epoch": 0.49167317982324293,
      "flos": 10524305197440.0,
      "grad_norm": 2.5346007002992805,
      "language_loss": 0.69590163,
      "learning_rate": 2.150976740617581e-06,
      "loss": 0.71783626,
      "num_input_tokens_seen": 88035670,
      "step": 4089,
      "time_per_iteration": 2.45009708404541
    },
    {
      "auxiliary_loss_clip": 0.01161298,
      "auxiliary_loss_mlp": 0.01030254,
      "balance_loss_clip": 1.05377293,
      "balance_loss_mlp": 1.02201641,
      "epoch": 0.49179342271388204,
      "flos": 25593427457280.0,
      "grad_norm": 1.8633110249632234,
      "language_loss": 0.71191609,
      "learning_rate": 2.150199980192006e-06,
      "loss": 0.73383164,
      "num_input_tokens_seen": 88054790,
      "step": 4090,
      "time_per_iteration": 2.5462887287139893
    },
    {
      "auxiliary_loss_clip": 0.0114873,
      "auxiliary_loss_mlp": 0.01024441,
      "balance_loss_clip": 1.04979455,
      "balance_loss_mlp": 1.0166862,
      "epoch": 0.49191366560452116,
      "flos": 21102043875840.0,
      "grad_norm": 1.6342879001809376,
      "language_loss": 0.80829811,
      "learning_rate": 2.1494231969812114e-06,
      "loss": 0.83002979,
      "num_input_tokens_seen": 88073780,
      "step": 4091,
      "time_per_iteration": 2.471823215484619
    },
    {
      "auxiliary_loss_clip": 0.01148388,
      "auxiliary_loss_mlp": 0.01028542,
      "balance_loss_clip": 1.05338955,
      "balance_loss_mlp": 1.02001905,
      "epoch": 0.4920339084951602,
      "flos": 26067520091520.0,
      "grad_norm": 2.4557446123295015,
      "language_loss": 0.81295007,
      "learning_rate": 2.1486463911030372e-06,
      "loss": 0.83471942,
      "num_input_tokens_seen": 88094430,
      "step": 4092,
      "time_per_iteration": 2.581411361694336
    },
    {
      "auxiliary_loss_clip": 0.01151775,
      "auxiliary_loss_mlp": 0.01031163,
      "balance_loss_clip": 1.04779994,
      "balance_loss_mlp": 1.02317047,
      "epoch": 0.4921541513857993,
      "flos": 25081269384960.0,
      "grad_norm": 1.8172882285606748,
      "language_loss": 0.74467409,
      "learning_rate": 2.147869562675324e-06,
      "loss": 0.76650345,
      "num_input_tokens_seen": 88113400,
      "step": 4093,
      "time_per_iteration": 2.5447893142700195
    },
    {
      "auxiliary_loss_clip": 0.01169473,
      "auxiliary_loss_mlp": 0.01027965,
      "balance_loss_clip": 1.05332804,
      "balance_loss_mlp": 1.01990676,
      "epoch": 0.49227439427643843,
      "flos": 24389809597440.0,
      "grad_norm": 1.7643488611528046,
      "language_loss": 0.72301012,
      "learning_rate": 2.147092711815915e-06,
      "loss": 0.74498451,
      "num_input_tokens_seen": 88132750,
      "step": 4094,
      "time_per_iteration": 2.4886093139648438
    },
    {
      "auxiliary_loss_clip": 0.01141163,
      "auxiliary_loss_mlp": 0.01022179,
      "balance_loss_clip": 1.05231571,
      "balance_loss_mlp": 1.01460588,
      "epoch": 0.4923946371670775,
      "flos": 11363753018880.0,
      "grad_norm": 2.443281012314946,
      "language_loss": 0.8587079,
      "learning_rate": 2.1463158386426593e-06,
      "loss": 0.88034129,
      "num_input_tokens_seen": 88150560,
      "step": 4095,
      "time_per_iteration": 2.4847512245178223
    },
    {
      "auxiliary_loss_clip": 0.01163391,
      "auxiliary_loss_mlp": 0.01029957,
      "balance_loss_clip": 1.05332351,
      "balance_loss_mlp": 1.02148676,
      "epoch": 0.4925148800577166,
      "flos": 30445964334720.0,
      "grad_norm": 1.9821149252294032,
      "language_loss": 0.7748816,
      "learning_rate": 2.145538943273407e-06,
      "loss": 0.7968151,
      "num_input_tokens_seen": 88170835,
      "step": 4096,
      "time_per_iteration": 2.5843424797058105
    },
    {
      "auxiliary_loss_clip": 0.01186129,
      "auxiliary_loss_mlp": 0.01030119,
      "balance_loss_clip": 1.05723906,
      "balance_loss_mlp": 1.02219796,
      "epoch": 0.49263512294835565,
      "flos": 20850454039680.0,
      "grad_norm": 2.932051404567651,
      "language_loss": 0.71612585,
      "learning_rate": 2.144762025826013e-06,
      "loss": 0.73828828,
      "num_input_tokens_seen": 88189925,
      "step": 4097,
      "time_per_iteration": 3.275627374649048
    },
    {
      "auxiliary_loss_clip": 0.01174187,
      "auxiliary_loss_mlp": 0.01030341,
      "balance_loss_clip": 1.05285192,
      "balance_loss_mlp": 1.02218711,
      "epoch": 0.49275536583899476,
      "flos": 23767477534080.0,
      "grad_norm": 2.255771670245462,
      "language_loss": 0.86852419,
      "learning_rate": 2.143985086418334e-06,
      "loss": 0.89056945,
      "num_input_tokens_seen": 88205105,
      "step": 4098,
      "time_per_iteration": 2.450673818588257
    },
    {
      "auxiliary_loss_clip": 0.0115596,
      "auxiliary_loss_mlp": 0.01022018,
      "balance_loss_clip": 1.05104613,
      "balance_loss_mlp": 1.01432848,
      "epoch": 0.4928756087296339,
      "flos": 22273522041600.0,
      "grad_norm": 1.4335540950301706,
      "language_loss": 0.76434034,
      "learning_rate": 2.1432081251682324e-06,
      "loss": 0.78612012,
      "num_input_tokens_seen": 88225475,
      "step": 4099,
      "time_per_iteration": 3.343742847442627
    },
    {
      "auxiliary_loss_clip": 0.01176135,
      "auxiliary_loss_mlp": 0.01026985,
      "balance_loss_clip": 1.06125724,
      "balance_loss_mlp": 1.0181222,
      "epoch": 0.49299585162027293,
      "flos": 19645471463040.0,
      "grad_norm": 2.216194006508457,
      "language_loss": 0.87102968,
      "learning_rate": 2.142431142193572e-06,
      "loss": 0.89306086,
      "num_input_tokens_seen": 88243255,
      "step": 4100,
      "time_per_iteration": 3.276048183441162
    },
    {
      "auxiliary_loss_clip": 0.011841,
      "auxiliary_loss_mlp": 0.01029629,
      "balance_loss_clip": 1.0565002,
      "balance_loss_mlp": 1.02131426,
      "epoch": 0.49311609451091204,
      "flos": 38837138497920.0,
      "grad_norm": 2.299635026533799,
      "language_loss": 0.7161544,
      "learning_rate": 2.1416541376122207e-06,
      "loss": 0.73829168,
      "num_input_tokens_seen": 88263435,
      "step": 4101,
      "time_per_iteration": 2.579281806945801
    },
    {
      "auxiliary_loss_clip": 0.01182973,
      "auxiliary_loss_mlp": 0.01029403,
      "balance_loss_clip": 1.0530324,
      "balance_loss_mlp": 1.02055156,
      "epoch": 0.49323633740155115,
      "flos": 28329102161280.0,
      "grad_norm": 1.7028832097159055,
      "language_loss": 0.73146141,
      "learning_rate": 2.1408771115420496e-06,
      "loss": 0.75358522,
      "num_input_tokens_seen": 88283295,
      "step": 4102,
      "time_per_iteration": 2.47206711769104
    },
    {
      "auxiliary_loss_clip": 0.01130977,
      "auxiliary_loss_mlp": 0.01029036,
      "balance_loss_clip": 1.05464578,
      "balance_loss_mlp": 1.02138281,
      "epoch": 0.4933565802921902,
      "flos": 21135584200320.0,
      "grad_norm": 2.000212652467221,
      "language_loss": 0.64809251,
      "learning_rate": 2.140100064100932e-06,
      "loss": 0.66969264,
      "num_input_tokens_seen": 88299270,
      "step": 4103,
      "time_per_iteration": 2.577216386795044
    },
    {
      "auxiliary_loss_clip": 0.01166449,
      "auxiliary_loss_mlp": 0.01023223,
      "balance_loss_clip": 1.05257654,
      "balance_loss_mlp": 1.01554906,
      "epoch": 0.4934768231828293,
      "flos": 18039007595520.0,
      "grad_norm": 1.821748772083299,
      "language_loss": 0.75829285,
      "learning_rate": 2.139322995406746e-06,
      "loss": 0.78018957,
      "num_input_tokens_seen": 88316905,
      "step": 4104,
      "time_per_iteration": 3.143646478652954
    },
    {
      "auxiliary_loss_clip": 0.01187829,
      "auxiliary_loss_mlp": 0.01032771,
      "balance_loss_clip": 1.05776906,
      "balance_loss_mlp": 1.02409816,
      "epoch": 0.4935970660734684,
      "flos": 23469957181440.0,
      "grad_norm": 1.8561560074483614,
      "language_loss": 0.79598165,
      "learning_rate": 2.1385459055773727e-06,
      "loss": 0.81818759,
      "num_input_tokens_seen": 88335095,
      "step": 4105,
      "time_per_iteration": 2.4261133670806885
    },
    {
      "auxiliary_loss_clip": 0.011128,
      "auxiliary_loss_mlp": 0.00762647,
      "balance_loss_clip": 1.04559731,
      "balance_loss_mlp": 1.00013137,
      "epoch": 0.4937173089641075,
      "flos": 64479258840960.0,
      "grad_norm": 2.1145639537237235,
      "language_loss": 0.73897099,
      "learning_rate": 2.137768794730696e-06,
      "loss": 0.75772548,
      "num_input_tokens_seen": 88358545,
      "step": 4106,
      "time_per_iteration": 2.948566436767578
    },
    {
      "auxiliary_loss_clip": 0.01160698,
      "auxiliary_loss_mlp": 0.01034146,
      "balance_loss_clip": 1.0540297,
      "balance_loss_mlp": 1.02558064,
      "epoch": 0.4938375518547466,
      "flos": 22346025644160.0,
      "grad_norm": 1.7450999679347463,
      "language_loss": 0.80337608,
      "learning_rate": 2.1369916629846026e-06,
      "loss": 0.82532454,
      "num_input_tokens_seen": 88378295,
      "step": 4107,
      "time_per_iteration": 2.536890983581543
    },
    {
      "auxiliary_loss_clip": 0.0115413,
      "auxiliary_loss_mlp": 0.01024966,
      "balance_loss_clip": 1.04833758,
      "balance_loss_mlp": 1.01715827,
      "epoch": 0.4939577947453857,
      "flos": 17858700299520.0,
      "grad_norm": 1.8202031863284833,
      "language_loss": 0.75075233,
      "learning_rate": 2.136214510456983e-06,
      "loss": 0.77254331,
      "num_input_tokens_seen": 88396750,
      "step": 4108,
      "time_per_iteration": 2.6008787155151367
    },
    {
      "auxiliary_loss_clip": 0.01048722,
      "auxiliary_loss_mlp": 0.00753078,
      "balance_loss_clip": 1.02279603,
      "balance_loss_mlp": 1.00006735,
      "epoch": 0.49407803763602476,
      "flos": 70066746875520.0,
      "grad_norm": 0.8904927835101719,
      "language_loss": 0.63189888,
      "learning_rate": 2.1354373372657296e-06,
      "loss": 0.64991689,
      "num_input_tokens_seen": 88455190,
      "step": 4109,
      "time_per_iteration": 3.1587002277374268
    },
    {
      "auxiliary_loss_clip": 0.01183679,
      "auxiliary_loss_mlp": 0.01028884,
      "balance_loss_clip": 1.05637908,
      "balance_loss_mlp": 1.02154064,
      "epoch": 0.49419828052666387,
      "flos": 24317485562880.0,
      "grad_norm": 1.9446445796968632,
      "language_loss": 0.71029574,
      "learning_rate": 2.1346601435287404e-06,
      "loss": 0.7324214,
      "num_input_tokens_seen": 88477460,
      "step": 4110,
      "time_per_iteration": 2.5125043392181396
    },
    {
      "auxiliary_loss_clip": 0.0115357,
      "auxiliary_loss_mlp": 0.01026553,
      "balance_loss_clip": 1.04996765,
      "balance_loss_mlp": 1.01864958,
      "epoch": 0.494318523417303,
      "flos": 29386060790400.0,
      "grad_norm": 2.0056009138406687,
      "language_loss": 0.80229628,
      "learning_rate": 2.1338829293639144e-06,
      "loss": 0.82409751,
      "num_input_tokens_seen": 88497820,
      "step": 4111,
      "time_per_iteration": 2.5772926807403564
    },
    {
      "auxiliary_loss_clip": 0.01127871,
      "auxiliary_loss_mlp": 0.01033919,
      "balance_loss_clip": 1.05036068,
      "balance_loss_mlp": 1.02529407,
      "epoch": 0.49443876630794203,
      "flos": 15268284195840.0,
      "grad_norm": 1.9464909524123395,
      "language_loss": 0.82966781,
      "learning_rate": 2.1331056948891547e-06,
      "loss": 0.8512857,
      "num_input_tokens_seen": 88514920,
      "step": 4112,
      "time_per_iteration": 2.5548248291015625
    },
    {
      "auxiliary_loss_clip": 0.01151468,
      "auxiliary_loss_mlp": 0.01026444,
      "balance_loss_clip": 1.05139184,
      "balance_loss_mlp": 1.01839781,
      "epoch": 0.49455900919858115,
      "flos": 12347453859840.0,
      "grad_norm": 2.153547646896936,
      "language_loss": 0.76733214,
      "learning_rate": 2.1323284402223666e-06,
      "loss": 0.78911126,
      "num_input_tokens_seen": 88530910,
      "step": 4113,
      "time_per_iteration": 2.4448986053466797
    },
    {
      "auxiliary_loss_clip": 0.01182954,
      "auxiliary_loss_mlp": 0.00761664,
      "balance_loss_clip": 1.05851507,
      "balance_loss_mlp": 1.00008357,
      "epoch": 0.4946792520892202,
      "flos": 22779610715520.0,
      "grad_norm": 1.894227028815412,
      "language_loss": 0.88390982,
      "learning_rate": 2.1315511654814597e-06,
      "loss": 0.90335602,
      "num_input_tokens_seen": 88549320,
      "step": 4114,
      "time_per_iteration": 2.449101209640503
    },
    {
      "auxiliary_loss_clip": 0.0114949,
      "auxiliary_loss_mlp": 0.01024801,
      "balance_loss_clip": 1.05316114,
      "balance_loss_mlp": 1.01755619,
      "epoch": 0.4947994949798593,
      "flos": 23148126299520.0,
      "grad_norm": 1.8064484786574229,
      "language_loss": 0.78082669,
      "learning_rate": 2.1307738707843456e-06,
      "loss": 0.80256957,
      "num_input_tokens_seen": 88568985,
      "step": 4115,
      "time_per_iteration": 2.5029611587524414
    },
    {
      "auxiliary_loss_clip": 0.01175774,
      "auxiliary_loss_mlp": 0.01026601,
      "balance_loss_clip": 1.05641603,
      "balance_loss_mlp": 1.01887,
      "epoch": 0.4949197378704984,
      "flos": 23659997063040.0,
      "grad_norm": 1.9267952837255575,
      "language_loss": 0.68972141,
      "learning_rate": 2.1299965562489385e-06,
      "loss": 0.71174514,
      "num_input_tokens_seen": 88588790,
      "step": 4116,
      "time_per_iteration": 2.5004801750183105
    },
    {
      "auxiliary_loss_clip": 0.01167111,
      "auxiliary_loss_mlp": 0.01030432,
      "balance_loss_clip": 1.05162835,
      "balance_loss_mlp": 1.02229583,
      "epoch": 0.4950399807611375,
      "flos": 26911493026560.0,
      "grad_norm": 1.404968374885798,
      "language_loss": 0.79080629,
      "learning_rate": 2.129219221993158e-06,
      "loss": 0.81278175,
      "num_input_tokens_seen": 88613575,
      "step": 4117,
      "time_per_iteration": 2.54606032371521
    },
    {
      "auxiliary_loss_clip": 0.01057268,
      "auxiliary_loss_mlp": 0.01006629,
      "balance_loss_clip": 1.02741957,
      "balance_loss_mlp": 1.00551462,
      "epoch": 0.4951602236517766,
      "flos": 67315270187520.0,
      "grad_norm": 0.7990890443105685,
      "language_loss": 0.59967893,
      "learning_rate": 2.128441868134924e-06,
      "loss": 0.62031788,
      "num_input_tokens_seen": 88675510,
      "step": 4118,
      "time_per_iteration": 3.1416573524475098
    },
    {
      "auxiliary_loss_clip": 0.01142471,
      "auxiliary_loss_mlp": 0.01027405,
      "balance_loss_clip": 1.04829574,
      "balance_loss_mlp": 1.01922154,
      "epoch": 0.4952804665424157,
      "flos": 19901442758400.0,
      "grad_norm": 2.0912139166573,
      "language_loss": 0.82745439,
      "learning_rate": 2.1276644947921606e-06,
      "loss": 0.84915316,
      "num_input_tokens_seen": 88694425,
      "step": 4119,
      "time_per_iteration": 2.5093836784362793
    },
    {
      "auxiliary_loss_clip": 0.0116913,
      "auxiliary_loss_mlp": 0.01026062,
      "balance_loss_clip": 1.05257046,
      "balance_loss_mlp": 1.01715124,
      "epoch": 0.49540070943305475,
      "flos": 18806813740800.0,
      "grad_norm": 1.9239056445013087,
      "language_loss": 0.8263731,
      "learning_rate": 2.126887102082795e-06,
      "loss": 0.84832501,
      "num_input_tokens_seen": 88714450,
      "step": 4120,
      "time_per_iteration": 2.4562063217163086
    },
    {
      "auxiliary_loss_clip": 0.0113999,
      "auxiliary_loss_mlp": 0.0102735,
      "balance_loss_clip": 1.04816628,
      "balance_loss_mlp": 1.01959276,
      "epoch": 0.49552095232369386,
      "flos": 24934179191040.0,
      "grad_norm": 1.9105544846739133,
      "language_loss": 0.70732427,
      "learning_rate": 2.126109690124757e-06,
      "loss": 0.72899771,
      "num_input_tokens_seen": 88735265,
      "step": 4121,
      "time_per_iteration": 2.5715651512145996
    },
    {
      "auxiliary_loss_clip": 0.01126929,
      "auxiliary_loss_mlp": 0.01027542,
      "balance_loss_clip": 1.04761755,
      "balance_loss_mlp": 1.01981711,
      "epoch": 0.495641195214333,
      "flos": 22857249962880.0,
      "grad_norm": 1.6611266414500356,
      "language_loss": 0.71172798,
      "learning_rate": 2.1253322590359786e-06,
      "loss": 0.73327267,
      "num_input_tokens_seen": 88754600,
      "step": 4122,
      "time_per_iteration": 2.574528455734253
    },
    {
      "auxiliary_loss_clip": 0.01166389,
      "auxiliary_loss_mlp": 0.01032753,
      "balance_loss_clip": 1.05131733,
      "balance_loss_mlp": 1.02505851,
      "epoch": 0.49576143810497203,
      "flos": 25769748343680.0,
      "grad_norm": 1.7271996843557909,
      "language_loss": 0.73734379,
      "learning_rate": 2.124554808934397e-06,
      "loss": 0.75933522,
      "num_input_tokens_seen": 88775180,
      "step": 4123,
      "time_per_iteration": 2.5132992267608643
    },
    {
      "auxiliary_loss_clip": 0.01119559,
      "auxiliary_loss_mlp": 0.01028642,
      "balance_loss_clip": 1.04577088,
      "balance_loss_mlp": 1.02005935,
      "epoch": 0.49588168099561114,
      "flos": 22128838058880.0,
      "grad_norm": 1.8237209943990293,
      "language_loss": 0.72868013,
      "learning_rate": 2.1237773399379496e-06,
      "loss": 0.75016212,
      "num_input_tokens_seen": 88796145,
      "step": 4124,
      "time_per_iteration": 3.4027979373931885
    },
    {
      "auxiliary_loss_clip": 0.01157245,
      "auxiliary_loss_mlp": 0.01025202,
      "balance_loss_clip": 1.04608119,
      "balance_loss_mlp": 1.01640153,
      "epoch": 0.49600192388625025,
      "flos": 24387331559040.0,
      "grad_norm": 9.384741885532225,
      "language_loss": 0.8693943,
      "learning_rate": 2.122999852164578e-06,
      "loss": 0.89121872,
      "num_input_tokens_seen": 88816765,
      "step": 4125,
      "time_per_iteration": 2.534491539001465
    },
    {
      "auxiliary_loss_clip": 0.01123971,
      "auxiliary_loss_mlp": 0.01024233,
      "balance_loss_clip": 1.04789424,
      "balance_loss_mlp": 1.01600778,
      "epoch": 0.4961221667768893,
      "flos": 22857429530880.0,
      "grad_norm": 2.3943681056355444,
      "language_loss": 0.58202177,
      "learning_rate": 2.122222345732227e-06,
      "loss": 0.60350382,
      "num_input_tokens_seen": 88836680,
      "step": 4126,
      "time_per_iteration": 3.4016165733337402
    },
    {
      "auxiliary_loss_clip": 0.01141351,
      "auxiliary_loss_mlp": 0.01027726,
      "balance_loss_clip": 1.048908,
      "balance_loss_mlp": 1.01952422,
      "epoch": 0.4962424096675284,
      "flos": 17858089768320.0,
      "grad_norm": 2.678943307408686,
      "language_loss": 0.83160228,
      "learning_rate": 2.121444820758843e-06,
      "loss": 0.85329306,
      "num_input_tokens_seen": 88855320,
      "step": 4127,
      "time_per_iteration": 3.260526180267334
    },
    {
      "auxiliary_loss_clip": 0.01125578,
      "auxiliary_loss_mlp": 0.01034139,
      "balance_loss_clip": 1.04996705,
      "balance_loss_mlp": 1.02537668,
      "epoch": 0.49636265255816747,
      "flos": 21793611404160.0,
      "grad_norm": 2.081068333847806,
      "language_loss": 0.78742272,
      "learning_rate": 2.120667277362376e-06,
      "loss": 0.80901992,
      "num_input_tokens_seen": 88874035,
      "step": 4128,
      "time_per_iteration": 2.5433716773986816
    },
    {
      "auxiliary_loss_clip": 0.0118723,
      "auxiliary_loss_mlp": 0.01036348,
      "balance_loss_clip": 1.05742288,
      "balance_loss_mlp": 1.02771091,
      "epoch": 0.4964828954488066,
      "flos": 16358603581440.0,
      "grad_norm": 2.201434720448348,
      "language_loss": 0.85090858,
      "learning_rate": 2.1198897156607796e-06,
      "loss": 0.87314433,
      "num_input_tokens_seen": 88891390,
      "step": 4129,
      "time_per_iteration": 2.4152348041534424
    },
    {
      "auxiliary_loss_clip": 0.01173403,
      "auxiliary_loss_mlp": 0.01029267,
      "balance_loss_clip": 1.05186415,
      "balance_loss_mlp": 1.02113104,
      "epoch": 0.4966031383394457,
      "flos": 24711101775360.0,
      "grad_norm": 2.9455213774524034,
      "language_loss": 0.73813653,
      "learning_rate": 2.1191121357720085e-06,
      "loss": 0.76016319,
      "num_input_tokens_seen": 88909450,
      "step": 4130,
      "time_per_iteration": 3.2572743892669678
    },
    {
      "auxiliary_loss_clip": 0.01120327,
      "auxiliary_loss_mlp": 0.01029676,
      "balance_loss_clip": 1.04868948,
      "balance_loss_mlp": 1.02118862,
      "epoch": 0.49672338123008475,
      "flos": 22930615491840.0,
      "grad_norm": 1.6450711028103961,
      "language_loss": 0.7504428,
      "learning_rate": 2.1183345378140206e-06,
      "loss": 0.77194291,
      "num_input_tokens_seen": 88929195,
      "step": 4131,
      "time_per_iteration": 2.6694369316101074
    },
    {
      "auxiliary_loss_clip": 0.01072872,
      "auxiliary_loss_mlp": 0.01002259,
      "balance_loss_clip": 1.01930702,
      "balance_loss_mlp": 1.0008105,
      "epoch": 0.49684362412072386,
      "flos": 65976736844160.0,
      "grad_norm": 0.8539052808092114,
      "language_loss": 0.61984456,
      "learning_rate": 2.1175569219047783e-06,
      "loss": 0.64059579,
      "num_input_tokens_seen": 88990635,
      "step": 4132,
      "time_per_iteration": 3.160383701324463
    },
    {
      "auxiliary_loss_clip": 0.0118328,
      "auxiliary_loss_mlp": 0.01027755,
      "balance_loss_clip": 1.05444455,
      "balance_loss_mlp": 1.01999497,
      "epoch": 0.49696386701136297,
      "flos": 19971288754560.0,
      "grad_norm": 1.6719445703853393,
      "language_loss": 0.73208475,
      "learning_rate": 2.1167792881622437e-06,
      "loss": 0.75419515,
      "num_input_tokens_seen": 89009655,
      "step": 4133,
      "time_per_iteration": 2.4385061264038086
    },
    {
      "auxiliary_loss_clip": 0.01153971,
      "auxiliary_loss_mlp": 0.01034489,
      "balance_loss_clip": 1.0534575,
      "balance_loss_mlp": 1.02666879,
      "epoch": 0.497084109902002,
      "flos": 24750819239040.0,
      "grad_norm": 1.7693898649399644,
      "language_loss": 0.81001109,
      "learning_rate": 2.116001636704384e-06,
      "loss": 0.83189565,
      "num_input_tokens_seen": 89030040,
      "step": 4134,
      "time_per_iteration": 2.5290114879608154
    },
    {
      "auxiliary_loss_clip": 0.01138425,
      "auxiliary_loss_mlp": 0.01032851,
      "balance_loss_clip": 1.05056131,
      "balance_loss_mlp": 1.02447033,
      "epoch": 0.49720435279264114,
      "flos": 21871825269120.0,
      "grad_norm": 1.8235047463180554,
      "language_loss": 0.80288315,
      "learning_rate": 2.1152239676491685e-06,
      "loss": 0.82459593,
      "num_input_tokens_seen": 89048145,
      "step": 4135,
      "time_per_iteration": 2.5922958850860596
    },
    {
      "auxiliary_loss_clip": 0.01159992,
      "auxiliary_loss_mlp": 0.01025529,
      "balance_loss_clip": 1.05013418,
      "balance_loss_mlp": 1.0177151,
      "epoch": 0.49732459568328025,
      "flos": 23805794367360.0,
      "grad_norm": 1.7542176573297985,
      "language_loss": 0.73184544,
      "learning_rate": 2.114446281114569e-06,
      "loss": 0.75370061,
      "num_input_tokens_seen": 89067165,
      "step": 4136,
      "time_per_iteration": 2.5260047912597656
    },
    {
      "auxiliary_loss_clip": 0.01147251,
      "auxiliary_loss_mlp": 0.01026575,
      "balance_loss_clip": 1.05117249,
      "balance_loss_mlp": 1.01786077,
      "epoch": 0.4974448385739193,
      "flos": 20047742853120.0,
      "grad_norm": 1.9369203888230022,
      "language_loss": 0.76220119,
      "learning_rate": 2.1136685772185587e-06,
      "loss": 0.78393948,
      "num_input_tokens_seen": 89086190,
      "step": 4137,
      "time_per_iteration": 2.489786386489868
    },
    {
      "auxiliary_loss_clip": 0.01152505,
      "auxiliary_loss_mlp": 0.00763387,
      "balance_loss_clip": 1.04498053,
      "balance_loss_mlp": 1.0000608,
      "epoch": 0.4975650814645584,
      "flos": 24821347593600.0,
      "grad_norm": 1.600728129345119,
      "language_loss": 0.78087401,
      "learning_rate": 2.1128908560791163e-06,
      "loss": 0.80003291,
      "num_input_tokens_seen": 89106020,
      "step": 4138,
      "time_per_iteration": 2.541379690170288
    },
    {
      "auxiliary_loss_clip": 0.01184245,
      "auxiliary_loss_mlp": 0.01029795,
      "balance_loss_clip": 1.05584574,
      "balance_loss_mlp": 1.02143252,
      "epoch": 0.4976853243551975,
      "flos": 19829477859840.0,
      "grad_norm": 2.1474548819042285,
      "language_loss": 0.78336895,
      "learning_rate": 2.1121131178142203e-06,
      "loss": 0.80550939,
      "num_input_tokens_seen": 89125385,
      "step": 4139,
      "time_per_iteration": 2.4247288703918457
    },
    {
      "auxiliary_loss_clip": 0.01156934,
      "auxiliary_loss_mlp": 0.01025583,
      "balance_loss_clip": 1.05096173,
      "balance_loss_mlp": 1.01781678,
      "epoch": 0.4978055672458366,
      "flos": 23142990654720.0,
      "grad_norm": 1.5999293712498972,
      "language_loss": 0.82329381,
      "learning_rate": 2.1113353625418544e-06,
      "loss": 0.845119,
      "num_input_tokens_seen": 89143935,
      "step": 4140,
      "time_per_iteration": 2.505744457244873
    },
    {
      "auxiliary_loss_clip": 0.01162455,
      "auxiliary_loss_mlp": 0.01029005,
      "balance_loss_clip": 1.05359507,
      "balance_loss_mlp": 1.02156019,
      "epoch": 0.4979258101364757,
      "flos": 15559914718080.0,
      "grad_norm": 1.754613911575738,
      "language_loss": 0.79090488,
      "learning_rate": 2.1105575903800017e-06,
      "loss": 0.81281954,
      "num_input_tokens_seen": 89162655,
      "step": 4141,
      "time_per_iteration": 2.427483081817627
    },
    {
      "auxiliary_loss_clip": 0.01173402,
      "auxiliary_loss_mlp": 0.01026574,
      "balance_loss_clip": 1.05192101,
      "balance_loss_mlp": 1.01840806,
      "epoch": 0.4980460530271148,
      "flos": 26356169784960.0,
      "grad_norm": 1.8894631223960248,
      "language_loss": 0.85268092,
      "learning_rate": 2.1097798014466502e-06,
      "loss": 0.8746807,
      "num_input_tokens_seen": 89182255,
      "step": 4142,
      "time_per_iteration": 2.484276533126831
    },
    {
      "auxiliary_loss_clip": 0.01173964,
      "auxiliary_loss_mlp": 0.01028953,
      "balance_loss_clip": 1.05452919,
      "balance_loss_mlp": 1.01988733,
      "epoch": 0.49816629591775385,
      "flos": 17274541415040.0,
      "grad_norm": 2.074310481810634,
      "language_loss": 0.5847829,
      "learning_rate": 2.109001995859791e-06,
      "loss": 0.606812,
      "num_input_tokens_seen": 89201155,
      "step": 4143,
      "time_per_iteration": 2.4370336532592773
    },
    {
      "auxiliary_loss_clip": 0.01061754,
      "auxiliary_loss_mlp": 0.01003484,
      "balance_loss_clip": 1.02219915,
      "balance_loss_mlp": 1.0021069,
      "epoch": 0.49828653880839296,
      "flos": 64930947344640.0,
      "grad_norm": 0.8104183401912067,
      "language_loss": 0.60123158,
      "learning_rate": 2.108224173737415e-06,
      "loss": 0.62188399,
      "num_input_tokens_seen": 89264455,
      "step": 4144,
      "time_per_iteration": 3.064518451690674
    },
    {
      "auxiliary_loss_clip": 0.01150527,
      "auxiliary_loss_mlp": 0.01030286,
      "balance_loss_clip": 1.04804778,
      "balance_loss_mlp": 1.02130914,
      "epoch": 0.498406781699032,
      "flos": 27484806003840.0,
      "grad_norm": 3.0373829126736895,
      "language_loss": 0.76386493,
      "learning_rate": 2.1074463351975183e-06,
      "loss": 0.78567308,
      "num_input_tokens_seen": 89283340,
      "step": 4145,
      "time_per_iteration": 2.5412256717681885
    },
    {
      "auxiliary_loss_clip": 0.01144585,
      "auxiliary_loss_mlp": 0.0102251,
      "balance_loss_clip": 1.04964185,
      "balance_loss_mlp": 1.0148474,
      "epoch": 0.49852702458967113,
      "flos": 31499870307840.0,
      "grad_norm": 1.6063259693606657,
      "language_loss": 0.71589231,
      "learning_rate": 2.106668480358098e-06,
      "loss": 0.73756331,
      "num_input_tokens_seen": 89303565,
      "step": 4146,
      "time_per_iteration": 2.6004159450531006
    },
    {
      "auxiliary_loss_clip": 0.01150463,
      "auxiliary_loss_mlp": 0.01026896,
      "balance_loss_clip": 1.04723525,
      "balance_loss_mlp": 1.01814604,
      "epoch": 0.49864726748031024,
      "flos": 22852868503680.0,
      "grad_norm": 1.567746630007518,
      "language_loss": 0.70865643,
      "learning_rate": 2.105890609337154e-06,
      "loss": 0.73043001,
      "num_input_tokens_seen": 89322080,
      "step": 4147,
      "time_per_iteration": 2.5667285919189453
    },
    {
      "auxiliary_loss_clip": 0.01083031,
      "auxiliary_loss_mlp": 0.01001448,
      "balance_loss_clip": 1.02080953,
      "balance_loss_mlp": 1.00014901,
      "epoch": 0.4987675103709493,
      "flos": 70405708544640.0,
      "grad_norm": 0.6878898786990562,
      "language_loss": 0.63853502,
      "learning_rate": 2.1051127222526883e-06,
      "loss": 0.65937972,
      "num_input_tokens_seen": 89394195,
      "step": 4148,
      "time_per_iteration": 3.1408419609069824
    },
    {
      "auxiliary_loss_clip": 0.01168398,
      "auxiliary_loss_mlp": 0.01023651,
      "balance_loss_clip": 1.05565691,
      "balance_loss_mlp": 1.01578307,
      "epoch": 0.4988877532615884,
      "flos": 28767571482240.0,
      "grad_norm": 1.5412398137947618,
      "language_loss": 0.8084116,
      "learning_rate": 2.1043348192227067e-06,
      "loss": 0.83033204,
      "num_input_tokens_seen": 89414565,
      "step": 4149,
      "time_per_iteration": 2.5256338119506836
    },
    {
      "auxiliary_loss_clip": 0.01129651,
      "auxiliary_loss_mlp": 0.01032881,
      "balance_loss_clip": 1.04869795,
      "balance_loss_mlp": 1.02454877,
      "epoch": 0.4990079961522275,
      "flos": 16872700988160.0,
      "grad_norm": 1.6977619615921062,
      "language_loss": 0.61629057,
      "learning_rate": 2.1035569003652156e-06,
      "loss": 0.63791585,
      "num_input_tokens_seen": 89433195,
      "step": 4150,
      "time_per_iteration": 3.3442342281341553
    },
    {
      "auxiliary_loss_clip": 0.01123107,
      "auxiliary_loss_mlp": 0.01037677,
      "balance_loss_clip": 1.04685307,
      "balance_loss_mlp": 1.02780664,
      "epoch": 0.4991282390428666,
      "flos": 13291042187520.0,
      "grad_norm": 1.9188090522970949,
      "language_loss": 0.81876165,
      "learning_rate": 2.1027789657982255e-06,
      "loss": 0.84036946,
      "num_input_tokens_seen": 89447410,
      "step": 4151,
      "time_per_iteration": 2.5223541259765625
    },
    {
      "auxiliary_loss_clip": 0.01127162,
      "auxiliary_loss_mlp": 0.01032447,
      "balance_loss_clip": 1.04864275,
      "balance_loss_mlp": 1.02450156,
      "epoch": 0.4992484819335057,
      "flos": 21537496454400.0,
      "grad_norm": 1.9915568983225955,
      "language_loss": 0.7703954,
      "learning_rate": 2.1020010156397482e-06,
      "loss": 0.79199147,
      "num_input_tokens_seen": 89464630,
      "step": 4152,
      "time_per_iteration": 2.558551788330078
    },
    {
      "auxiliary_loss_clip": 0.01169351,
      "auxiliary_loss_mlp": 0.01033487,
      "balance_loss_clip": 1.05259895,
      "balance_loss_mlp": 1.02545202,
      "epoch": 0.4993687248241448,
      "flos": 24860095390080.0,
      "grad_norm": 1.4786463022889875,
      "language_loss": 0.7766124,
      "learning_rate": 2.101223050007797e-06,
      "loss": 0.79864079,
      "num_input_tokens_seen": 89483180,
      "step": 4153,
      "time_per_iteration": 3.3580849170684814
    },
    {
      "auxiliary_loss_clip": 0.0108163,
      "auxiliary_loss_mlp": 0.01001306,
      "balance_loss_clip": 1.01904166,
      "balance_loss_mlp": 0.99999505,
      "epoch": 0.49948896771478385,
      "flos": 62941602453120.0,
      "grad_norm": 0.8207227722631733,
      "language_loss": 0.53765684,
      "learning_rate": 2.1004450690203904e-06,
      "loss": 0.55848616,
      "num_input_tokens_seen": 89539260,
      "step": 4154,
      "time_per_iteration": 3.8548424243927
    },
    {
      "auxiliary_loss_clip": 0.01081326,
      "auxiliary_loss_mlp": 0.01001137,
      "balance_loss_clip": 1.01902056,
      "balance_loss_mlp": 0.99988502,
      "epoch": 0.49960921060542296,
      "flos": 68284213516800.0,
      "grad_norm": 0.8536652669784582,
      "language_loss": 0.63358331,
      "learning_rate": 2.099667072795546e-06,
      "loss": 0.65440792,
      "num_input_tokens_seen": 89601380,
      "step": 4155,
      "time_per_iteration": 3.084392547607422
    },
    {
      "auxiliary_loss_clip": 0.01166483,
      "auxiliary_loss_mlp": 0.01030603,
      "balance_loss_clip": 1.05074334,
      "balance_loss_mlp": 1.02206814,
      "epoch": 0.49972945349606207,
      "flos": 23659350618240.0,
      "grad_norm": 2.3055516959853253,
      "language_loss": 0.79888976,
      "learning_rate": 2.0988890614512864e-06,
      "loss": 0.82086068,
      "num_input_tokens_seen": 89621270,
      "step": 4156,
      "time_per_iteration": 2.5225977897644043
    },
    {
      "auxiliary_loss_clip": 0.01159267,
      "auxiliary_loss_mlp": 0.01029429,
      "balance_loss_clip": 1.05502033,
      "balance_loss_mlp": 1.02136457,
      "epoch": 0.4998496963867011,
      "flos": 19755825022080.0,
      "grad_norm": 1.6344915540691656,
      "language_loss": 0.84120947,
      "learning_rate": 2.098111035105635e-06,
      "loss": 0.86309636,
      "num_input_tokens_seen": 89639695,
      "step": 4157,
      "time_per_iteration": 3.2427964210510254
    },
    {
      "auxiliary_loss_clip": 0.01126301,
      "auxiliary_loss_mlp": 0.01029097,
      "balance_loss_clip": 1.05217063,
      "balance_loss_mlp": 1.02112246,
      "epoch": 0.49996993927734024,
      "flos": 22265728790400.0,
      "grad_norm": 1.7160573327998834,
      "language_loss": 0.73278666,
      "learning_rate": 2.0973329938766176e-06,
      "loss": 0.75434065,
      "num_input_tokens_seen": 89657125,
      "step": 4158,
      "time_per_iteration": 2.596588134765625
    },
    {
      "auxiliary_loss_clip": 0.01165641,
      "auxiliary_loss_mlp": 0.01033458,
      "balance_loss_clip": 1.05397642,
      "balance_loss_mlp": 1.02487493,
      "epoch": 0.5000901821679793,
      "flos": 23327212533120.0,
      "grad_norm": 1.9087083097330289,
      "language_loss": 0.78850073,
      "learning_rate": 2.0965549378822618e-06,
      "loss": 0.81049168,
      "num_input_tokens_seen": 89678415,
      "step": 4159,
      "time_per_iteration": 2.580095052719116
    },
    {
      "auxiliary_loss_clip": 0.01091393,
      "auxiliary_loss_mlp": 0.01026239,
      "balance_loss_clip": 1.04507875,
      "balance_loss_mlp": 1.01816249,
      "epoch": 0.5002104250586185,
      "flos": 20339014239360.0,
      "grad_norm": 1.9060454826206714,
      "language_loss": 0.83911598,
      "learning_rate": 2.095776867240599e-06,
      "loss": 0.8602922,
      "num_input_tokens_seen": 89695405,
      "step": 4160,
      "time_per_iteration": 2.6216256618499756
    },
    {
      "auxiliary_loss_clip": 0.01133653,
      "auxiliary_loss_mlp": 0.01029621,
      "balance_loss_clip": 1.0473839,
      "balance_loss_mlp": 1.02227426,
      "epoch": 0.5003306679492575,
      "flos": 13991372634240.0,
      "grad_norm": 1.845389233110712,
      "language_loss": 0.82811105,
      "learning_rate": 2.094998782069661e-06,
      "loss": 0.84974384,
      "num_input_tokens_seen": 89713110,
      "step": 4161,
      "time_per_iteration": 2.5136258602142334
    },
    {
      "auxiliary_loss_clip": 0.01183639,
      "auxiliary_loss_mlp": 0.01027165,
      "balance_loss_clip": 1.05594206,
      "balance_loss_mlp": 1.01898742,
      "epoch": 0.5004509108398966,
      "flos": 27672762896640.0,
      "grad_norm": 1.7248011203696096,
      "language_loss": 0.75443238,
      "learning_rate": 2.0942206824874845e-06,
      "loss": 0.77654046,
      "num_input_tokens_seen": 89735885,
      "step": 4162,
      "time_per_iteration": 2.490328550338745
    },
    {
      "auxiliary_loss_clip": 0.01170035,
      "auxiliary_loss_mlp": 0.01027891,
      "balance_loss_clip": 1.05594885,
      "balance_loss_mlp": 1.0193975,
      "epoch": 0.5005711537305357,
      "flos": 14976186796800.0,
      "grad_norm": 2.1625795840130104,
      "language_loss": 0.79006231,
      "learning_rate": 2.093442568612105e-06,
      "loss": 0.81204152,
      "num_input_tokens_seen": 89753690,
      "step": 4163,
      "time_per_iteration": 2.4314751625061035
    },
    {
      "auxiliary_loss_clip": 0.01182263,
      "auxiliary_loss_mlp": 0.01025346,
      "balance_loss_clip": 1.05275989,
      "balance_loss_mlp": 1.0171802,
      "epoch": 0.5006913966211748,
      "flos": 26503259978880.0,
      "grad_norm": 1.4703431591048903,
      "language_loss": 0.85319805,
      "learning_rate": 2.0926644405615613e-06,
      "loss": 0.87527406,
      "num_input_tokens_seen": 89774590,
      "step": 4164,
      "time_per_iteration": 2.4616031646728516
    },
    {
      "auxiliary_loss_clip": 0.01133387,
      "auxiliary_loss_mlp": 0.01028397,
      "balance_loss_clip": 1.04901135,
      "balance_loss_mlp": 1.02048099,
      "epoch": 0.5008116395118138,
      "flos": 20449295971200.0,
      "grad_norm": 1.8117700571286253,
      "language_loss": 0.81258881,
      "learning_rate": 2.091886298453897e-06,
      "loss": 0.83420658,
      "num_input_tokens_seen": 89792775,
      "step": 4165,
      "time_per_iteration": 2.5090725421905518
    },
    {
      "auxiliary_loss_clip": 0.01166732,
      "auxiliary_loss_mlp": 0.01024058,
      "balance_loss_clip": 1.05182672,
      "balance_loss_mlp": 1.01641965,
      "epoch": 0.500931882402453,
      "flos": 21579871524480.0,
      "grad_norm": 2.1610045685585133,
      "language_loss": 0.73066109,
      "learning_rate": 2.091108142407153e-06,
      "loss": 0.75256902,
      "num_input_tokens_seen": 89811515,
      "step": 4166,
      "time_per_iteration": 2.4603617191314697
    },
    {
      "auxiliary_loss_clip": 0.01058418,
      "auxiliary_loss_mlp": 0.01003957,
      "balance_loss_clip": 1.029953,
      "balance_loss_mlp": 1.00227606,
      "epoch": 0.5010521252930921,
      "flos": 57785011925760.0,
      "grad_norm": 0.8383630926993236,
      "language_loss": 0.62405729,
      "learning_rate": 2.090329972539377e-06,
      "loss": 0.64468098,
      "num_input_tokens_seen": 89870080,
      "step": 4167,
      "time_per_iteration": 3.124333143234253
    },
    {
      "auxiliary_loss_clip": 0.01092424,
      "auxiliary_loss_mlp": 0.01030355,
      "balance_loss_clip": 1.04495597,
      "balance_loss_mlp": 1.02255905,
      "epoch": 0.5011723681837311,
      "flos": 18625500864000.0,
      "grad_norm": 1.7024647686693317,
      "language_loss": 0.68343723,
      "learning_rate": 2.089551788968616e-06,
      "loss": 0.70466501,
      "num_input_tokens_seen": 89888045,
      "step": 4168,
      "time_per_iteration": 2.5765953063964844
    },
    {
      "auxiliary_loss_clip": 0.0107902,
      "auxiliary_loss_mlp": 0.01003842,
      "balance_loss_clip": 1.01689529,
      "balance_loss_mlp": 1.00264382,
      "epoch": 0.5012926110743702,
      "flos": 55883146608000.0,
      "grad_norm": 0.834716188739814,
      "language_loss": 0.60811985,
      "learning_rate": 2.08877359181292e-06,
      "loss": 0.62894845,
      "num_input_tokens_seen": 89944610,
      "step": 4169,
      "time_per_iteration": 2.985989809036255
    },
    {
      "auxiliary_loss_clip": 0.01142338,
      "auxiliary_loss_mlp": 0.01026952,
      "balance_loss_clip": 1.04551339,
      "balance_loss_mlp": 1.01925445,
      "epoch": 0.5014128539650093,
      "flos": 24238266117120.0,
      "grad_norm": 2.4386148547423683,
      "language_loss": 0.85715073,
      "learning_rate": 2.0879953811903396e-06,
      "loss": 0.87884367,
      "num_input_tokens_seen": 89959495,
      "step": 4170,
      "time_per_iteration": 2.5322375297546387
    },
    {
      "auxiliary_loss_clip": 0.01167103,
      "auxiliary_loss_mlp": 0.0103121,
      "balance_loss_clip": 1.0531826,
      "balance_loss_mlp": 1.02312195,
      "epoch": 0.5015330968556484,
      "flos": 27527468382720.0,
      "grad_norm": 1.7939722933105977,
      "language_loss": 0.7837528,
      "learning_rate": 2.08721715721893e-06,
      "loss": 0.80573595,
      "num_input_tokens_seen": 89978820,
      "step": 4171,
      "time_per_iteration": 2.5158846378326416
    },
    {
      "auxiliary_loss_clip": 0.01167839,
      "auxiliary_loss_mlp": 0.01030489,
      "balance_loss_clip": 1.05332685,
      "balance_loss_mlp": 1.0225141,
      "epoch": 0.5016533397462875,
      "flos": 23800802376960.0,
      "grad_norm": 1.747416038381673,
      "language_loss": 0.76668239,
      "learning_rate": 2.0864389200167477e-06,
      "loss": 0.78866565,
      "num_input_tokens_seen": 89997075,
      "step": 4172,
      "time_per_iteration": 2.47259521484375
    },
    {
      "auxiliary_loss_clip": 0.01170802,
      "auxiliary_loss_mlp": 0.00762675,
      "balance_loss_clip": 1.05285501,
      "balance_loss_mlp": 1.00014329,
      "epoch": 0.5017735826369266,
      "flos": 25295009264640.0,
      "grad_norm": 1.919798091280079,
      "language_loss": 0.78957605,
      "learning_rate": 2.0856606697018504e-06,
      "loss": 0.80891085,
      "num_input_tokens_seen": 90015085,
      "step": 4173,
      "time_per_iteration": 2.4933574199676514
    },
    {
      "auxiliary_loss_clip": 0.01149911,
      "auxiliary_loss_mlp": 0.01029961,
      "balance_loss_clip": 1.04977274,
      "balance_loss_mlp": 1.02165246,
      "epoch": 0.5018938255275657,
      "flos": 16873203778560.0,
      "grad_norm": 2.8500797342025574,
      "language_loss": 0.72968054,
      "learning_rate": 2.084882406392297e-06,
      "loss": 0.75147927,
      "num_input_tokens_seen": 90033045,
      "step": 4174,
      "time_per_iteration": 2.478384256362915
    },
    {
      "auxiliary_loss_clip": 0.011482,
      "auxiliary_loss_mlp": 0.01027174,
      "balance_loss_clip": 1.05123162,
      "balance_loss_mlp": 1.01927614,
      "epoch": 0.5020140684182047,
      "flos": 25515429073920.0,
      "grad_norm": 2.494079964511907,
      "language_loss": 0.70868623,
      "learning_rate": 2.0841041302061496e-06,
      "loss": 0.7304399,
      "num_input_tokens_seen": 90052505,
      "step": 4175,
      "time_per_iteration": 2.5475215911865234
    },
    {
      "auxiliary_loss_clip": 0.01142428,
      "auxiliary_loss_mlp": 0.01033949,
      "balance_loss_clip": 1.04672444,
      "balance_loss_mlp": 1.02592075,
      "epoch": 0.5021343113088439,
      "flos": 23659278791040.0,
      "grad_norm": 1.9479204045875718,
      "language_loss": 0.75687671,
      "learning_rate": 2.083325841261473e-06,
      "loss": 0.77864051,
      "num_input_tokens_seen": 90071565,
      "step": 4176,
      "time_per_iteration": 2.512023687362671
    },
    {
      "auxiliary_loss_clip": 0.01145399,
      "auxiliary_loss_mlp": 0.01025994,
      "balance_loss_clip": 1.04672837,
      "balance_loss_mlp": 1.01819158,
      "epoch": 0.502254554199483,
      "flos": 24534673148160.0,
      "grad_norm": 2.429520816426436,
      "language_loss": 0.66624248,
      "learning_rate": 2.0825475396763322e-06,
      "loss": 0.68795645,
      "num_input_tokens_seen": 90092215,
      "step": 4177,
      "time_per_iteration": 3.3415985107421875
    },
    {
      "auxiliary_loss_clip": 0.01098364,
      "auxiliary_loss_mlp": 0.01029623,
      "balance_loss_clip": 1.04720092,
      "balance_loss_mlp": 1.02111149,
      "epoch": 0.502374797090122,
      "flos": 34240285607040.0,
      "grad_norm": 1.3806383394855086,
      "language_loss": 0.65732622,
      "learning_rate": 2.081769225568796e-06,
      "loss": 0.67860615,
      "num_input_tokens_seen": 90114665,
      "step": 4178,
      "time_per_iteration": 2.730598211288452
    },
    {
      "auxiliary_loss_clip": 0.01168322,
      "auxiliary_loss_mlp": 0.0103396,
      "balance_loss_clip": 1.04987943,
      "balance_loss_mlp": 1.02526331,
      "epoch": 0.5024950399807612,
      "flos": 26031106679040.0,
      "grad_norm": 1.489354769754311,
      "language_loss": 0.75870955,
      "learning_rate": 2.0809908990569327e-06,
      "loss": 0.78073239,
      "num_input_tokens_seen": 90136445,
      "step": 4179,
      "time_per_iteration": 3.362760543823242
    },
    {
      "auxiliary_loss_clip": 0.01153864,
      "auxiliary_loss_mlp": 0.01029152,
      "balance_loss_clip": 1.05078077,
      "balance_loss_mlp": 1.02071762,
      "epoch": 0.5026152828714002,
      "flos": 21252438120960.0,
      "grad_norm": 1.6577920262841377,
      "language_loss": 0.78895926,
      "learning_rate": 2.0802125602588146e-06,
      "loss": 0.81078947,
      "num_input_tokens_seen": 90155710,
      "step": 4180,
      "time_per_iteration": 3.326996088027954
    },
    {
      "auxiliary_loss_clip": 0.0118223,
      "auxiliary_loss_mlp": 0.01033684,
      "balance_loss_clip": 1.05484164,
      "balance_loss_mlp": 1.02550054,
      "epoch": 0.5027355257620393,
      "flos": 30956111245440.0,
      "grad_norm": 1.8507698758535343,
      "language_loss": 0.66382253,
      "learning_rate": 2.0794342092925146e-06,
      "loss": 0.68598163,
      "num_input_tokens_seen": 90176845,
      "step": 4181,
      "time_per_iteration": 2.507242441177368
    },
    {
      "auxiliary_loss_clip": 0.01172468,
      "auxiliary_loss_mlp": 0.01031505,
      "balance_loss_clip": 1.05574775,
      "balance_loss_mlp": 1.02333903,
      "epoch": 0.5028557686526784,
      "flos": 24791147233920.0,
      "grad_norm": 1.928388142887104,
      "language_loss": 0.68008,
      "learning_rate": 2.078655846276108e-06,
      "loss": 0.70211977,
      "num_input_tokens_seen": 90197175,
      "step": 4182,
      "time_per_iteration": 2.4964892864227295
    },
    {
      "auxiliary_loss_clip": 0.01148994,
      "auxiliary_loss_mlp": 0.01028131,
      "balance_loss_clip": 1.05040455,
      "balance_loss_mlp": 1.01983976,
      "epoch": 0.5029760115433175,
      "flos": 22966992990720.0,
      "grad_norm": 2.137462344561256,
      "language_loss": 0.68710399,
      "learning_rate": 2.0778774713276727e-06,
      "loss": 0.70887524,
      "num_input_tokens_seen": 90216650,
      "step": 4183,
      "time_per_iteration": 3.2987024784088135
    },
    {
      "auxiliary_loss_clip": 0.01164778,
      "auxiliary_loss_mlp": 0.01029592,
      "balance_loss_clip": 1.04963803,
      "balance_loss_mlp": 1.02106857,
      "epoch": 0.5030962544339566,
      "flos": 15305164485120.0,
      "grad_norm": 2.1077179033705704,
      "language_loss": 0.68073726,
      "learning_rate": 2.077099084565287e-06,
      "loss": 0.70268095,
      "num_input_tokens_seen": 90234055,
      "step": 4184,
      "time_per_iteration": 2.4242732524871826
    },
    {
      "auxiliary_loss_clip": 0.01147657,
      "auxiliary_loss_mlp": 0.01028727,
      "balance_loss_clip": 1.0478301,
      "balance_loss_mlp": 1.02064419,
      "epoch": 0.5032164973245957,
      "flos": 24494847943680.0,
      "grad_norm": 2.3040838560828556,
      "language_loss": 0.65544063,
      "learning_rate": 2.0763206861070313e-06,
      "loss": 0.67720449,
      "num_input_tokens_seen": 90253115,
      "step": 4185,
      "time_per_iteration": 2.5097901821136475
    },
    {
      "auxiliary_loss_clip": 0.01185076,
      "auxiliary_loss_mlp": 0.01032256,
      "balance_loss_clip": 1.05550051,
      "balance_loss_mlp": 1.02372706,
      "epoch": 0.5033367402152348,
      "flos": 16213452721920.0,
      "grad_norm": 2.0072480334691063,
      "language_loss": 0.75391686,
      "learning_rate": 2.0755422760709876e-06,
      "loss": 0.77609015,
      "num_input_tokens_seen": 90270515,
      "step": 4186,
      "time_per_iteration": 2.399305820465088
    },
    {
      "auxiliary_loss_clip": 0.01119322,
      "auxiliary_loss_mlp": 0.01033237,
      "balance_loss_clip": 1.04549003,
      "balance_loss_mlp": 1.024845,
      "epoch": 0.5034569831058738,
      "flos": 21391375927680.0,
      "grad_norm": 1.9119030243796675,
      "language_loss": 0.76888931,
      "learning_rate": 2.0747638545752417e-06,
      "loss": 0.79041493,
      "num_input_tokens_seen": 90289075,
      "step": 4187,
      "time_per_iteration": 2.5306806564331055
    },
    {
      "auxiliary_loss_clip": 0.01154907,
      "auxiliary_loss_mlp": 0.01026302,
      "balance_loss_clip": 1.05417061,
      "balance_loss_mlp": 1.01818991,
      "epoch": 0.503577225996513,
      "flos": 20558751690240.0,
      "grad_norm": 2.0261212518530884,
      "language_loss": 0.83200598,
      "learning_rate": 2.073985421737878e-06,
      "loss": 0.85381806,
      "num_input_tokens_seen": 90306385,
      "step": 4188,
      "time_per_iteration": 2.4842543601989746
    },
    {
      "auxiliary_loss_clip": 0.01171887,
      "auxiliary_loss_mlp": 0.01026604,
      "balance_loss_clip": 1.05360103,
      "balance_loss_mlp": 1.01855719,
      "epoch": 0.5036974688871521,
      "flos": 27229157930880.0,
      "grad_norm": 2.016543792409451,
      "language_loss": 0.74444741,
      "learning_rate": 2.0732069776769844e-06,
      "loss": 0.76643234,
      "num_input_tokens_seen": 90323795,
      "step": 4189,
      "time_per_iteration": 2.517836570739746
    },
    {
      "auxiliary_loss_clip": 0.01184411,
      "auxiliary_loss_mlp": 0.01029907,
      "balance_loss_clip": 1.05654585,
      "balance_loss_mlp": 1.02113903,
      "epoch": 0.5038177117777911,
      "flos": 20412164286720.0,
      "grad_norm": 2.0890636437834433,
      "language_loss": 0.73028249,
      "learning_rate": 2.072428522510651e-06,
      "loss": 0.75242567,
      "num_input_tokens_seen": 90340360,
      "step": 4190,
      "time_per_iteration": 2.4159305095672607
    },
    {
      "auxiliary_loss_clip": 0.01133694,
      "auxiliary_loss_mlp": 0.01028127,
      "balance_loss_clip": 1.04952073,
      "balance_loss_mlp": 1.02043819,
      "epoch": 0.5039379546684303,
      "flos": 21907987286400.0,
      "grad_norm": 2.1990393357000757,
      "language_loss": 0.76394939,
      "learning_rate": 2.071650056356968e-06,
      "loss": 0.78556764,
      "num_input_tokens_seen": 90357900,
      "step": 4191,
      "time_per_iteration": 2.5060040950775146
    },
    {
      "auxiliary_loss_clip": 0.01181761,
      "auxiliary_loss_mlp": 0.01033053,
      "balance_loss_clip": 1.05405402,
      "balance_loss_mlp": 1.02518511,
      "epoch": 0.5040581975590693,
      "flos": 20010718909440.0,
      "grad_norm": 1.872558425835631,
      "language_loss": 0.79949474,
      "learning_rate": 2.070871579334028e-06,
      "loss": 0.82164288,
      "num_input_tokens_seen": 90377010,
      "step": 4192,
      "time_per_iteration": 2.428006887435913
    },
    {
      "auxiliary_loss_clip": 0.01180101,
      "auxiliary_loss_mlp": 0.01027685,
      "balance_loss_clip": 1.05277407,
      "balance_loss_mlp": 1.01966202,
      "epoch": 0.5041784404497084,
      "flos": 20959837931520.0,
      "grad_norm": 3.869194947901325,
      "language_loss": 0.71947128,
      "learning_rate": 2.0700930915599264e-06,
      "loss": 0.74154913,
      "num_input_tokens_seen": 90396740,
      "step": 4193,
      "time_per_iteration": 2.4468986988067627
    },
    {
      "auxiliary_loss_clip": 0.01182096,
      "auxiliary_loss_mlp": 0.01030377,
      "balance_loss_clip": 1.05380321,
      "balance_loss_mlp": 1.02240825,
      "epoch": 0.5042986833403476,
      "flos": 12495082757760.0,
      "grad_norm": 1.9991825328208133,
      "language_loss": 0.78426707,
      "learning_rate": 2.0693145931527583e-06,
      "loss": 0.80639184,
      "num_input_tokens_seen": 90413220,
      "step": 4194,
      "time_per_iteration": 2.3959057331085205
    },
    {
      "auxiliary_loss_clip": 0.01148845,
      "auxiliary_loss_mlp": 0.01031137,
      "balance_loss_clip": 1.0500226,
      "balance_loss_mlp": 1.0230068,
      "epoch": 0.5044189262309866,
      "flos": 29202305788800.0,
      "grad_norm": 1.5362573874502132,
      "language_loss": 0.7786479,
      "learning_rate": 2.068536084230622e-06,
      "loss": 0.8004477,
      "num_input_tokens_seen": 90435085,
      "step": 4195,
      "time_per_iteration": 2.545450210571289
    },
    {
      "auxiliary_loss_clip": 0.01168461,
      "auxiliary_loss_mlp": 0.01034119,
      "balance_loss_clip": 1.05380082,
      "balance_loss_mlp": 1.02468359,
      "epoch": 0.5045391691216257,
      "flos": 23873198238720.0,
      "grad_norm": 1.9581666163537965,
      "language_loss": 0.88449043,
      "learning_rate": 2.067757564911616e-06,
      "loss": 0.90651625,
      "num_input_tokens_seen": 90453660,
      "step": 4196,
      "time_per_iteration": 2.4743988513946533
    },
    {
      "auxiliary_loss_clip": 0.01160993,
      "auxiliary_loss_mlp": 0.00763363,
      "balance_loss_clip": 1.05141652,
      "balance_loss_mlp": 1.00016475,
      "epoch": 0.5046594120122648,
      "flos": 24644990793600.0,
      "grad_norm": 3.0464101752092607,
      "language_loss": 0.92837322,
      "learning_rate": 2.0669790353138407e-06,
      "loss": 0.9476167,
      "num_input_tokens_seen": 90472625,
      "step": 4197,
      "time_per_iteration": 2.523627281188965
    },
    {
      "auxiliary_loss_clip": 0.01136121,
      "auxiliary_loss_mlp": 0.00763012,
      "balance_loss_clip": 1.05122519,
      "balance_loss_mlp": 1.00025356,
      "epoch": 0.5047796549029039,
      "flos": 23362835846400.0,
      "grad_norm": 2.06997513878877,
      "language_loss": 0.73087668,
      "learning_rate": 2.0662004955553995e-06,
      "loss": 0.74986798,
      "num_input_tokens_seen": 90492325,
      "step": 4198,
      "time_per_iteration": 2.530907154083252
    },
    {
      "auxiliary_loss_clip": 0.01148733,
      "auxiliary_loss_mlp": 0.01023642,
      "balance_loss_clip": 1.04856849,
      "balance_loss_mlp": 1.01592374,
      "epoch": 0.5048998977935429,
      "flos": 17304095329920.0,
      "grad_norm": 1.9088304713640585,
      "language_loss": 0.76845455,
      "learning_rate": 2.065421945754395e-06,
      "loss": 0.7901783,
      "num_input_tokens_seen": 90510055,
      "step": 4199,
      "time_per_iteration": 2.4702606201171875
    },
    {
      "auxiliary_loss_clip": 0.01127468,
      "auxiliary_loss_mlp": 0.01027449,
      "balance_loss_clip": 1.05005312,
      "balance_loss_mlp": 1.01991773,
      "epoch": 0.505020140684182,
      "flos": 34856979235200.0,
      "grad_norm": 1.592849459414632,
      "language_loss": 0.78183734,
      "learning_rate": 2.0646433860289344e-06,
      "loss": 0.80338645,
      "num_input_tokens_seen": 90528980,
      "step": 4200,
      "time_per_iteration": 2.6616246700286865
    },
    {
      "auxiliary_loss_clip": 0.01172984,
      "auxiliary_loss_mlp": 0.00763655,
      "balance_loss_clip": 1.05288565,
      "balance_loss_mlp": 1.00022566,
      "epoch": 0.5051403835748212,
      "flos": 24863974058880.0,
      "grad_norm": 2.2449877647319285,
      "language_loss": 0.82668477,
      "learning_rate": 2.0638648164971233e-06,
      "loss": 0.84605116,
      "num_input_tokens_seen": 90547445,
      "step": 4201,
      "time_per_iteration": 2.486485242843628
    },
    {
      "auxiliary_loss_clip": 0.01152563,
      "auxiliary_loss_mlp": 0.0102952,
      "balance_loss_clip": 1.05217338,
      "balance_loss_mlp": 1.02192688,
      "epoch": 0.5052606264654602,
      "flos": 20959694277120.0,
      "grad_norm": 1.8321090215994578,
      "language_loss": 0.88836199,
      "learning_rate": 2.06308623727707e-06,
      "loss": 0.91018283,
      "num_input_tokens_seen": 90567545,
      "step": 4202,
      "time_per_iteration": 2.488079071044922
    },
    {
      "auxiliary_loss_clip": 0.01162169,
      "auxiliary_loss_mlp": 0.01027565,
      "balance_loss_clip": 1.0506134,
      "balance_loss_mlp": 1.01898205,
      "epoch": 0.5053808693560993,
      "flos": 19642382893440.0,
      "grad_norm": 2.260185856820954,
      "language_loss": 0.76533687,
      "learning_rate": 2.0623076484868846e-06,
      "loss": 0.78723413,
      "num_input_tokens_seen": 90585000,
      "step": 4203,
      "time_per_iteration": 3.2754476070404053
    },
    {
      "auxiliary_loss_clip": 0.01057443,
      "auxiliary_loss_mlp": 0.01004305,
      "balance_loss_clip": 1.02130198,
      "balance_loss_mlp": 1.00317848,
      "epoch": 0.5055011122467384,
      "flos": 67504915019520.0,
      "grad_norm": 0.8469472341861977,
      "language_loss": 0.60696357,
      "learning_rate": 2.061529050244679e-06,
      "loss": 0.627581,
      "num_input_tokens_seen": 90644745,
      "step": 4204,
      "time_per_iteration": 3.034193515777588
    },
    {
      "auxiliary_loss_clip": 0.0114668,
      "auxiliary_loss_mlp": 0.01022801,
      "balance_loss_clip": 1.05068588,
      "balance_loss_mlp": 1.01427186,
      "epoch": 0.5056213551373775,
      "flos": 16872952383360.0,
      "grad_norm": 2.2513044406990224,
      "language_loss": 0.74369597,
      "learning_rate": 2.060750442668565e-06,
      "loss": 0.76539075,
      "num_input_tokens_seen": 90662500,
      "step": 4205,
      "time_per_iteration": 2.5891470909118652
    },
    {
      "auxiliary_loss_clip": 0.01169881,
      "auxiliary_loss_mlp": 0.01030287,
      "balance_loss_clip": 1.05556941,
      "balance_loss_mlp": 1.02191257,
      "epoch": 0.5057415980280165,
      "flos": 15334179696000.0,
      "grad_norm": 2.49031724293243,
      "language_loss": 0.64106423,
      "learning_rate": 2.059971825876657e-06,
      "loss": 0.66306591,
      "num_input_tokens_seen": 90677010,
      "step": 4206,
      "time_per_iteration": 2.414470672607422
    },
    {
      "auxiliary_loss_clip": 0.01170493,
      "auxiliary_loss_mlp": 0.01026891,
      "balance_loss_clip": 1.05423498,
      "balance_loss_mlp": 1.01861501,
      "epoch": 0.5058618409186557,
      "flos": 19025976574080.0,
      "grad_norm": 1.9791528682941184,
      "language_loss": 0.7663753,
      "learning_rate": 2.0591931999870713e-06,
      "loss": 0.78834915,
      "num_input_tokens_seen": 90695935,
      "step": 4207,
      "time_per_iteration": 4.025703191757202
    },
    {
      "auxiliary_loss_clip": 0.01065987,
      "auxiliary_loss_mlp": 0.01001798,
      "balance_loss_clip": 1.0204407,
      "balance_loss_mlp": 1.00072539,
      "epoch": 0.5059820838092948,
      "flos": 63453114080640.0,
      "grad_norm": 0.8289903686563834,
      "language_loss": 0.57591641,
      "learning_rate": 2.0584145651179234e-06,
      "loss": 0.59659427,
      "num_input_tokens_seen": 90751645,
      "step": 4208,
      "time_per_iteration": 3.039763927459717
    },
    {
      "auxiliary_loss_clip": 0.01155303,
      "auxiliary_loss_mlp": 0.00762425,
      "balance_loss_clip": 1.05463147,
      "balance_loss_mlp": 1.00016546,
      "epoch": 0.5061023266999338,
      "flos": 15441803821440.0,
      "grad_norm": 2.5622089982981513,
      "language_loss": 0.80391634,
      "learning_rate": 2.0576359213873327e-06,
      "loss": 0.82309365,
      "num_input_tokens_seen": 90766795,
      "step": 4209,
      "time_per_iteration": 2.4564402103424072
    },
    {
      "auxiliary_loss_clip": 0.01160104,
      "auxiliary_loss_mlp": 0.0102846,
      "balance_loss_clip": 1.04809022,
      "balance_loss_mlp": 1.02016318,
      "epoch": 0.506222569590573,
      "flos": 22451063990400.0,
      "grad_norm": 4.534793872075395,
      "language_loss": 0.70908451,
      "learning_rate": 2.056857268913419e-06,
      "loss": 0.73097014,
      "num_input_tokens_seen": 90786845,
      "step": 4210,
      "time_per_iteration": 3.2950143814086914
    },
    {
      "auxiliary_loss_clip": 0.01169104,
      "auxiliary_loss_mlp": 0.01027517,
      "balance_loss_clip": 1.05562043,
      "balance_loss_mlp": 1.0197382,
      "epoch": 0.506342812481212,
      "flos": 17558665994880.0,
      "grad_norm": 2.105991022313912,
      "language_loss": 0.83813787,
      "learning_rate": 2.056078607814303e-06,
      "loss": 0.86010408,
      "num_input_tokens_seen": 90802630,
      "step": 4211,
      "time_per_iteration": 2.411654233932495
    },
    {
      "auxiliary_loss_clip": 0.01168489,
      "auxiliary_loss_mlp": 0.01023847,
      "balance_loss_clip": 1.05369461,
      "balance_loss_mlp": 1.01562178,
      "epoch": 0.5064630553718511,
      "flos": 23402050519680.0,
      "grad_norm": 3.9351652022734696,
      "language_loss": 0.78508145,
      "learning_rate": 2.055299938208106e-06,
      "loss": 0.80700481,
      "num_input_tokens_seen": 90823620,
      "step": 4212,
      "time_per_iteration": 2.4986555576324463
    },
    {
      "auxiliary_loss_clip": 0.01174589,
      "auxiliary_loss_mlp": 0.01031573,
      "balance_loss_clip": 1.05630445,
      "balance_loss_mlp": 1.02319312,
      "epoch": 0.5065832982624903,
      "flos": 23987035416960.0,
      "grad_norm": 1.5746707234394322,
      "language_loss": 0.86405993,
      "learning_rate": 2.0545212602129526e-06,
      "loss": 0.88612157,
      "num_input_tokens_seen": 90843475,
      "step": 4213,
      "time_per_iteration": 2.480971574783325
    },
    {
      "auxiliary_loss_clip": 0.01143534,
      "auxiliary_loss_mlp": 0.01032386,
      "balance_loss_clip": 1.04665661,
      "balance_loss_mlp": 1.02359986,
      "epoch": 0.5067035411531293,
      "flos": 21503058289920.0,
      "grad_norm": 2.0019384238611893,
      "language_loss": 0.66305214,
      "learning_rate": 2.0537425739469673e-06,
      "loss": 0.68481135,
      "num_input_tokens_seen": 90862410,
      "step": 4214,
      "time_per_iteration": 2.48228120803833
    },
    {
      "auxiliary_loss_clip": 0.01074146,
      "auxiliary_loss_mlp": 0.01002184,
      "balance_loss_clip": 1.02130651,
      "balance_loss_mlp": 1.00107503,
      "epoch": 0.5068237840437684,
      "flos": 65934397687680.0,
      "grad_norm": 0.8419990211827253,
      "language_loss": 0.59471959,
      "learning_rate": 2.052963879528276e-06,
      "loss": 0.61548293,
      "num_input_tokens_seen": 90922280,
      "step": 4215,
      "time_per_iteration": 3.0256857872009277
    },
    {
      "auxiliary_loss_clip": 0.01169888,
      "auxiliary_loss_mlp": 0.0102373,
      "balance_loss_clip": 1.05455613,
      "balance_loss_mlp": 1.0157249,
      "epoch": 0.5069440269344075,
      "flos": 27264206626560.0,
      "grad_norm": 2.0111282731624147,
      "language_loss": 0.76598889,
      "learning_rate": 2.052185177075007e-06,
      "loss": 0.787925,
      "num_input_tokens_seen": 90941850,
      "step": 4216,
      "time_per_iteration": 2.5299932956695557
    },
    {
      "auxiliary_loss_clip": 0.0117132,
      "auxiliary_loss_mlp": 0.01031496,
      "balance_loss_clip": 1.05420351,
      "balance_loss_mlp": 1.02381301,
      "epoch": 0.5070642698250466,
      "flos": 23366319465600.0,
      "grad_norm": 1.6919714070192025,
      "language_loss": 0.8306703,
      "learning_rate": 2.051406466705288e-06,
      "loss": 0.85269845,
      "num_input_tokens_seen": 90961390,
      "step": 4217,
      "time_per_iteration": 2.4793636798858643
    },
    {
      "auxiliary_loss_clip": 0.01180918,
      "auxiliary_loss_mlp": 0.01027469,
      "balance_loss_clip": 1.05301011,
      "balance_loss_mlp": 1.01961899,
      "epoch": 0.5071845127156857,
      "flos": 20340127560960.0,
      "grad_norm": 1.829978519391727,
      "language_loss": 0.81053251,
      "learning_rate": 2.0506277485372486e-06,
      "loss": 0.83261639,
      "num_input_tokens_seen": 90980215,
      "step": 4218,
      "time_per_iteration": 2.4224464893341064
    },
    {
      "auxiliary_loss_clip": 0.01164191,
      "auxiliary_loss_mlp": 0.01032584,
      "balance_loss_clip": 1.05398405,
      "balance_loss_mlp": 1.02469254,
      "epoch": 0.5073047556063248,
      "flos": 12092955022080.0,
      "grad_norm": 1.8871532714754766,
      "language_loss": 0.67037475,
      "learning_rate": 2.04984902268902e-06,
      "loss": 0.69234252,
      "num_input_tokens_seen": 90997415,
      "step": 4219,
      "time_per_iteration": 2.4345171451568604
    },
    {
      "auxiliary_loss_clip": 0.01172974,
      "auxiliary_loss_mlp": 0.01029291,
      "balance_loss_clip": 1.05083847,
      "balance_loss_mlp": 1.02015352,
      "epoch": 0.5074249984969639,
      "flos": 19682854542720.0,
      "grad_norm": 2.601554207508566,
      "language_loss": 0.75460339,
      "learning_rate": 2.0490702892787345e-06,
      "loss": 0.77662605,
      "num_input_tokens_seen": 91016475,
      "step": 4220,
      "time_per_iteration": 2.4609060287475586
    },
    {
      "auxiliary_loss_clip": 0.01160354,
      "auxiliary_loss_mlp": 0.01031914,
      "balance_loss_clip": 1.05006528,
      "balance_loss_mlp": 1.02389121,
      "epoch": 0.5075452413876029,
      "flos": 28765703975040.0,
      "grad_norm": 2.192762152780997,
      "language_loss": 0.62366754,
      "learning_rate": 2.0482915484245246e-06,
      "loss": 0.64559019,
      "num_input_tokens_seen": 91038095,
      "step": 4221,
      "time_per_iteration": 2.505283832550049
    },
    {
      "auxiliary_loss_clip": 0.01117755,
      "auxiliary_loss_mlp": 0.01031162,
      "balance_loss_clip": 1.04777181,
      "balance_loss_mlp": 1.02243543,
      "epoch": 0.5076654842782421,
      "flos": 20339445202560.0,
      "grad_norm": 2.369725711986272,
      "language_loss": 0.84143519,
      "learning_rate": 2.047512800244526e-06,
      "loss": 0.86292434,
      "num_input_tokens_seen": 91053360,
      "step": 4222,
      "time_per_iteration": 2.531452178955078
    },
    {
      "auxiliary_loss_clip": 0.01167282,
      "auxiliary_loss_mlp": 0.01025243,
      "balance_loss_clip": 1.05313158,
      "balance_loss_mlp": 1.01694572,
      "epoch": 0.5077857271688812,
      "flos": 26359653404160.0,
      "grad_norm": 3.7636683653919003,
      "language_loss": 0.78961837,
      "learning_rate": 2.046734044856873e-06,
      "loss": 0.81154358,
      "num_input_tokens_seen": 91072770,
      "step": 4223,
      "time_per_iteration": 2.49385142326355
    },
    {
      "auxiliary_loss_clip": 0.01167541,
      "auxiliary_loss_mlp": 0.01029025,
      "balance_loss_clip": 1.05351055,
      "balance_loss_mlp": 1.02124691,
      "epoch": 0.5079059700595202,
      "flos": 21798962530560.0,
      "grad_norm": 1.897913170136811,
      "language_loss": 0.81250632,
      "learning_rate": 2.045955282379702e-06,
      "loss": 0.83447206,
      "num_input_tokens_seen": 91091430,
      "step": 4224,
      "time_per_iteration": 2.4464786052703857
    },
    {
      "auxiliary_loss_clip": 0.01164505,
      "auxiliary_loss_mlp": 0.01030386,
      "balance_loss_clip": 1.04941392,
      "balance_loss_mlp": 1.02194023,
      "epoch": 0.5080262129501594,
      "flos": 13187943175680.0,
      "grad_norm": 2.6944983267968388,
      "language_loss": 0.76109529,
      "learning_rate": 2.045176512931152e-06,
      "loss": 0.78304422,
      "num_input_tokens_seen": 91106060,
      "step": 4225,
      "time_per_iteration": 2.4173262119293213
    },
    {
      "auxiliary_loss_clip": 0.01143044,
      "auxiliary_loss_mlp": 0.01024754,
      "balance_loss_clip": 1.04974222,
      "balance_loss_mlp": 1.01703537,
      "epoch": 0.5081464558407984,
      "flos": 25301473712640.0,
      "grad_norm": 2.337237949388475,
      "language_loss": 0.76053709,
      "learning_rate": 2.0443977366293604e-06,
      "loss": 0.78221506,
      "num_input_tokens_seen": 91124100,
      "step": 4226,
      "time_per_iteration": 2.541816473007202
    },
    {
      "auxiliary_loss_clip": 0.01111162,
      "auxiliary_loss_mlp": 0.01032203,
      "balance_loss_clip": 1.04539144,
      "balance_loss_mlp": 1.02336955,
      "epoch": 0.5082666987314375,
      "flos": 30951226995840.0,
      "grad_norm": 1.6492625686937972,
      "language_loss": 0.76927078,
      "learning_rate": 2.043618953592468e-06,
      "loss": 0.79070437,
      "num_input_tokens_seen": 91146555,
      "step": 4227,
      "time_per_iteration": 2.661424160003662
    },
    {
      "auxiliary_loss_clip": 0.01155349,
      "auxiliary_loss_mlp": 0.01030147,
      "balance_loss_clip": 1.05310082,
      "balance_loss_mlp": 1.02155852,
      "epoch": 0.5083869416220766,
      "flos": 19682495406720.0,
      "grad_norm": 1.5038381746403708,
      "language_loss": 0.81146085,
      "learning_rate": 2.0428401639386144e-06,
      "loss": 0.83331573,
      "num_input_tokens_seen": 91167120,
      "step": 4228,
      "time_per_iteration": 2.524024724960327
    },
    {
      "auxiliary_loss_clip": 0.0105483,
      "auxiliary_loss_mlp": 0.01003571,
      "balance_loss_clip": 1.02039552,
      "balance_loss_mlp": 1.00253987,
      "epoch": 0.5085071845127157,
      "flos": 71817535589760.0,
      "grad_norm": 0.8183674946183168,
      "language_loss": 0.58113456,
      "learning_rate": 2.042061367785943e-06,
      "loss": 0.60171854,
      "num_input_tokens_seen": 91220260,
      "step": 4229,
      "time_per_iteration": 3.0499186515808105
    },
    {
      "auxiliary_loss_clip": 0.01143538,
      "auxiliary_loss_mlp": 0.0103079,
      "balance_loss_clip": 1.04867339,
      "balance_loss_mlp": 1.022493,
      "epoch": 0.5086274274033548,
      "flos": 35951608252800.0,
      "grad_norm": 2.068272594458992,
      "language_loss": 0.74709278,
      "learning_rate": 2.041282565252594e-06,
      "loss": 0.76883602,
      "num_input_tokens_seen": 91240425,
      "step": 4230,
      "time_per_iteration": 3.4399161338806152
    },
    {
      "auxiliary_loss_clip": 0.01140722,
      "auxiliary_loss_mlp": 0.01027647,
      "balance_loss_clip": 1.04955876,
      "balance_loss_mlp": 1.01965976,
      "epoch": 0.5087476702939938,
      "flos": 23513732881920.0,
      "grad_norm": 1.9248725173440195,
      "language_loss": 0.77212548,
      "learning_rate": 2.040503756456714e-06,
      "loss": 0.79380912,
      "num_input_tokens_seen": 91259635,
      "step": 4231,
      "time_per_iteration": 2.5535709857940674
    },
    {
      "auxiliary_loss_clip": 0.01160407,
      "auxiliary_loss_mlp": 0.0103112,
      "balance_loss_clip": 1.04937959,
      "balance_loss_mlp": 1.02263844,
      "epoch": 0.508867913184633,
      "flos": 15122091841920.0,
      "grad_norm": 1.8720259744479968,
      "language_loss": 0.78788388,
      "learning_rate": 2.0397249415164456e-06,
      "loss": 0.80979919,
      "num_input_tokens_seen": 91276990,
      "step": 4232,
      "time_per_iteration": 2.434131622314453
    },
    {
      "auxiliary_loss_clip": 0.01145754,
      "auxiliary_loss_mlp": 0.01027276,
      "balance_loss_clip": 1.04710853,
      "balance_loss_mlp": 1.01895523,
      "epoch": 0.508988156075272,
      "flos": 25885309374720.0,
      "grad_norm": 1.7991617762802905,
      "language_loss": 0.80021429,
      "learning_rate": 2.0389461205499354e-06,
      "loss": 0.82194459,
      "num_input_tokens_seen": 91296125,
      "step": 4233,
      "time_per_iteration": 3.3895139694213867
    },
    {
      "auxiliary_loss_clip": 0.01141641,
      "auxiliary_loss_mlp": 0.01029267,
      "balance_loss_clip": 1.04964304,
      "balance_loss_mlp": 1.02150595,
      "epoch": 0.5091083989659111,
      "flos": 13844857057920.0,
      "grad_norm": 2.59532224720696,
      "language_loss": 0.73436964,
      "learning_rate": 2.03816729367533e-06,
      "loss": 0.75607872,
      "num_input_tokens_seen": 91314280,
      "step": 4234,
      "time_per_iteration": 3.333515167236328
    },
    {
      "auxiliary_loss_clip": 0.01158594,
      "auxiliary_loss_mlp": 0.0103558,
      "balance_loss_clip": 1.05407727,
      "balance_loss_mlp": 1.02714014,
      "epoch": 0.5092286418565503,
      "flos": 21104881050240.0,
      "grad_norm": 2.296637304685272,
      "language_loss": 0.71788311,
      "learning_rate": 2.0373884610107765e-06,
      "loss": 0.73982489,
      "num_input_tokens_seen": 91334595,
      "step": 4235,
      "time_per_iteration": 2.520890235900879
    },
    {
      "auxiliary_loss_clip": 0.01168491,
      "auxiliary_loss_mlp": 0.01027052,
      "balance_loss_clip": 1.04858708,
      "balance_loss_mlp": 1.01885009,
      "epoch": 0.5093488847471893,
      "flos": 18621298972800.0,
      "grad_norm": 3.6892371380204545,
      "language_loss": 0.69298148,
      "learning_rate": 2.0366096226744225e-06,
      "loss": 0.71493697,
      "num_input_tokens_seen": 91349790,
      "step": 4236,
      "time_per_iteration": 2.4222216606140137
    },
    {
      "auxiliary_loss_clip": 0.01157311,
      "auxiliary_loss_mlp": 0.01035518,
      "balance_loss_clip": 1.05011976,
      "balance_loss_mlp": 1.02727509,
      "epoch": 0.5094691276378284,
      "flos": 23803783205760.0,
      "grad_norm": 1.780245047229767,
      "language_loss": 0.7679432,
      "learning_rate": 2.035830778784418e-06,
      "loss": 0.78987145,
      "num_input_tokens_seen": 91370465,
      "step": 4237,
      "time_per_iteration": 3.2091763019561768
    },
    {
      "auxiliary_loss_clip": 0.01156599,
      "auxiliary_loss_mlp": 0.01025372,
      "balance_loss_clip": 1.05279922,
      "balance_loss_mlp": 1.01682496,
      "epoch": 0.5095893705284675,
      "flos": 17420410546560.0,
      "grad_norm": 1.8942234409410152,
      "language_loss": 0.79986608,
      "learning_rate": 2.0350519294589134e-06,
      "loss": 0.82168573,
      "num_input_tokens_seen": 91388505,
      "step": 4238,
      "time_per_iteration": 2.4703798294067383
    },
    {
      "auxiliary_loss_clip": 0.01118546,
      "auxiliary_loss_mlp": 0.01023291,
      "balance_loss_clip": 1.04449272,
      "balance_loss_mlp": 1.01473761,
      "epoch": 0.5097096134191066,
      "flos": 25849362839040.0,
      "grad_norm": 1.6666450294808899,
      "language_loss": 0.82938766,
      "learning_rate": 2.0342730748160588e-06,
      "loss": 0.850806,
      "num_input_tokens_seen": 91408970,
      "step": 4239,
      "time_per_iteration": 2.5937604904174805
    },
    {
      "auxiliary_loss_clip": 0.01151448,
      "auxiliary_loss_mlp": 0.01028605,
      "balance_loss_clip": 1.04785407,
      "balance_loss_mlp": 1.02061224,
      "epoch": 0.5098298563097456,
      "flos": 27745122844800.0,
      "grad_norm": 2.869188560325743,
      "language_loss": 0.706734,
      "learning_rate": 2.033494214974006e-06,
      "loss": 0.72853452,
      "num_input_tokens_seen": 91430115,
      "step": 4240,
      "time_per_iteration": 2.5369811058044434
    },
    {
      "auxiliary_loss_clip": 0.01141895,
      "auxiliary_loss_mlp": 0.01029812,
      "balance_loss_clip": 1.04915464,
      "balance_loss_mlp": 1.02198005,
      "epoch": 0.5099500992003848,
      "flos": 21358913011200.0,
      "grad_norm": 1.703418559764768,
      "language_loss": 0.83784044,
      "learning_rate": 2.0327153500509067e-06,
      "loss": 0.85955751,
      "num_input_tokens_seen": 91449140,
      "step": 4241,
      "time_per_iteration": 2.4743709564208984
    },
    {
      "auxiliary_loss_clip": 0.01155068,
      "auxiliary_loss_mlp": 0.01030003,
      "balance_loss_clip": 1.05198455,
      "balance_loss_mlp": 1.02207017,
      "epoch": 0.5100703420910239,
      "flos": 19865999013120.0,
      "grad_norm": 2.0116333807045126,
      "language_loss": 0.84781206,
      "learning_rate": 2.031936480164916e-06,
      "loss": 0.86966276,
      "num_input_tokens_seen": 91466880,
      "step": 4242,
      "time_per_iteration": 2.4793429374694824
    },
    {
      "auxiliary_loss_clip": 0.01149133,
      "auxiliary_loss_mlp": 0.01028188,
      "balance_loss_clip": 1.05286491,
      "balance_loss_mlp": 1.02001047,
      "epoch": 0.5101905849816629,
      "flos": 24648797635200.0,
      "grad_norm": 4.389376168812962,
      "language_loss": 0.79635608,
      "learning_rate": 2.0311576054341857e-06,
      "loss": 0.8181293,
      "num_input_tokens_seen": 91487495,
      "step": 4243,
      "time_per_iteration": 2.5181028842926025
    },
    {
      "auxiliary_loss_clip": 0.01183291,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 1.05620539,
      "balance_loss_mlp": 1.01855731,
      "epoch": 0.5103108278723021,
      "flos": 22930076787840.0,
      "grad_norm": 1.7259891666790357,
      "language_loss": 0.62566996,
      "learning_rate": 2.0303787259768715e-06,
      "loss": 0.64777017,
      "num_input_tokens_seen": 91508395,
      "step": 4244,
      "time_per_iteration": 2.4243245124816895
    },
    {
      "auxiliary_loss_clip": 0.0115381,
      "auxiliary_loss_mlp": 0.01029591,
      "balance_loss_clip": 1.05220366,
      "balance_loss_mlp": 1.0216701,
      "epoch": 0.5104310707629411,
      "flos": 21506613736320.0,
      "grad_norm": 2.111999253410287,
      "language_loss": 0.6946708,
      "learning_rate": 2.0295998419111294e-06,
      "loss": 0.71650481,
      "num_input_tokens_seen": 91525685,
      "step": 4245,
      "time_per_iteration": 2.466531991958618
    },
    {
      "auxiliary_loss_clip": 0.01111886,
      "auxiliary_loss_mlp": 0.01031936,
      "balance_loss_clip": 1.04429841,
      "balance_loss_mlp": 1.02376997,
      "epoch": 0.5105513136535802,
      "flos": 14903180403840.0,
      "grad_norm": 3.5010966158179557,
      "language_loss": 0.733778,
      "learning_rate": 2.028820953355115e-06,
      "loss": 0.75521624,
      "num_input_tokens_seen": 91543785,
      "step": 4246,
      "time_per_iteration": 2.544123888015747
    },
    {
      "auxiliary_loss_clip": 0.01158315,
      "auxiliary_loss_mlp": 0.01029634,
      "balance_loss_clip": 1.04906607,
      "balance_loss_mlp": 1.02098274,
      "epoch": 0.5106715565442194,
      "flos": 22602212421120.0,
      "grad_norm": 1.6737099940698206,
      "language_loss": 0.7864427,
      "learning_rate": 2.0280420604269834e-06,
      "loss": 0.80832219,
      "num_input_tokens_seen": 91563325,
      "step": 4247,
      "time_per_iteration": 2.4977834224700928
    },
    {
      "auxiliary_loss_clip": 0.01067746,
      "auxiliary_loss_mlp": 0.0100102,
      "balance_loss_clip": 1.01934361,
      "balance_loss_mlp": 1.00002432,
      "epoch": 0.5107917994348584,
      "flos": 71027645558400.0,
      "grad_norm": 0.7001215243755687,
      "language_loss": 0.58915651,
      "learning_rate": 2.027263163244895e-06,
      "loss": 0.60984409,
      "num_input_tokens_seen": 91632450,
      "step": 4248,
      "time_per_iteration": 3.191516160964966
    },
    {
      "auxiliary_loss_clip": 0.01166075,
      "auxiliary_loss_mlp": 0.01028669,
      "balance_loss_clip": 1.05384707,
      "balance_loss_mlp": 1.02083135,
      "epoch": 0.5109120423254975,
      "flos": 24827416992000.0,
      "grad_norm": 1.6677924342726513,
      "language_loss": 0.74236333,
      "learning_rate": 2.026484261927005e-06,
      "loss": 0.76431084,
      "num_input_tokens_seen": 91651945,
      "step": 4249,
      "time_per_iteration": 2.4837357997894287
    },
    {
      "auxiliary_loss_clip": 0.01174665,
      "auxiliary_loss_mlp": 0.01027265,
      "balance_loss_clip": 1.05639207,
      "balance_loss_mlp": 1.01845551,
      "epoch": 0.5110322852161366,
      "flos": 21247661612160.0,
      "grad_norm": 2.009971883001086,
      "language_loss": 0.74127877,
      "learning_rate": 2.025705356591475e-06,
      "loss": 0.76329809,
      "num_input_tokens_seen": 91669635,
      "step": 4250,
      "time_per_iteration": 2.4449050426483154
    },
    {
      "auxiliary_loss_clip": 0.01045319,
      "auxiliary_loss_mlp": 0.00753165,
      "balance_loss_clip": 1.01980639,
      "balance_loss_mlp": 0.99992067,
      "epoch": 0.5111525281067757,
      "flos": 66457114358400.0,
      "grad_norm": 0.755468175708005,
      "language_loss": 0.57979679,
      "learning_rate": 2.024926447356462e-06,
      "loss": 0.59778166,
      "num_input_tokens_seen": 91731920,
      "step": 4251,
      "time_per_iteration": 3.0303187370300293
    },
    {
      "auxiliary_loss_clip": 0.01165466,
      "auxiliary_loss_mlp": 0.01033915,
      "balance_loss_clip": 1.05213022,
      "balance_loss_mlp": 1.02505744,
      "epoch": 0.5112727709974147,
      "flos": 14866731077760.0,
      "grad_norm": 2.3387553783665362,
      "language_loss": 0.78711414,
      "learning_rate": 2.024147534340127e-06,
      "loss": 0.8091079,
      "num_input_tokens_seen": 91749780,
      "step": 4252,
      "time_per_iteration": 2.4719812870025635
    },
    {
      "auxiliary_loss_clip": 0.01148648,
      "auxiliary_loss_mlp": 0.01026464,
      "balance_loss_clip": 1.04688764,
      "balance_loss_mlp": 1.01856589,
      "epoch": 0.5113930138880539,
      "flos": 21177600134400.0,
      "grad_norm": 1.7655880521304388,
      "language_loss": 0.79851246,
      "learning_rate": 2.02336861766063e-06,
      "loss": 0.82026362,
      "num_input_tokens_seen": 91768840,
      "step": 4253,
      "time_per_iteration": 2.5480082035064697
    },
    {
      "auxiliary_loss_clip": 0.01174237,
      "auxiliary_loss_mlp": 0.01027646,
      "balance_loss_clip": 1.05411017,
      "balance_loss_mlp": 1.01945055,
      "epoch": 0.511513256778693,
      "flos": 20409111630720.0,
      "grad_norm": 1.7485085200015165,
      "language_loss": 0.78985453,
      "learning_rate": 2.0225896974361327e-06,
      "loss": 0.81187338,
      "num_input_tokens_seen": 91788945,
      "step": 4254,
      "time_per_iteration": 2.5228259563446045
    },
    {
      "auxiliary_loss_clip": 0.01052463,
      "auxiliary_loss_mlp": 0.01003924,
      "balance_loss_clip": 1.02267146,
      "balance_loss_mlp": 1.0027678,
      "epoch": 0.511633499669332,
      "flos": 69879975131520.0,
      "grad_norm": 0.8621654082889129,
      "language_loss": 0.59980214,
      "learning_rate": 2.0218107737847962e-06,
      "loss": 0.62036598,
      "num_input_tokens_seen": 91850990,
      "step": 4255,
      "time_per_iteration": 3.131788730621338
    },
    {
      "auxiliary_loss_clip": 0.01181696,
      "auxiliary_loss_mlp": 0.01028919,
      "balance_loss_clip": 1.05496347,
      "balance_loss_mlp": 1.02102172,
      "epoch": 0.5117537425599712,
      "flos": 24097855852800.0,
      "grad_norm": 1.8905596839496766,
      "language_loss": 0.7469821,
      "learning_rate": 2.0210318468247826e-06,
      "loss": 0.76908821,
      "num_input_tokens_seen": 91869960,
      "step": 4256,
      "time_per_iteration": 3.301748275756836
    },
    {
      "auxiliary_loss_clip": 0.01152331,
      "auxiliary_loss_mlp": 0.01025683,
      "balance_loss_clip": 1.04946208,
      "balance_loss_mlp": 1.01835787,
      "epoch": 0.5118739854506102,
      "flos": 20959550622720.0,
      "grad_norm": 1.766194154064807,
      "language_loss": 0.81982529,
      "learning_rate": 2.020252916674255e-06,
      "loss": 0.84160548,
      "num_input_tokens_seen": 91889075,
      "step": 4257,
      "time_per_iteration": 2.518026113510132
    },
    {
      "auxiliary_loss_clip": 0.01167079,
      "auxiliary_loss_mlp": 0.01026932,
      "balance_loss_clip": 1.05014861,
      "balance_loss_mlp": 1.01886749,
      "epoch": 0.5119942283412493,
      "flos": 17457326749440.0,
      "grad_norm": 1.8069203388510389,
      "language_loss": 0.81198859,
      "learning_rate": 2.019473983451375e-06,
      "loss": 0.8339287,
      "num_input_tokens_seen": 91907495,
      "step": 4258,
      "time_per_iteration": 2.4647676944732666
    },
    {
      "auxiliary_loss_clip": 0.01144511,
      "auxiliary_loss_mlp": 0.01027322,
      "balance_loss_clip": 1.04867578,
      "balance_loss_mlp": 1.01919746,
      "epoch": 0.5121144712318885,
      "flos": 21066743784960.0,
      "grad_norm": 2.0220925179159495,
      "language_loss": 0.71626347,
      "learning_rate": 2.0186950472743076e-06,
      "loss": 0.7379818,
      "num_input_tokens_seen": 91927400,
      "step": 4259,
      "time_per_iteration": 2.5598652362823486
    },
    {
      "auxiliary_loss_clip": 0.01181697,
      "auxiliary_loss_mlp": 0.01027339,
      "balance_loss_clip": 1.0537957,
      "balance_loss_mlp": 1.01917958,
      "epoch": 0.5122347141225275,
      "flos": 19860791541120.0,
      "grad_norm": 1.6777330566778843,
      "language_loss": 0.74287784,
      "learning_rate": 2.0179161082612162e-06,
      "loss": 0.76496816,
      "num_input_tokens_seen": 91946790,
      "step": 4260,
      "time_per_iteration": 3.3007566928863525
    },
    {
      "auxiliary_loss_clip": 0.01147798,
      "auxiliary_loss_mlp": 0.01027528,
      "balance_loss_clip": 1.04789376,
      "balance_loss_mlp": 1.01999366,
      "epoch": 0.5123549570131666,
      "flos": 22528487756160.0,
      "grad_norm": 1.9901296992834492,
      "language_loss": 0.7300002,
      "learning_rate": 2.017137166530266e-06,
      "loss": 0.75175345,
      "num_input_tokens_seen": 91966325,
      "step": 4261,
      "time_per_iteration": 3.289705514907837
    },
    {
      "auxiliary_loss_clip": 0.01154071,
      "auxiliary_loss_mlp": 0.01029635,
      "balance_loss_clip": 1.04925203,
      "balance_loss_mlp": 1.02190161,
      "epoch": 0.5124751999038056,
      "flos": 20333375804160.0,
      "grad_norm": 2.0714490609464873,
      "language_loss": 0.79990327,
      "learning_rate": 2.0163582221996213e-06,
      "loss": 0.82174033,
      "num_input_tokens_seen": 91984700,
      "step": 4262,
      "time_per_iteration": 2.6133012771606445
    },
    {
      "auxiliary_loss_clip": 0.01153555,
      "auxiliary_loss_mlp": 0.01027758,
      "balance_loss_clip": 1.05056953,
      "balance_loss_mlp": 1.0198065,
      "epoch": 0.5125954427944448,
      "flos": 39785970211200.0,
      "grad_norm": 2.4268200798706934,
      "language_loss": 0.68229413,
      "learning_rate": 2.015579275387446e-06,
      "loss": 0.70410722,
      "num_input_tokens_seen": 92010020,
      "step": 4263,
      "time_per_iteration": 3.3962647914886475
    },
    {
      "auxiliary_loss_clip": 0.01144284,
      "auxiliary_loss_mlp": 0.01027971,
      "balance_loss_clip": 1.05089891,
      "balance_loss_mlp": 1.02004635,
      "epoch": 0.5127156856850839,
      "flos": 29205394358400.0,
      "grad_norm": 1.9084650485949497,
      "language_loss": 0.68326336,
      "learning_rate": 2.0148003262119085e-06,
      "loss": 0.70498592,
      "num_input_tokens_seen": 92030990,
      "step": 4264,
      "time_per_iteration": 2.5450198650360107
    },
    {
      "auxiliary_loss_clip": 0.01137384,
      "auxiliary_loss_mlp": 0.01028729,
      "balance_loss_clip": 1.04927349,
      "balance_loss_mlp": 1.0205214,
      "epoch": 0.5128359285757229,
      "flos": 13553693412480.0,
      "grad_norm": 1.8918325748929428,
      "language_loss": 0.76884156,
      "learning_rate": 2.0140213747911728e-06,
      "loss": 0.79050267,
      "num_input_tokens_seen": 92049525,
      "step": 4265,
      "time_per_iteration": 2.5075795650482178
    },
    {
      "auxiliary_loss_clip": 0.01135409,
      "auxiliary_loss_mlp": 0.01030036,
      "balance_loss_clip": 1.04993486,
      "balance_loss_mlp": 1.02166152,
      "epoch": 0.5129561714663621,
      "flos": 25192089820800.0,
      "grad_norm": 1.9729055609448587,
      "language_loss": 0.80425715,
      "learning_rate": 2.013242421243406e-06,
      "loss": 0.82591158,
      "num_input_tokens_seen": 92068430,
      "step": 4266,
      "time_per_iteration": 2.5770020484924316
    },
    {
      "auxiliary_loss_clip": 0.01125901,
      "auxiliary_loss_mlp": 0.01021837,
      "balance_loss_clip": 1.05133057,
      "balance_loss_mlp": 1.01396894,
      "epoch": 0.5130764143570011,
      "flos": 18150223080960.0,
      "grad_norm": 1.4989578344089376,
      "language_loss": 0.7892133,
      "learning_rate": 2.012463465686774e-06,
      "loss": 0.81069064,
      "num_input_tokens_seen": 92088180,
      "step": 4267,
      "time_per_iteration": 2.601551055908203
    },
    {
      "auxiliary_loss_clip": 0.01050355,
      "auxiliary_loss_mlp": 0.01004367,
      "balance_loss_clip": 1.03345513,
      "balance_loss_mlp": 1.00287092,
      "epoch": 0.5131966572476402,
      "flos": 59794896418560.0,
      "grad_norm": 0.7681886659106236,
      "language_loss": 0.54740709,
      "learning_rate": 2.0116845082394446e-06,
      "loss": 0.56795424,
      "num_input_tokens_seen": 92153015,
      "step": 4268,
      "time_per_iteration": 3.1891331672668457
    },
    {
      "auxiliary_loss_clip": 0.01172765,
      "auxiliary_loss_mlp": 0.01024595,
      "balance_loss_clip": 1.05345535,
      "balance_loss_mlp": 1.01644742,
      "epoch": 0.5133169001382794,
      "flos": 18515219132160.0,
      "grad_norm": 1.853034544470862,
      "language_loss": 0.78480369,
      "learning_rate": 2.0109055490195836e-06,
      "loss": 0.80677724,
      "num_input_tokens_seen": 92171470,
      "step": 4269,
      "time_per_iteration": 2.454500913619995
    },
    {
      "auxiliary_loss_clip": 0.01113876,
      "auxiliary_loss_mlp": 0.01025866,
      "balance_loss_clip": 1.04113901,
      "balance_loss_mlp": 1.01811767,
      "epoch": 0.5134371430289184,
      "flos": 15523537219200.0,
      "grad_norm": 2.2869226140867918,
      "language_loss": 0.64670295,
      "learning_rate": 2.0101265881453605e-06,
      "loss": 0.66810036,
      "num_input_tokens_seen": 92189945,
      "step": 4270,
      "time_per_iteration": 2.58471941947937
    },
    {
      "auxiliary_loss_clip": 0.01149587,
      "auxiliary_loss_mlp": 0.01031538,
      "balance_loss_clip": 1.05353785,
      "balance_loss_mlp": 1.02380776,
      "epoch": 0.5135573859195575,
      "flos": 21433786911360.0,
      "grad_norm": 2.088858675280562,
      "language_loss": 0.78387833,
      "learning_rate": 2.009347625734941e-06,
      "loss": 0.80568963,
      "num_input_tokens_seen": 92209855,
      "step": 4271,
      "time_per_iteration": 2.5243873596191406
    },
    {
      "auxiliary_loss_clip": 0.01187983,
      "auxiliary_loss_mlp": 0.01027625,
      "balance_loss_clip": 1.0594821,
      "balance_loss_mlp": 1.01976943,
      "epoch": 0.5136776288101966,
      "flos": 17712651600000.0,
      "grad_norm": 2.1007574975095444,
      "language_loss": 0.74510866,
      "learning_rate": 2.0085686619064954e-06,
      "loss": 0.76726472,
      "num_input_tokens_seen": 92226295,
      "step": 4272,
      "time_per_iteration": 2.4277310371398926
    },
    {
      "auxiliary_loss_clip": 0.01173347,
      "auxiliary_loss_mlp": 0.01030694,
      "balance_loss_clip": 1.056095,
      "balance_loss_mlp": 1.02272475,
      "epoch": 0.5137978717008357,
      "flos": 16581680997120.0,
      "grad_norm": 3.1822277116267355,
      "language_loss": 0.82629323,
      "learning_rate": 2.00778969677819e-06,
      "loss": 0.84833366,
      "num_input_tokens_seen": 92243330,
      "step": 4273,
      "time_per_iteration": 2.473097324371338
    },
    {
      "auxiliary_loss_clip": 0.0115095,
      "auxiliary_loss_mlp": 0.01024046,
      "balance_loss_clip": 1.05058813,
      "balance_loss_mlp": 1.01648211,
      "epoch": 0.5139181145914747,
      "flos": 20668243322880.0,
      "grad_norm": 1.9084091067323563,
      "language_loss": 0.63895261,
      "learning_rate": 2.0070107304681934e-06,
      "loss": 0.66070259,
      "num_input_tokens_seen": 92262285,
      "step": 4274,
      "time_per_iteration": 2.5414175987243652
    },
    {
      "auxiliary_loss_clip": 0.01139716,
      "auxiliary_loss_mlp": 0.01027266,
      "balance_loss_clip": 1.05352378,
      "balance_loss_mlp": 1.01905286,
      "epoch": 0.5140383574821139,
      "flos": 32926996546560.0,
      "grad_norm": 1.8882609542090956,
      "language_loss": 0.78203666,
      "learning_rate": 2.006231763094675e-06,
      "loss": 0.80370641,
      "num_input_tokens_seen": 92283305,
      "step": 4275,
      "time_per_iteration": 2.6354713439941406
    },
    {
      "auxiliary_loss_clip": 0.01149281,
      "auxiliary_loss_mlp": 0.0102304,
      "balance_loss_clip": 1.05418932,
      "balance_loss_mlp": 1.01544595,
      "epoch": 0.514158600372753,
      "flos": 19537093152000.0,
      "grad_norm": 2.4021131350950373,
      "language_loss": 0.87360674,
      "learning_rate": 2.0054527947758027e-06,
      "loss": 0.89532995,
      "num_input_tokens_seen": 92302105,
      "step": 4276,
      "time_per_iteration": 2.600318670272827
    },
    {
      "auxiliary_loss_clip": 0.01071156,
      "auxiliary_loss_mlp": 0.01003584,
      "balance_loss_clip": 1.02186275,
      "balance_loss_mlp": 1.00257671,
      "epoch": 0.514278843263392,
      "flos": 62523855279360.0,
      "grad_norm": 0.7206748185050432,
      "language_loss": 0.55898213,
      "learning_rate": 2.004673825629746e-06,
      "loss": 0.57972956,
      "num_input_tokens_seen": 92362885,
      "step": 4277,
      "time_per_iteration": 3.039154291152954
    },
    {
      "auxiliary_loss_clip": 0.0114906,
      "auxiliary_loss_mlp": 0.01028679,
      "balance_loss_clip": 1.05074024,
      "balance_loss_mlp": 1.02107334,
      "epoch": 0.5143990861540312,
      "flos": 25882328545920.0,
      "grad_norm": 1.5859270425814573,
      "language_loss": 0.72312629,
      "learning_rate": 2.0038948557746744e-06,
      "loss": 0.74490374,
      "num_input_tokens_seen": 92384740,
      "step": 4278,
      "time_per_iteration": 2.527926206588745
    },
    {
      "auxiliary_loss_clip": 0.01162618,
      "auxiliary_loss_mlp": 0.01026902,
      "balance_loss_clip": 1.05229259,
      "balance_loss_mlp": 1.01935351,
      "epoch": 0.5145193290446702,
      "flos": 23330660238720.0,
      "grad_norm": 1.7345291500477011,
      "language_loss": 0.75129551,
      "learning_rate": 2.0031158853287558e-06,
      "loss": 0.77319074,
      "num_input_tokens_seen": 92405175,
      "step": 4279,
      "time_per_iteration": 2.477201461791992
    },
    {
      "auxiliary_loss_clip": 0.011537,
      "auxiliary_loss_mlp": 0.01030551,
      "balance_loss_clip": 1.05419528,
      "balance_loss_mlp": 1.02298689,
      "epoch": 0.5146395719353093,
      "flos": 22856603518080.0,
      "grad_norm": 2.1843884344457845,
      "language_loss": 0.70376295,
      "learning_rate": 2.0023369144101593e-06,
      "loss": 0.72560543,
      "num_input_tokens_seen": 92423345,
      "step": 4280,
      "time_per_iteration": 2.5090935230255127
    },
    {
      "auxiliary_loss_clip": 0.01142817,
      "auxiliary_loss_mlp": 0.01027925,
      "balance_loss_clip": 1.04797149,
      "balance_loss_mlp": 1.02022433,
      "epoch": 0.5147598148259485,
      "flos": 26391577616640.0,
      "grad_norm": 1.6592812695804138,
      "language_loss": 0.76773685,
      "learning_rate": 2.0015579431370555e-06,
      "loss": 0.78944433,
      "num_input_tokens_seen": 92445025,
      "step": 4281,
      "time_per_iteration": 2.5510001182556152
    },
    {
      "auxiliary_loss_clip": 0.01164272,
      "auxiliary_loss_mlp": 0.01025841,
      "balance_loss_clip": 1.05373812,
      "balance_loss_mlp": 1.01799083,
      "epoch": 0.5148800577165875,
      "flos": 29965694561280.0,
      "grad_norm": 2.785029311071165,
      "language_loss": 0.69829583,
      "learning_rate": 2.000778971627612e-06,
      "loss": 0.72019702,
      "num_input_tokens_seen": 92464490,
      "step": 4282,
      "time_per_iteration": 2.52500057220459
    },
    {
      "auxiliary_loss_clip": 0.0114497,
      "auxiliary_loss_mlp": 0.01032913,
      "balance_loss_clip": 1.0489862,
      "balance_loss_mlp": 1.02492595,
      "epoch": 0.5150003006072266,
      "flos": 17931383470080.0,
      "grad_norm": 1.9897955407156322,
      "language_loss": 0.90266895,
      "learning_rate": 2e-06,
      "loss": 0.92444777,
      "num_input_tokens_seen": 92482085,
      "step": 4283,
      "time_per_iteration": 3.3161795139312744
    },
    {
      "auxiliary_loss_clip": 0.01179945,
      "auxiliary_loss_mlp": 0.01028807,
      "balance_loss_clip": 1.05495954,
      "balance_loss_mlp": 1.02117157,
      "epoch": 0.5151205434978657,
      "flos": 18478733892480.0,
      "grad_norm": 1.699246641932537,
      "language_loss": 0.85659552,
      "learning_rate": 1.9992210283723878e-06,
      "loss": 0.87868309,
      "num_input_tokens_seen": 92499325,
      "step": 4284,
      "time_per_iteration": 2.410841226577759
    },
    {
      "auxiliary_loss_clip": 0.01180323,
      "auxiliary_loss_mlp": 0.01027353,
      "balance_loss_clip": 1.0563767,
      "balance_loss_mlp": 1.01995599,
      "epoch": 0.5152407863885048,
      "flos": 25341263003520.0,
      "grad_norm": 1.6415695985877476,
      "language_loss": 0.79565156,
      "learning_rate": 1.9984420568629448e-06,
      "loss": 0.81772828,
      "num_input_tokens_seen": 92522090,
      "step": 4285,
      "time_per_iteration": 2.527308940887451
    },
    {
      "auxiliary_loss_clip": 0.01167141,
      "auxiliary_loss_mlp": 0.0102579,
      "balance_loss_clip": 1.05316257,
      "balance_loss_mlp": 1.0186851,
      "epoch": 0.5153610292791438,
      "flos": 18329740277760.0,
      "grad_norm": 2.0560583649139854,
      "language_loss": 0.78259861,
      "learning_rate": 1.9976630855898405e-06,
      "loss": 0.80452788,
      "num_input_tokens_seen": 92539845,
      "step": 4286,
      "time_per_iteration": 2.4286062717437744
    },
    {
      "auxiliary_loss_clip": 0.01146903,
      "auxiliary_loss_mlp": 0.01024541,
      "balance_loss_clip": 1.04628944,
      "balance_loss_mlp": 1.01700091,
      "epoch": 0.515481272169783,
      "flos": 30409945971840.0,
      "grad_norm": 2.008342408011769,
      "language_loss": 0.74458647,
      "learning_rate": 1.9968841146712445e-06,
      "loss": 0.76630092,
      "num_input_tokens_seen": 92559460,
      "step": 4287,
      "time_per_iteration": 3.4090259075164795
    },
    {
      "auxiliary_loss_clip": 0.01109598,
      "auxiliary_loss_mlp": 0.00762022,
      "balance_loss_clip": 1.04694295,
      "balance_loss_mlp": 1.00020337,
      "epoch": 0.5156015150604221,
      "flos": 23037305863680.0,
      "grad_norm": 1.6638910783958147,
      "language_loss": 0.71525955,
      "learning_rate": 1.996105144225326e-06,
      "loss": 0.73397571,
      "num_input_tokens_seen": 92579695,
      "step": 4288,
      "time_per_iteration": 3.406270742416382
    },
    {
      "auxiliary_loss_clip": 0.01166683,
      "auxiliary_loss_mlp": 0.01029889,
      "balance_loss_clip": 1.05421185,
      "balance_loss_mlp": 1.02243233,
      "epoch": 0.5157217579510611,
      "flos": 17858556645120.0,
      "grad_norm": 1.7709147666094829,
      "language_loss": 0.78625512,
      "learning_rate": 1.995326174370254e-06,
      "loss": 0.80822086,
      "num_input_tokens_seen": 92598795,
      "step": 4289,
      "time_per_iteration": 2.4571619033813477
    },
    {
      "auxiliary_loss_clip": 0.01163046,
      "auxiliary_loss_mlp": 0.00761795,
      "balance_loss_clip": 1.05121469,
      "balance_loss_mlp": 1.00013304,
      "epoch": 0.5158420008417003,
      "flos": 19171486569600.0,
      "grad_norm": 1.557681297857005,
      "language_loss": 0.73106652,
      "learning_rate": 1.994547205224197e-06,
      "loss": 0.75031495,
      "num_input_tokens_seen": 92617700,
      "step": 4290,
      "time_per_iteration": 3.2195868492126465
    },
    {
      "auxiliary_loss_clip": 0.01145769,
      "auxiliary_loss_mlp": 0.0102658,
      "balance_loss_clip": 1.05079579,
      "balance_loss_mlp": 1.01839888,
      "epoch": 0.5159622437323393,
      "flos": 22419534827520.0,
      "grad_norm": 1.9993505968268424,
      "language_loss": 0.67429018,
      "learning_rate": 1.993768236905325e-06,
      "loss": 0.69601369,
      "num_input_tokens_seen": 92638370,
      "step": 4291,
      "time_per_iteration": 2.5067594051361084
    },
    {
      "auxiliary_loss_clip": 0.01147966,
      "auxiliary_loss_mlp": 0.01025063,
      "balance_loss_clip": 1.05018187,
      "balance_loss_mlp": 1.01724887,
      "epoch": 0.5160824866229784,
      "flos": 24603010773120.0,
      "grad_norm": 2.5267436636036757,
      "language_loss": 0.65778154,
      "learning_rate": 1.992989269531807e-06,
      "loss": 0.67951179,
      "num_input_tokens_seen": 92657180,
      "step": 4292,
      "time_per_iteration": 2.5515949726104736
    },
    {
      "auxiliary_loss_clip": 0.01152177,
      "auxiliary_loss_mlp": 0.01026493,
      "balance_loss_clip": 1.04967022,
      "balance_loss_mlp": 1.01865482,
      "epoch": 0.5162027295136175,
      "flos": 18002737837440.0,
      "grad_norm": 2.4856415194819648,
      "language_loss": 0.68858957,
      "learning_rate": 1.99221030322181e-06,
      "loss": 0.71037626,
      "num_input_tokens_seen": 92673985,
      "step": 4293,
      "time_per_iteration": 2.4755516052246094
    },
    {
      "auxiliary_loss_clip": 0.01155531,
      "auxiliary_loss_mlp": 0.01027698,
      "balance_loss_clip": 1.05152452,
      "balance_loss_mlp": 1.02004457,
      "epoch": 0.5163229724042566,
      "flos": 27344611221120.0,
      "grad_norm": 1.6513786060270488,
      "language_loss": 0.80928516,
      "learning_rate": 1.991431338093505e-06,
      "loss": 0.83111745,
      "num_input_tokens_seen": 92696340,
      "step": 4294,
      "time_per_iteration": 2.544196605682373
    },
    {
      "auxiliary_loss_clip": 0.01151707,
      "auxiliary_loss_mlp": 0.01026931,
      "balance_loss_clip": 1.0545224,
      "balance_loss_mlp": 1.0196445,
      "epoch": 0.5164432152948957,
      "flos": 21762764599680.0,
      "grad_norm": 2.8668974819848865,
      "language_loss": 0.79280865,
      "learning_rate": 1.9906523742650587e-06,
      "loss": 0.8145951,
      "num_input_tokens_seen": 92715200,
      "step": 4295,
      "time_per_iteration": 2.4912047386169434
    },
    {
      "auxiliary_loss_clip": 0.01180272,
      "auxiliary_loss_mlp": 0.01031478,
      "balance_loss_clip": 1.05210805,
      "balance_loss_mlp": 1.02317548,
      "epoch": 0.5165634581855347,
      "flos": 25550334115200.0,
      "grad_norm": 2.05186771821765,
      "language_loss": 0.77610034,
      "learning_rate": 1.9898734118546397e-06,
      "loss": 0.79821783,
      "num_input_tokens_seen": 92735150,
      "step": 4296,
      "time_per_iteration": 2.498521327972412
    },
    {
      "auxiliary_loss_clip": 0.01100435,
      "auxiliary_loss_mlp": 0.01025664,
      "balance_loss_clip": 1.04560828,
      "balance_loss_mlp": 1.01712275,
      "epoch": 0.5166837010761739,
      "flos": 19901191363200.0,
      "grad_norm": 1.5975486388966356,
      "language_loss": 0.80442047,
      "learning_rate": 1.989094450980416e-06,
      "loss": 0.82568151,
      "num_input_tokens_seen": 92755250,
      "step": 4297,
      "time_per_iteration": 2.6359899044036865
    },
    {
      "auxiliary_loss_clip": 0.01164353,
      "auxiliary_loss_mlp": 0.01023905,
      "balance_loss_clip": 1.05279386,
      "balance_loss_mlp": 1.01618052,
      "epoch": 0.516803943966813,
      "flos": 26646076454400.0,
      "grad_norm": 2.10310171739443,
      "language_loss": 0.76911175,
      "learning_rate": 1.9883154917605556e-06,
      "loss": 0.79099429,
      "num_input_tokens_seen": 92774460,
      "step": 4298,
      "time_per_iteration": 2.5876362323760986
    },
    {
      "auxiliary_loss_clip": 0.01178971,
      "auxiliary_loss_mlp": 0.01022804,
      "balance_loss_clip": 1.05338526,
      "balance_loss_mlp": 1.01564538,
      "epoch": 0.516924186857452,
      "flos": 19682854542720.0,
      "grad_norm": 1.9547209466611781,
      "language_loss": 0.83364004,
      "learning_rate": 1.9875365343132262e-06,
      "loss": 0.85565782,
      "num_input_tokens_seen": 92791580,
      "step": 4299,
      "time_per_iteration": 2.420444965362549
    },
    {
      "auxiliary_loss_clip": 0.01165741,
      "auxiliary_loss_mlp": 0.00762273,
      "balance_loss_clip": 1.05377007,
      "balance_loss_mlp": 1.00012457,
      "epoch": 0.5170444297480912,
      "flos": 15956583586560.0,
      "grad_norm": 3.028424899236101,
      "language_loss": 0.84598947,
      "learning_rate": 1.9867575787565946e-06,
      "loss": 0.86526966,
      "num_input_tokens_seen": 92806240,
      "step": 4300,
      "time_per_iteration": 2.4310429096221924
    },
    {
      "auxiliary_loss_clip": 0.01167281,
      "auxiliary_loss_mlp": 0.01025512,
      "balance_loss_clip": 1.05279255,
      "balance_loss_mlp": 1.01728082,
      "epoch": 0.5171646726387302,
      "flos": 14174157968640.0,
      "grad_norm": 2.4605083133120766,
      "language_loss": 0.86404359,
      "learning_rate": 1.9859786252088275e-06,
      "loss": 0.88597155,
      "num_input_tokens_seen": 92823420,
      "step": 4301,
      "time_per_iteration": 2.463771104812622
    },
    {
      "auxiliary_loss_clip": 0.01141744,
      "auxiliary_loss_mlp": 0.01031076,
      "balance_loss_clip": 1.05039752,
      "balance_loss_mlp": 1.02247512,
      "epoch": 0.5172849155293693,
      "flos": 23578550974080.0,
      "grad_norm": 2.897042630731926,
      "language_loss": 0.66240561,
      "learning_rate": 1.9851996737880914e-06,
      "loss": 0.68413377,
      "num_input_tokens_seen": 92838605,
      "step": 4302,
      "time_per_iteration": 2.521503448486328
    },
    {
      "auxiliary_loss_clip": 0.01171635,
      "auxiliary_loss_mlp": 0.01031522,
      "balance_loss_clip": 1.05365467,
      "balance_loss_mlp": 1.02348089,
      "epoch": 0.5174051584200084,
      "flos": 14283541860480.0,
      "grad_norm": 2.123928898927342,
      "language_loss": 0.74533945,
      "learning_rate": 1.9844207246125537e-06,
      "loss": 0.767371,
      "num_input_tokens_seen": 92855185,
      "step": 4303,
      "time_per_iteration": 2.484696865081787
    },
    {
      "auxiliary_loss_clip": 0.01148577,
      "auxiliary_loss_mlp": 0.01024082,
      "balance_loss_clip": 1.04949856,
      "balance_loss_mlp": 1.01698649,
      "epoch": 0.5175254013106475,
      "flos": 37889384192640.0,
      "grad_norm": 1.7874384254251288,
      "language_loss": 0.68360448,
      "learning_rate": 1.983641777800379e-06,
      "loss": 0.70533103,
      "num_input_tokens_seen": 92877830,
      "step": 4304,
      "time_per_iteration": 2.6318607330322266
    },
    {
      "auxiliary_loss_clip": 0.01062865,
      "auxiliary_loss_mlp": 0.01005569,
      "balance_loss_clip": 1.01982236,
      "balance_loss_mlp": 1.00444841,
      "epoch": 0.5176456442012866,
      "flos": 68549737829760.0,
      "grad_norm": 0.8793157424073644,
      "language_loss": 0.58793604,
      "learning_rate": 1.9828628334697343e-06,
      "loss": 0.60862041,
      "num_input_tokens_seen": 92945040,
      "step": 4305,
      "time_per_iteration": 3.240938901901245
    },
    {
      "auxiliary_loss_clip": 0.01065804,
      "auxiliary_loss_mlp": 0.01004215,
      "balance_loss_clip": 1.0218854,
      "balance_loss_mlp": 1.00299871,
      "epoch": 0.5177658870919257,
      "flos": 64084137235200.0,
      "grad_norm": 0.7698451296104786,
      "language_loss": 0.54749763,
      "learning_rate": 1.982083891738784e-06,
      "loss": 0.56819779,
      "num_input_tokens_seen": 93005910,
      "step": 4306,
      "time_per_iteration": 3.0985476970672607
    },
    {
      "auxiliary_loss_clip": 0.01147166,
      "auxiliary_loss_mlp": 0.01026133,
      "balance_loss_clip": 1.05288994,
      "balance_loss_mlp": 1.01856327,
      "epoch": 0.5178861299825648,
      "flos": 26651248012800.0,
      "grad_norm": 1.5003269605184575,
      "language_loss": 0.82572871,
      "learning_rate": 1.9813049527256923e-06,
      "loss": 0.8474617,
      "num_input_tokens_seen": 93026305,
      "step": 4307,
      "time_per_iteration": 2.54730224609375
    },
    {
      "auxiliary_loss_clip": 0.01135141,
      "auxiliary_loss_mlp": 0.01028255,
      "balance_loss_clip": 1.04671049,
      "balance_loss_mlp": 1.02073932,
      "epoch": 0.5180063728732038,
      "flos": 17931886260480.0,
      "grad_norm": 2.4991705447756334,
      "language_loss": 0.81758183,
      "learning_rate": 1.9805260165486252e-06,
      "loss": 0.83921587,
      "num_input_tokens_seen": 93045675,
      "step": 4308,
      "time_per_iteration": 2.5303757190704346
    },
    {
      "auxiliary_loss_clip": 0.01165803,
      "auxiliary_loss_mlp": 0.01022941,
      "balance_loss_clip": 1.05437827,
      "balance_loss_mlp": 1.01526403,
      "epoch": 0.518126615763843,
      "flos": 19500895221120.0,
      "grad_norm": 2.2166133050093895,
      "language_loss": 0.86462986,
      "learning_rate": 1.9797470833257457e-06,
      "loss": 0.88651729,
      "num_input_tokens_seen": 93065375,
      "step": 4309,
      "time_per_iteration": 3.323747158050537
    },
    {
      "auxiliary_loss_clip": 0.01167648,
      "auxiliary_loss_mlp": 0.01027318,
      "balance_loss_clip": 1.05438471,
      "balance_loss_mlp": 1.01899672,
      "epoch": 0.5182468586544821,
      "flos": 20704082117760.0,
      "grad_norm": 2.261365916362636,
      "language_loss": 0.77514726,
      "learning_rate": 1.9789681531752177e-06,
      "loss": 0.79709697,
      "num_input_tokens_seen": 93085595,
      "step": 4310,
      "time_per_iteration": 2.465726613998413
    },
    {
      "auxiliary_loss_clip": 0.01119335,
      "auxiliary_loss_mlp": 0.01025563,
      "balance_loss_clip": 1.04865932,
      "balance_loss_mlp": 1.01830888,
      "epoch": 0.5183671015451211,
      "flos": 23112107936640.0,
      "grad_norm": 1.5614632562759347,
      "language_loss": 0.72529924,
      "learning_rate": 1.978189226215204e-06,
      "loss": 0.74674821,
      "num_input_tokens_seen": 93106140,
      "step": 4311,
      "time_per_iteration": 2.572096586227417
    },
    {
      "auxiliary_loss_clip": 0.01179527,
      "auxiliary_loss_mlp": 0.01027023,
      "balance_loss_clip": 1.0537281,
      "balance_loss_mlp": 1.01897609,
      "epoch": 0.5184873444357603,
      "flos": 17597090568960.0,
      "grad_norm": 2.070529119022739,
      "language_loss": 0.76982635,
      "learning_rate": 1.9774103025638675e-06,
      "loss": 0.79189187,
      "num_input_tokens_seen": 93124265,
      "step": 4312,
      "time_per_iteration": 2.4041223526000977
    },
    {
      "auxiliary_loss_clip": 0.01127932,
      "auxiliary_loss_mlp": 0.01022869,
      "balance_loss_clip": 1.05382943,
      "balance_loss_mlp": 1.01555252,
      "epoch": 0.5186075873263993,
      "flos": 24936800883840.0,
      "grad_norm": 1.5604120131675898,
      "language_loss": 0.76241541,
      "learning_rate": 1.9766313823393696e-06,
      "loss": 0.78392339,
      "num_input_tokens_seen": 93145130,
      "step": 4313,
      "time_per_iteration": 2.588914632797241
    },
    {
      "auxiliary_loss_clip": 0.01117137,
      "auxiliary_loss_mlp": 0.01028235,
      "balance_loss_clip": 1.04318452,
      "balance_loss_mlp": 1.02031946,
      "epoch": 0.5187278302170384,
      "flos": 15190106244480.0,
      "grad_norm": 2.02884406468073,
      "language_loss": 0.68983698,
      "learning_rate": 1.975852465659873e-06,
      "loss": 0.71129072,
      "num_input_tokens_seen": 93161110,
      "step": 4314,
      "time_per_iteration": 3.3884284496307373
    },
    {
      "auxiliary_loss_clip": 0.01169039,
      "auxiliary_loss_mlp": 0.01030757,
      "balance_loss_clip": 1.05429661,
      "balance_loss_mlp": 1.02309155,
      "epoch": 0.5188480731076776,
      "flos": 25009412227200.0,
      "grad_norm": 2.0468833794700854,
      "language_loss": 0.69994009,
      "learning_rate": 1.9750735526435377e-06,
      "loss": 0.72193801,
      "num_input_tokens_seen": 93178055,
      "step": 4315,
      "time_per_iteration": 3.3257477283477783
    },
    {
      "auxiliary_loss_clip": 0.01153357,
      "auxiliary_loss_mlp": 0.01025859,
      "balance_loss_clip": 1.05405092,
      "balance_loss_mlp": 1.01817012,
      "epoch": 0.5189683159983166,
      "flos": 24790141653120.0,
      "grad_norm": 2.150479590116937,
      "language_loss": 0.78674585,
      "learning_rate": 1.974294643408525e-06,
      "loss": 0.80853796,
      "num_input_tokens_seen": 93195850,
      "step": 4316,
      "time_per_iteration": 3.349114418029785
    },
    {
      "auxiliary_loss_clip": 0.01167859,
      "auxiliary_loss_mlp": 0.01029102,
      "balance_loss_clip": 1.05047369,
      "balance_loss_mlp": 1.02165174,
      "epoch": 0.5190885588889557,
      "flos": 24754266944640.0,
      "grad_norm": 4.668884775065209,
      "language_loss": 0.67125702,
      "learning_rate": 1.9735157380729947e-06,
      "loss": 0.69322664,
      "num_input_tokens_seen": 93216260,
      "step": 4317,
      "time_per_iteration": 2.485344409942627
    },
    {
      "auxiliary_loss_clip": 0.01153076,
      "auxiliary_loss_mlp": 0.01022755,
      "balance_loss_clip": 1.04970074,
      "balance_loss_mlp": 1.01567125,
      "epoch": 0.5192088017795948,
      "flos": 24712646060160.0,
      "grad_norm": 1.8294509732764275,
      "language_loss": 0.84534085,
      "learning_rate": 1.9727368367551053e-06,
      "loss": 0.86709917,
      "num_input_tokens_seen": 93234810,
      "step": 4318,
      "time_per_iteration": 2.523404121398926
    },
    {
      "auxiliary_loss_clip": 0.01138333,
      "auxiliary_loss_mlp": 0.01025726,
      "balance_loss_clip": 1.04777312,
      "balance_loss_mlp": 1.01807272,
      "epoch": 0.5193290446702339,
      "flos": 27229588894080.0,
      "grad_norm": 1.9492075566846674,
      "language_loss": 0.68504083,
      "learning_rate": 1.9719579395730164e-06,
      "loss": 0.70668137,
      "num_input_tokens_seen": 93254185,
      "step": 4319,
      "time_per_iteration": 2.528857946395874
    },
    {
      "auxiliary_loss_clip": 0.01181693,
      "auxiliary_loss_mlp": 0.01023542,
      "balance_loss_clip": 1.05628467,
      "balance_loss_mlp": 1.01593351,
      "epoch": 0.5194492875608729,
      "flos": 11473352392320.0,
      "grad_norm": 2.370936817676572,
      "language_loss": 0.93126476,
      "learning_rate": 1.9711790466448854e-06,
      "loss": 0.95331705,
      "num_input_tokens_seen": 93268205,
      "step": 4320,
      "time_per_iteration": 2.37650990486145
    },
    {
      "auxiliary_loss_clip": 0.01128511,
      "auxiliary_loss_mlp": 0.01033846,
      "balance_loss_clip": 1.04931116,
      "balance_loss_mlp": 1.02584112,
      "epoch": 0.5195695304515121,
      "flos": 20338906498560.0,
      "grad_norm": 2.0874242588894107,
      "language_loss": 0.71438462,
      "learning_rate": 1.9704001580888704e-06,
      "loss": 0.73600817,
      "num_input_tokens_seen": 93286945,
      "step": 4321,
      "time_per_iteration": 2.5591723918914795
    },
    {
      "auxiliary_loss_clip": 0.0114406,
      "auxiliary_loss_mlp": 0.00762592,
      "balance_loss_clip": 1.04641712,
      "balance_loss_mlp": 1.00028682,
      "epoch": 0.5196897733421512,
      "flos": 20048317470720.0,
      "grad_norm": 3.0046490005712054,
      "language_loss": 0.8701604,
      "learning_rate": 1.9696212740231283e-06,
      "loss": 0.88922691,
      "num_input_tokens_seen": 93305595,
      "step": 4322,
      "time_per_iteration": 2.4885165691375732
    },
    {
      "auxiliary_loss_clip": 0.01172126,
      "auxiliary_loss_mlp": 0.01024591,
      "balance_loss_clip": 1.0511024,
      "balance_loss_mlp": 1.01629353,
      "epoch": 0.5198100162327902,
      "flos": 23805507058560.0,
      "grad_norm": 2.139084722899328,
      "language_loss": 0.82582247,
      "learning_rate": 1.9688423945658146e-06,
      "loss": 0.84778965,
      "num_input_tokens_seen": 93326460,
      "step": 4323,
      "time_per_iteration": 2.487976312637329
    },
    {
      "auxiliary_loss_clip": 0.01113229,
      "auxiliary_loss_mlp": 0.0102657,
      "balance_loss_clip": 1.0415988,
      "balance_loss_mlp": 1.01821399,
      "epoch": 0.5199302591234293,
      "flos": 24023951619840.0,
      "grad_norm": 1.985345452166628,
      "language_loss": 0.7154538,
      "learning_rate": 1.9680635198350845e-06,
      "loss": 0.73685181,
      "num_input_tokens_seen": 93346170,
      "step": 4324,
      "time_per_iteration": 2.576594829559326
    },
    {
      "auxiliary_loss_clip": 0.01165487,
      "auxiliary_loss_mlp": 0.01032973,
      "balance_loss_clip": 1.04954374,
      "balance_loss_mlp": 1.02458119,
      "epoch": 0.5200505020140684,
      "flos": 26359366095360.0,
      "grad_norm": 2.012401828540083,
      "language_loss": 0.72556275,
      "learning_rate": 1.967284649949093e-06,
      "loss": 0.74754739,
      "num_input_tokens_seen": 93365380,
      "step": 4325,
      "time_per_iteration": 2.512523651123047
    },
    {
      "auxiliary_loss_clip": 0.01135533,
      "auxiliary_loss_mlp": 0.01030607,
      "balance_loss_clip": 1.04761183,
      "balance_loss_mlp": 1.02251267,
      "epoch": 0.5201707449047075,
      "flos": 39604262284800.0,
      "grad_norm": 1.7634997804006274,
      "language_loss": 0.72408974,
      "learning_rate": 1.966505785025994e-06,
      "loss": 0.74575114,
      "num_input_tokens_seen": 93387285,
      "step": 4326,
      "time_per_iteration": 2.67545485496521
    },
    {
      "auxiliary_loss_clip": 0.0113476,
      "auxiliary_loss_mlp": 0.01027572,
      "balance_loss_clip": 1.04808664,
      "balance_loss_mlp": 1.01947165,
      "epoch": 0.5202909877953465,
      "flos": 53682788292480.0,
      "grad_norm": 1.791571356702319,
      "language_loss": 0.76148891,
      "learning_rate": 1.965726925183941e-06,
      "loss": 0.78311223,
      "num_input_tokens_seen": 93410390,
      "step": 4327,
      "time_per_iteration": 2.8161420822143555
    },
    {
      "auxiliary_loss_clip": 0.0118042,
      "auxiliary_loss_mlp": 0.01022503,
      "balance_loss_clip": 1.05524254,
      "balance_loss_mlp": 1.01525545,
      "epoch": 0.5204112306859857,
      "flos": 19537021324800.0,
      "grad_norm": 1.7586306918358665,
      "language_loss": 0.84700787,
      "learning_rate": 1.964948070541087e-06,
      "loss": 0.86903709,
      "num_input_tokens_seen": 93429050,
      "step": 4328,
      "time_per_iteration": 2.4287045001983643
    },
    {
      "auxiliary_loss_clip": 0.01153016,
      "auxiliary_loss_mlp": 0.01030736,
      "balance_loss_clip": 1.04685843,
      "balance_loss_mlp": 1.02280903,
      "epoch": 0.5205314735766248,
      "flos": 15304697608320.0,
      "grad_norm": 2.4499090544097184,
      "language_loss": 0.69810385,
      "learning_rate": 1.9641692212155816e-06,
      "loss": 0.71994138,
      "num_input_tokens_seen": 93446815,
      "step": 4329,
      "time_per_iteration": 2.451185464859009
    },
    {
      "auxiliary_loss_clip": 0.01123511,
      "auxiliary_loss_mlp": 0.01031179,
      "balance_loss_clip": 1.05195141,
      "balance_loss_mlp": 1.02306128,
      "epoch": 0.5206517164672638,
      "flos": 59263701160320.0,
      "grad_norm": 1.9996389057800341,
      "language_loss": 0.72494018,
      "learning_rate": 1.9633903773255777e-06,
      "loss": 0.74648714,
      "num_input_tokens_seen": 93469130,
      "step": 4330,
      "time_per_iteration": 2.8791377544403076
    },
    {
      "auxiliary_loss_clip": 0.01176565,
      "auxiliary_loss_mlp": 0.0102619,
      "balance_loss_clip": 1.05012131,
      "balance_loss_mlp": 1.01829851,
      "epoch": 0.520771959357903,
      "flos": 26871129118080.0,
      "grad_norm": 1.7086111102781905,
      "language_loss": 0.7495327,
      "learning_rate": 1.9626115389892237e-06,
      "loss": 0.77156025,
      "num_input_tokens_seen": 93489920,
      "step": 4331,
      "time_per_iteration": 2.498196601867676
    },
    {
      "auxiliary_loss_clip": 0.01143066,
      "auxiliary_loss_mlp": 0.01024783,
      "balance_loss_clip": 1.04956031,
      "balance_loss_mlp": 1.01695704,
      "epoch": 0.520892202248542,
      "flos": 26907075653760.0,
      "grad_norm": 2.7606361747193757,
      "language_loss": 0.85726553,
      "learning_rate": 1.96183270632467e-06,
      "loss": 0.87894404,
      "num_input_tokens_seen": 93509770,
      "step": 4332,
      "time_per_iteration": 2.6455507278442383
    },
    {
      "auxiliary_loss_clip": 0.0112943,
      "auxiliary_loss_mlp": 0.00762793,
      "balance_loss_clip": 1.04587126,
      "balance_loss_mlp": 1.0001626,
      "epoch": 0.5210124451391811,
      "flos": 25849434666240.0,
      "grad_norm": 1.7626853461122032,
      "language_loss": 0.79265553,
      "learning_rate": 1.9610538794500644e-06,
      "loss": 0.81157768,
      "num_input_tokens_seen": 93529320,
      "step": 4333,
      "time_per_iteration": 2.5401036739349365
    },
    {
      "auxiliary_loss_clip": 0.01051956,
      "auxiliary_loss_mlp": 0.01003295,
      "balance_loss_clip": 1.01970232,
      "balance_loss_mlp": 1.00224614,
      "epoch": 0.5211326880298203,
      "flos": 70553804319360.0,
      "grad_norm": 0.7727918858617991,
      "language_loss": 0.59494221,
      "learning_rate": 1.9602750584835542e-06,
      "loss": 0.61549473,
      "num_input_tokens_seen": 93595255,
      "step": 4334,
      "time_per_iteration": 3.2483842372894287
    },
    {
      "auxiliary_loss_clip": 0.01147472,
      "auxiliary_loss_mlp": 0.0102291,
      "balance_loss_clip": 1.04837263,
      "balance_loss_mlp": 1.01527476,
      "epoch": 0.5212529309204593,
      "flos": 15628898787840.0,
      "grad_norm": 1.9299286669331854,
      "language_loss": 0.82301176,
      "learning_rate": 1.959496243543286e-06,
      "loss": 0.84471554,
      "num_input_tokens_seen": 93613135,
      "step": 4335,
      "time_per_iteration": 3.2852776050567627
    },
    {
      "auxiliary_loss_clip": 0.01169145,
      "auxiliary_loss_mlp": 0.01036039,
      "balance_loss_clip": 1.05653191,
      "balance_loss_mlp": 1.02782249,
      "epoch": 0.5213731738110984,
      "flos": 26242655829120.0,
      "grad_norm": 2.013150200395761,
      "language_loss": 0.79077005,
      "learning_rate": 1.9587174347474057e-06,
      "loss": 0.81282187,
      "num_input_tokens_seen": 93629645,
      "step": 4336,
      "time_per_iteration": 2.4735939502716064
    },
    {
      "auxiliary_loss_clip": 0.011091,
      "auxiliary_loss_mlp": 0.01029109,
      "balance_loss_clip": 1.04402423,
      "balance_loss_mlp": 1.02120566,
      "epoch": 0.5214934167017375,
      "flos": 19418407637760.0,
      "grad_norm": 2.116526222975386,
      "language_loss": 0.82434964,
      "learning_rate": 1.9579386322140574e-06,
      "loss": 0.84573174,
      "num_input_tokens_seen": 93645325,
      "step": 4337,
      "time_per_iteration": 2.5194547176361084
    },
    {
      "auxiliary_loss_clip": 0.0118239,
      "auxiliary_loss_mlp": 0.00762601,
      "balance_loss_clip": 1.05463576,
      "balance_loss_mlp": 1.00025547,
      "epoch": 0.5216136595923766,
      "flos": 30955788023040.0,
      "grad_norm": 1.7550947998859048,
      "language_loss": 0.80700946,
      "learning_rate": 1.9571598360613854e-06,
      "loss": 0.82645935,
      "num_input_tokens_seen": 93668200,
      "step": 4338,
      "time_per_iteration": 2.488835334777832
    },
    {
      "auxiliary_loss_clip": 0.01134789,
      "auxiliary_loss_mlp": 0.01026559,
      "balance_loss_clip": 1.04458201,
      "balance_loss_mlp": 1.01879263,
      "epoch": 0.5217339024830157,
      "flos": 21945047143680.0,
      "grad_norm": 2.0776521547677693,
      "language_loss": 0.6973902,
      "learning_rate": 1.956381046407532e-06,
      "loss": 0.71900368,
      "num_input_tokens_seen": 93688495,
      "step": 4339,
      "time_per_iteration": 2.475827932357788
    },
    {
      "auxiliary_loss_clip": 0.01132451,
      "auxiliary_loss_mlp": 0.01031366,
      "balance_loss_clip": 1.04680729,
      "balance_loss_mlp": 1.02343869,
      "epoch": 0.5218541453736548,
      "flos": 20923209037440.0,
      "grad_norm": 1.7810824338663827,
      "language_loss": 0.86097682,
      "learning_rate": 1.9556022633706394e-06,
      "loss": 0.88261497,
      "num_input_tokens_seen": 93707285,
      "step": 4340,
      "time_per_iteration": 3.3773446083068848
    },
    {
      "auxiliary_loss_clip": 0.01143484,
      "auxiliary_loss_mlp": 0.01029277,
      "balance_loss_clip": 1.04903436,
      "balance_loss_mlp": 1.02143347,
      "epoch": 0.5219743882642939,
      "flos": 23951663498880.0,
      "grad_norm": 1.6106960482397852,
      "language_loss": 0.79703474,
      "learning_rate": 1.954823487068848e-06,
      "loss": 0.8187623,
      "num_input_tokens_seen": 93727495,
      "step": 4341,
      "time_per_iteration": 3.3566136360168457
    },
    {
      "auxiliary_loss_clip": 0.01165518,
      "auxiliary_loss_mlp": 0.01028864,
      "balance_loss_clip": 1.05290639,
      "balance_loss_mlp": 1.02135944,
      "epoch": 0.5220946311549329,
      "flos": 28799280213120.0,
      "grad_norm": 1.672869638190347,
      "language_loss": 0.80807132,
      "learning_rate": 1.9540447176202976e-06,
      "loss": 0.83001512,
      "num_input_tokens_seen": 93748740,
      "step": 4342,
      "time_per_iteration": 2.521245241165161
    },
    {
      "auxiliary_loss_clip": 0.01067541,
      "auxiliary_loss_mlp": 0.01001622,
      "balance_loss_clip": 1.01848412,
      "balance_loss_mlp": 1.00064456,
      "epoch": 0.5222148740455721,
      "flos": 67189369017600.0,
      "grad_norm": 0.8879401036442628,
      "language_loss": 0.60693657,
      "learning_rate": 1.9532659551431272e-06,
      "loss": 0.62762821,
      "num_input_tokens_seen": 93815770,
      "step": 4343,
      "time_per_iteration": 3.2200825214385986
    },
    {
      "auxiliary_loss_clip": 0.01166344,
      "auxiliary_loss_mlp": 0.01026637,
      "balance_loss_clip": 1.05193317,
      "balance_loss_mlp": 1.01924598,
      "epoch": 0.5223351169362112,
      "flos": 61856164339200.0,
      "grad_norm": 1.5510103453136361,
      "language_loss": 0.67752343,
      "learning_rate": 1.9524871997554744e-06,
      "loss": 0.69945323,
      "num_input_tokens_seen": 93843530,
      "step": 4344,
      "time_per_iteration": 3.5065371990203857
    },
    {
      "auxiliary_loss_clip": 0.01165432,
      "auxiliary_loss_mlp": 0.01026911,
      "balance_loss_clip": 1.05167866,
      "balance_loss_mlp": 1.0191927,
      "epoch": 0.5224553598268502,
      "flos": 14647388676480.0,
      "grad_norm": 5.207205442187628,
      "language_loss": 0.81162918,
      "learning_rate": 1.951708451575475e-06,
      "loss": 0.8335526,
      "num_input_tokens_seen": 93860595,
      "step": 4345,
      "time_per_iteration": 2.4215919971466064
    },
    {
      "auxiliary_loss_clip": 0.01143969,
      "auxiliary_loss_mlp": 0.01030587,
      "balance_loss_clip": 1.04691207,
      "balance_loss_mlp": 1.02298737,
      "epoch": 0.5225756027174894,
      "flos": 14826043946880.0,
      "grad_norm": 2.2067932277279616,
      "language_loss": 0.82118744,
      "learning_rate": 1.9509297107212657e-06,
      "loss": 0.84293306,
      "num_input_tokens_seen": 93877365,
      "step": 4346,
      "time_per_iteration": 2.481962203979492
    },
    {
      "auxiliary_loss_clip": 0.01175944,
      "auxiliary_loss_mlp": 0.01025648,
      "balance_loss_clip": 1.05192351,
      "balance_loss_mlp": 1.01818001,
      "epoch": 0.5226958456081284,
      "flos": 23512009029120.0,
      "grad_norm": 1.7004833271668003,
      "language_loss": 0.79269499,
      "learning_rate": 1.95015097731098e-06,
      "loss": 0.81471092,
      "num_input_tokens_seen": 93896855,
      "step": 4347,
      "time_per_iteration": 2.435988664627075
    },
    {
      "auxiliary_loss_clip": 0.01177566,
      "auxiliary_loss_mlp": 0.01025499,
      "balance_loss_clip": 1.05310202,
      "balance_loss_mlp": 1.0179503,
      "epoch": 0.5228160884987675,
      "flos": 19062928690560.0,
      "grad_norm": 4.298942695852673,
      "language_loss": 0.81880325,
      "learning_rate": 1.949372251462751e-06,
      "loss": 0.8408339,
      "num_input_tokens_seen": 93914270,
      "step": 4348,
      "time_per_iteration": 2.410475492477417
    },
    {
      "auxiliary_loss_clip": 0.01136318,
      "auxiliary_loss_mlp": 0.00761949,
      "balance_loss_clip": 1.04916906,
      "balance_loss_mlp": 1.00028086,
      "epoch": 0.5229363313894067,
      "flos": 21063224252160.0,
      "grad_norm": 1.9589558864326888,
      "language_loss": 0.82839519,
      "learning_rate": 1.9485935332947124e-06,
      "loss": 0.84737784,
      "num_input_tokens_seen": 93932180,
      "step": 4349,
      "time_per_iteration": 2.55025577545166
    },
    {
      "auxiliary_loss_clip": 0.01143372,
      "auxiliary_loss_mlp": 0.01023202,
      "balance_loss_clip": 1.04954123,
      "balance_loss_mlp": 1.01607358,
      "epoch": 0.5230565742800457,
      "flos": 14830389492480.0,
      "grad_norm": 2.524751867110256,
      "language_loss": 0.84168446,
      "learning_rate": 1.947814822924993e-06,
      "loss": 0.86335015,
      "num_input_tokens_seen": 93949690,
      "step": 4350,
      "time_per_iteration": 2.4523684978485107
    },
    {
      "auxiliary_loss_clip": 0.01177895,
      "auxiliary_loss_mlp": 0.01030901,
      "balance_loss_clip": 1.05479002,
      "balance_loss_mlp": 1.02355731,
      "epoch": 0.5231768171706848,
      "flos": 25813021253760.0,
      "grad_norm": 1.8300090896213734,
      "language_loss": 0.82848084,
      "learning_rate": 1.9470361204717236e-06,
      "loss": 0.85056877,
      "num_input_tokens_seen": 93968830,
      "step": 4351,
      "time_per_iteration": 2.571258068084717
    },
    {
      "auxiliary_loss_clip": 0.01139389,
      "auxiliary_loss_mlp": 0.00762291,
      "balance_loss_clip": 1.0495981,
      "balance_loss_mlp": 1.00025225,
      "epoch": 0.5232970600613239,
      "flos": 22743807834240.0,
      "grad_norm": 1.693903220217893,
      "language_loss": 0.80695319,
      "learning_rate": 1.9462574260530326e-06,
      "loss": 0.82596999,
      "num_input_tokens_seen": 93989110,
      "step": 4352,
      "time_per_iteration": 2.5304713249206543
    },
    {
      "auxiliary_loss_clip": 0.01154177,
      "auxiliary_loss_mlp": 0.0102684,
      "balance_loss_clip": 1.04898679,
      "balance_loss_mlp": 1.01900768,
      "epoch": 0.523417302951963,
      "flos": 17310703432320.0,
      "grad_norm": 1.7964250712837848,
      "language_loss": 0.80760342,
      "learning_rate": 1.9454787397870472e-06,
      "loss": 0.82941359,
      "num_input_tokens_seen": 94006430,
      "step": 4353,
      "time_per_iteration": 2.4470584392547607
    },
    {
      "auxiliary_loss_clip": 0.01103339,
      "auxiliary_loss_mlp": 0.01028481,
      "balance_loss_clip": 1.04832923,
      "balance_loss_mlp": 1.02044952,
      "epoch": 0.523537545842602,
      "flos": 18551740285440.0,
      "grad_norm": 1.770964479083237,
      "language_loss": 0.71991873,
      "learning_rate": 1.944700061791894e-06,
      "loss": 0.74123693,
      "num_input_tokens_seen": 94024825,
      "step": 4354,
      "time_per_iteration": 2.6146578788757324
    },
    {
      "auxiliary_loss_clip": 0.01162298,
      "auxiliary_loss_mlp": 0.01030004,
      "balance_loss_clip": 1.05246806,
      "balance_loss_mlp": 1.02224326,
      "epoch": 0.5236577887332411,
      "flos": 19719267955200.0,
      "grad_norm": 2.612381994926738,
      "language_loss": 0.65407026,
      "learning_rate": 1.943921392185698e-06,
      "loss": 0.67599326,
      "num_input_tokens_seen": 94043450,
      "step": 4355,
      "time_per_iteration": 2.5120255947113037
    },
    {
      "auxiliary_loss_clip": 0.01152573,
      "auxiliary_loss_mlp": 0.01026797,
      "balance_loss_clip": 1.0486176,
      "balance_loss_mlp": 1.01908731,
      "epoch": 0.5237780316238803,
      "flos": 23550218121600.0,
      "grad_norm": 2.1571417274222675,
      "language_loss": 0.77540857,
      "learning_rate": 1.9431427310865814e-06,
      "loss": 0.79720229,
      "num_input_tokens_seen": 94063055,
      "step": 4356,
      "time_per_iteration": 2.5682945251464844
    },
    {
      "auxiliary_loss_clip": 0.01123109,
      "auxiliary_loss_mlp": 0.0102904,
      "balance_loss_clip": 1.04862618,
      "balance_loss_mlp": 1.02081442,
      "epoch": 0.5238982745145193,
      "flos": 22491894775680.0,
      "grad_norm": 1.6814132975443963,
      "language_loss": 0.78262562,
      "learning_rate": 1.942364078612667e-06,
      "loss": 0.80414712,
      "num_input_tokens_seen": 94081785,
      "step": 4357,
      "time_per_iteration": 2.5321431159973145
    },
    {
      "auxiliary_loss_clip": 0.0114305,
      "auxiliary_loss_mlp": 0.01024575,
      "balance_loss_clip": 1.04884946,
      "balance_loss_mlp": 1.01734471,
      "epoch": 0.5240185174051584,
      "flos": 27088927234560.0,
      "grad_norm": 2.023984857391796,
      "language_loss": 0.75546193,
      "learning_rate": 1.9415854348820765e-06,
      "loss": 0.77713811,
      "num_input_tokens_seen": 94101635,
      "step": 4358,
      "time_per_iteration": 2.5493717193603516
    },
    {
      "auxiliary_loss_clip": 0.01168851,
      "auxiliary_loss_mlp": 0.01027763,
      "balance_loss_clip": 1.05036139,
      "balance_loss_mlp": 1.01940095,
      "epoch": 0.5241387602957975,
      "flos": 22674680110080.0,
      "grad_norm": 2.0470814232962895,
      "language_loss": 0.67465502,
      "learning_rate": 1.940806800012929e-06,
      "loss": 0.69662118,
      "num_input_tokens_seen": 94121705,
      "step": 4359,
      "time_per_iteration": 2.460353374481201
    },
    {
      "auxiliary_loss_clip": 0.01117597,
      "auxiliary_loss_mlp": 0.00762406,
      "balance_loss_clip": 1.04751372,
      "balance_loss_mlp": 1.00024033,
      "epoch": 0.5242590031864366,
      "flos": 40553453134080.0,
      "grad_norm": 1.6004649444778154,
      "language_loss": 0.63456345,
      "learning_rate": 1.9400281741233432e-06,
      "loss": 0.65336347,
      "num_input_tokens_seen": 94146595,
      "step": 4360,
      "time_per_iteration": 2.725044012069702
    },
    {
      "auxiliary_loss_clip": 0.01039033,
      "auxiliary_loss_mlp": 0.01008526,
      "balance_loss_clip": 1.01704049,
      "balance_loss_mlp": 1.00757241,
      "epoch": 0.5243792460770756,
      "flos": 66676313105280.0,
      "grad_norm": 0.6555360742272462,
      "language_loss": 0.52552974,
      "learning_rate": 1.939249557331435e-06,
      "loss": 0.54600537,
      "num_input_tokens_seen": 94212410,
      "step": 4361,
      "time_per_iteration": 3.1416094303131104
    },
    {
      "auxiliary_loss_clip": 0.01144121,
      "auxiliary_loss_mlp": 0.01029342,
      "balance_loss_clip": 1.05019045,
      "balance_loss_mlp": 1.02224588,
      "epoch": 0.5244994889677148,
      "flos": 28183663992960.0,
      "grad_norm": 1.913655570960772,
      "language_loss": 0.72607535,
      "learning_rate": 1.938470949755321e-06,
      "loss": 0.74781001,
      "num_input_tokens_seen": 94232290,
      "step": 4362,
      "time_per_iteration": 3.411875009536743
    },
    {
      "auxiliary_loss_clip": 0.01047553,
      "auxiliary_loss_mlp": 0.01006545,
      "balance_loss_clip": 1.01656938,
      "balance_loss_mlp": 1.00555539,
      "epoch": 0.5246197318583539,
      "flos": 65950379239680.0,
      "grad_norm": 0.81850348095091,
      "language_loss": 0.55699503,
      "learning_rate": 1.937692351513115e-06,
      "loss": 0.57753599,
      "num_input_tokens_seen": 94291285,
      "step": 4363,
      "time_per_iteration": 3.0562753677368164
    },
    {
      "auxiliary_loss_clip": 0.01168707,
      "auxiliary_loss_mlp": 0.01026146,
      "balance_loss_clip": 1.05165184,
      "balance_loss_mlp": 1.0186832,
      "epoch": 0.5247399747489929,
      "flos": 21033490769280.0,
      "grad_norm": 1.541235835049601,
      "language_loss": 0.80282021,
      "learning_rate": 1.9369137627229297e-06,
      "loss": 0.82476878,
      "num_input_tokens_seen": 94309685,
      "step": 4364,
      "time_per_iteration": 2.455899477005005
    },
    {
      "auxiliary_loss_clip": 0.01163184,
      "auxiliary_loss_mlp": 0.0102686,
      "balance_loss_clip": 1.05228126,
      "balance_loss_mlp": 1.0191952,
      "epoch": 0.5248602176396321,
      "flos": 19025940660480.0,
      "grad_norm": 2.280632731269297,
      "language_loss": 0.88028109,
      "learning_rate": 1.936135183502877e-06,
      "loss": 0.90218151,
      "num_input_tokens_seen": 94326985,
      "step": 4365,
      "time_per_iteration": 2.469602108001709
    },
    {
      "auxiliary_loss_clip": 0.01139628,
      "auxiliary_loss_mlp": 0.01024871,
      "balance_loss_clip": 1.04844642,
      "balance_loss_mlp": 1.01735473,
      "epoch": 0.5249804605302711,
      "flos": 22200084685440.0,
      "grad_norm": 2.1876929504225076,
      "language_loss": 0.79830396,
      "learning_rate": 1.935356613971066e-06,
      "loss": 0.81994891,
      "num_input_tokens_seen": 94347645,
      "step": 4366,
      "time_per_iteration": 3.388249397277832
    },
    {
      "auxiliary_loss_clip": 0.01146819,
      "auxiliary_loss_mlp": 0.00761961,
      "balance_loss_clip": 1.04911792,
      "balance_loss_mlp": 1.0002172,
      "epoch": 0.5251007034209102,
      "flos": 23805686626560.0,
      "grad_norm": 1.7929640130798927,
      "language_loss": 0.76504159,
      "learning_rate": 1.9345780542456047e-06,
      "loss": 0.78412938,
      "num_input_tokens_seen": 94367020,
      "step": 4367,
      "time_per_iteration": 2.51507568359375
    },
    {
      "auxiliary_loss_clip": 0.0115283,
      "auxiliary_loss_mlp": 0.01028503,
      "balance_loss_clip": 1.04873621,
      "balance_loss_mlp": 1.02101707,
      "epoch": 0.5252209463115494,
      "flos": 23294605962240.0,
      "grad_norm": 2.4416101440400015,
      "language_loss": 0.71907568,
      "learning_rate": 1.9337995044446007e-06,
      "loss": 0.74088907,
      "num_input_tokens_seen": 94385860,
      "step": 4368,
      "time_per_iteration": 3.278944492340088
    },
    {
      "auxiliary_loss_clip": 0.01168199,
      "auxiliary_loss_mlp": 0.01028193,
      "balance_loss_clip": 1.05153382,
      "balance_loss_mlp": 1.0203433,
      "epoch": 0.5253411892021884,
      "flos": 19828687760640.0,
      "grad_norm": 2.044737898576299,
      "language_loss": 0.79873681,
      "learning_rate": 1.9330209646861596e-06,
      "loss": 0.82070065,
      "num_input_tokens_seen": 94405010,
      "step": 4369,
      "time_per_iteration": 2.4421024322509766
    },
    {
      "auxiliary_loss_clip": 0.0114528,
      "auxiliary_loss_mlp": 0.01029998,
      "balance_loss_clip": 1.04863429,
      "balance_loss_mlp": 1.02292848,
      "epoch": 0.5254614320928275,
      "flos": 24133730561280.0,
      "grad_norm": 1.9552026212971447,
      "language_loss": 0.77821803,
      "learning_rate": 1.9322424350883843e-06,
      "loss": 0.79997075,
      "num_input_tokens_seen": 94426845,
      "step": 4370,
      "time_per_iteration": 2.5154197216033936
    },
    {
      "auxiliary_loss_clip": 0.01150655,
      "auxiliary_loss_mlp": 0.01028791,
      "balance_loss_clip": 1.04898167,
      "balance_loss_mlp": 1.02157307,
      "epoch": 0.5255816749834666,
      "flos": 24644954880000.0,
      "grad_norm": 2.5500733863189264,
      "language_loss": 0.78570747,
      "learning_rate": 1.931463915769379e-06,
      "loss": 0.80750191,
      "num_input_tokens_seen": 94446960,
      "step": 4371,
      "time_per_iteration": 3.2991833686828613
    },
    {
      "auxiliary_loss_clip": 0.01118164,
      "auxiliary_loss_mlp": 0.01024478,
      "balance_loss_clip": 1.04520321,
      "balance_loss_mlp": 1.01684523,
      "epoch": 0.5257019178741057,
      "flos": 14136595320960.0,
      "grad_norm": 2.242927013176805,
      "language_loss": 0.74336219,
      "learning_rate": 1.930685406847242e-06,
      "loss": 0.76478863,
      "num_input_tokens_seen": 94461535,
      "step": 4372,
      "time_per_iteration": 2.5128729343414307
    },
    {
      "auxiliary_loss_clip": 0.01144863,
      "auxiliary_loss_mlp": 0.01027176,
      "balance_loss_clip": 1.04809499,
      "balance_loss_mlp": 1.01987433,
      "epoch": 0.5258221607647448,
      "flos": 23548961145600.0,
      "grad_norm": 1.4621954235803154,
      "language_loss": 0.81530154,
      "learning_rate": 1.9299069084400734e-06,
      "loss": 0.83702195,
      "num_input_tokens_seen": 94482395,
      "step": 4373,
      "time_per_iteration": 2.51153564453125
    },
    {
      "auxiliary_loss_clip": 0.01133198,
      "auxiliary_loss_mlp": 0.01026445,
      "balance_loss_clip": 1.05137515,
      "balance_loss_mlp": 1.01860726,
      "epoch": 0.5259424036553839,
      "flos": 24966103403520.0,
      "grad_norm": 1.9663180526975068,
      "language_loss": 0.69542682,
      "learning_rate": 1.9291284206659717e-06,
      "loss": 0.71702325,
      "num_input_tokens_seen": 94500580,
      "step": 4374,
      "time_per_iteration": 2.5352210998535156
    },
    {
      "auxiliary_loss_clip": 0.01179488,
      "auxiliary_loss_mlp": 0.01021644,
      "balance_loss_clip": 1.05401397,
      "balance_loss_mlp": 1.01370513,
      "epoch": 0.526062646546023,
      "flos": 28763908295040.0,
      "grad_norm": 2.2382752326694977,
      "language_loss": 0.7157954,
      "learning_rate": 1.928349943643032e-06,
      "loss": 0.73780674,
      "num_input_tokens_seen": 94519680,
      "step": 4375,
      "time_per_iteration": 2.4739067554473877
    },
    {
      "auxiliary_loss_clip": 0.01157843,
      "auxiliary_loss_mlp": 0.01028292,
      "balance_loss_clip": 1.05037689,
      "balance_loss_mlp": 1.02052212,
      "epoch": 0.526182889436662,
      "flos": 22821375254400.0,
      "grad_norm": 1.8576834313800896,
      "language_loss": 0.81786835,
      "learning_rate": 1.9275714774893493e-06,
      "loss": 0.83972967,
      "num_input_tokens_seen": 94539135,
      "step": 4376,
      "time_per_iteration": 2.4604790210723877
    },
    {
      "auxiliary_loss_clip": 0.01123334,
      "auxiliary_loss_mlp": 0.01028427,
      "balance_loss_clip": 1.04401994,
      "balance_loss_mlp": 1.02013004,
      "epoch": 0.5263031323273012,
      "flos": 22929466256640.0,
      "grad_norm": 2.182211073509432,
      "language_loss": 0.73040879,
      "learning_rate": 1.9267930223230154e-06,
      "loss": 0.75192642,
      "num_input_tokens_seen": 94557610,
      "step": 4377,
      "time_per_iteration": 2.5353589057922363
    },
    {
      "auxiliary_loss_clip": 0.01152057,
      "auxiliary_loss_mlp": 0.01027588,
      "balance_loss_clip": 1.05085623,
      "balance_loss_mlp": 1.0202837,
      "epoch": 0.5264233752179402,
      "flos": 17748634049280.0,
      "grad_norm": 2.4095937786058936,
      "language_loss": 0.779791,
      "learning_rate": 1.9260145782621224e-06,
      "loss": 0.80158746,
      "num_input_tokens_seen": 94575390,
      "step": 4378,
      "time_per_iteration": 2.4543614387512207
    },
    {
      "auxiliary_loss_clip": 0.01147088,
      "auxiliary_loss_mlp": 0.01026594,
      "balance_loss_clip": 1.05152142,
      "balance_loss_mlp": 1.01914334,
      "epoch": 0.5265436181085793,
      "flos": 24421626069120.0,
      "grad_norm": 1.947542888374729,
      "language_loss": 0.87916994,
      "learning_rate": 1.925236145424758e-06,
      "loss": 0.9009068,
      "num_input_tokens_seen": 94594210,
      "step": 4379,
      "time_per_iteration": 2.508543014526367
    },
    {
      "auxiliary_loss_clip": 0.0106866,
      "auxiliary_loss_mlp": 0.01002682,
      "balance_loss_clip": 1.01650429,
      "balance_loss_mlp": 1.00184727,
      "epoch": 0.5266638609992185,
      "flos": 69207298156800.0,
      "grad_norm": 0.697470827288729,
      "language_loss": 0.57586503,
      "learning_rate": 1.924457723929012e-06,
      "loss": 0.59657848,
      "num_input_tokens_seen": 94665020,
      "step": 4380,
      "time_per_iteration": 3.174663782119751
    },
    {
      "auxiliary_loss_clip": 0.01162604,
      "auxiliary_loss_mlp": 0.01024759,
      "balance_loss_clip": 1.05070293,
      "balance_loss_mlp": 1.01727295,
      "epoch": 0.5267841038898575,
      "flos": 20738699850240.0,
      "grad_norm": 2.304291087725399,
      "language_loss": 0.83023971,
      "learning_rate": 1.9236793138929685e-06,
      "loss": 0.85211337,
      "num_input_tokens_seen": 94684290,
      "step": 4381,
      "time_per_iteration": 2.45731258392334
    },
    {
      "auxiliary_loss_clip": 0.01167551,
      "auxiliary_loss_mlp": 0.01025076,
      "balance_loss_clip": 1.05054665,
      "balance_loss_mlp": 1.01741672,
      "epoch": 0.5269043467804966,
      "flos": 17234392988160.0,
      "grad_norm": 1.9429832661019535,
      "language_loss": 0.81147134,
      "learning_rate": 1.9229009154347133e-06,
      "loss": 0.83339763,
      "num_input_tokens_seen": 94701880,
      "step": 4382,
      "time_per_iteration": 2.4214773178100586
    },
    {
      "auxiliary_loss_clip": 0.01106013,
      "auxiliary_loss_mlp": 0.00761869,
      "balance_loss_clip": 1.04324234,
      "balance_loss_mlp": 1.00023746,
      "epoch": 0.5270245896711357,
      "flos": 18223157646720.0,
      "grad_norm": 2.029056836800655,
      "language_loss": 0.80564171,
      "learning_rate": 1.922122528672327e-06,
      "loss": 0.82432055,
      "num_input_tokens_seen": 94720545,
      "step": 4383,
      "time_per_iteration": 2.529179096221924
    },
    {
      "auxiliary_loss_clip": 0.0117295,
      "auxiliary_loss_mlp": 0.01023731,
      "balance_loss_clip": 1.05089211,
      "balance_loss_mlp": 1.01635194,
      "epoch": 0.5271448325617748,
      "flos": 21287558643840.0,
      "grad_norm": 2.386289761773235,
      "language_loss": 0.780711,
      "learning_rate": 1.9213441537238914e-06,
      "loss": 0.80267781,
      "num_input_tokens_seen": 94737420,
      "step": 4384,
      "time_per_iteration": 2.4164559841156006
    },
    {
      "auxiliary_loss_clip": 0.01030032,
      "auxiliary_loss_mlp": 0.01001423,
      "balance_loss_clip": 1.02040553,
      "balance_loss_mlp": 1.00047517,
      "epoch": 0.5272650754524139,
      "flos": 65495497403520.0,
      "grad_norm": 0.841077346623064,
      "language_loss": 0.57383412,
      "learning_rate": 1.920565790707485e-06,
      "loss": 0.5941487,
      "num_input_tokens_seen": 94802810,
      "step": 4385,
      "time_per_iteration": 3.2633578777313232
    },
    {
      "auxiliary_loss_clip": 0.01129808,
      "auxiliary_loss_mlp": 0.01027898,
      "balance_loss_clip": 1.04693341,
      "balance_loss_mlp": 1.0197506,
      "epoch": 0.527385318343053,
      "flos": 19676426008320.0,
      "grad_norm": 1.8964439843109433,
      "language_loss": 0.65567815,
      "learning_rate": 1.9197874397411853e-06,
      "loss": 0.67725527,
      "num_input_tokens_seen": 94819440,
      "step": 4386,
      "time_per_iteration": 2.5258615016937256
    },
    {
      "auxiliary_loss_clip": 0.01134758,
      "auxiliary_loss_mlp": 0.01033394,
      "balance_loss_clip": 1.04434276,
      "balance_loss_mlp": 1.02490032,
      "epoch": 0.5275055612336921,
      "flos": 12712018947840.0,
      "grad_norm": 3.2809255173526526,
      "language_loss": 0.66693175,
      "learning_rate": 1.919009100943067e-06,
      "loss": 0.6886133,
      "num_input_tokens_seen": 94835130,
      "step": 4387,
      "time_per_iteration": 2.471652030944824
    },
    {
      "auxiliary_loss_clip": 0.01132024,
      "auxiliary_loss_mlp": 0.010246,
      "balance_loss_clip": 1.0480566,
      "balance_loss_mlp": 1.01664257,
      "epoch": 0.5276258041243311,
      "flos": 17749029098880.0,
      "grad_norm": 1.8054861978937125,
      "language_loss": 0.65553266,
      "learning_rate": 1.9182307744312043e-06,
      "loss": 0.67709893,
      "num_input_tokens_seen": 94852235,
      "step": 4388,
      "time_per_iteration": 3.363699436187744
    },
    {
      "auxiliary_loss_clip": 0.01150688,
      "auxiliary_loss_mlp": 0.0103115,
      "balance_loss_clip": 1.04755569,
      "balance_loss_mlp": 1.02365184,
      "epoch": 0.5277460470149702,
      "flos": 22710447077760.0,
      "grad_norm": 2.4880963675929513,
      "language_loss": 0.76714778,
      "learning_rate": 1.9174524603236676e-06,
      "loss": 0.78896612,
      "num_input_tokens_seen": 94871185,
      "step": 4389,
      "time_per_iteration": 2.585955858230591
    },
    {
      "auxiliary_loss_clip": 0.01151704,
      "auxiliary_loss_mlp": 0.01025511,
      "balance_loss_clip": 1.05048561,
      "balance_loss_mlp": 1.01736891,
      "epoch": 0.5278662899056094,
      "flos": 19902699734400.0,
      "grad_norm": 2.0773213747718327,
      "language_loss": 0.75944352,
      "learning_rate": 1.916674158738527e-06,
      "loss": 0.78121567,
      "num_input_tokens_seen": 94890090,
      "step": 4390,
      "time_per_iteration": 2.494558572769165
    },
    {
      "auxiliary_loss_clip": 0.01130051,
      "auxiliary_loss_mlp": 0.00763286,
      "balance_loss_clip": 1.04871428,
      "balance_loss_mlp": 1.00020909,
      "epoch": 0.5279865327962484,
      "flos": 18005215875840.0,
      "grad_norm": 1.8242723600729802,
      "language_loss": 0.60296273,
      "learning_rate": 1.9158958697938506e-06,
      "loss": 0.62189615,
      "num_input_tokens_seen": 94908470,
      "step": 4391,
      "time_per_iteration": 2.5011210441589355
    },
    {
      "auxiliary_loss_clip": 0.01142681,
      "auxiliary_loss_mlp": 0.01028389,
      "balance_loss_clip": 1.0460099,
      "balance_loss_mlp": 1.0204463,
      "epoch": 0.5281067756868875,
      "flos": 15924443892480.0,
      "grad_norm": 2.660231527652515,
      "language_loss": 0.85814232,
      "learning_rate": 1.9151175936077032e-06,
      "loss": 0.87985301,
      "num_input_tokens_seen": 94923440,
      "step": 4392,
      "time_per_iteration": 3.3119781017303467
    },
    {
      "auxiliary_loss_clip": 0.01159666,
      "auxiliary_loss_mlp": 0.0102813,
      "balance_loss_clip": 1.05074072,
      "balance_loss_mlp": 1.02007723,
      "epoch": 0.5282270185775266,
      "flos": 19426488197760.0,
      "grad_norm": 1.6654065150361346,
      "language_loss": 0.79473078,
      "learning_rate": 1.9143393302981507e-06,
      "loss": 0.81660873,
      "num_input_tokens_seen": 94941125,
      "step": 4393,
      "time_per_iteration": 2.439785957336426
    },
    {
      "auxiliary_loss_clip": 0.01150025,
      "auxiliary_loss_mlp": 0.01024902,
      "balance_loss_clip": 1.04806554,
      "balance_loss_mlp": 1.01733208,
      "epoch": 0.5283472614681657,
      "flos": 16399613934720.0,
      "grad_norm": 1.705385640586101,
      "language_loss": 0.83006084,
      "learning_rate": 1.913561079983252e-06,
      "loss": 0.8518101,
      "num_input_tokens_seen": 94959950,
      "step": 4394,
      "time_per_iteration": 3.2972543239593506
    },
    {
      "auxiliary_loss_clip": 0.01153413,
      "auxiliary_loss_mlp": 0.01035054,
      "balance_loss_clip": 1.04833102,
      "balance_loss_mlp": 1.02601802,
      "epoch": 0.5284675043588047,
      "flos": 26760524163840.0,
      "grad_norm": 2.0889060258322156,
      "language_loss": 0.74984896,
      "learning_rate": 1.9127828427810693e-06,
      "loss": 0.77173364,
      "num_input_tokens_seen": 94980515,
      "step": 4395,
      "time_per_iteration": 2.5200607776641846
    },
    {
      "auxiliary_loss_clip": 0.01142611,
      "auxiliary_loss_mlp": 0.01028513,
      "balance_loss_clip": 1.04810238,
      "balance_loss_mlp": 1.02055538,
      "epoch": 0.5285877472494439,
      "flos": 19899898473600.0,
      "grad_norm": 2.9597928023796944,
      "language_loss": 0.81171829,
      "learning_rate": 1.9120046188096607e-06,
      "loss": 0.83342957,
      "num_input_tokens_seen": 94998560,
      "step": 4396,
      "time_per_iteration": 2.5485148429870605
    },
    {
      "auxiliary_loss_clip": 0.01151945,
      "auxiliary_loss_mlp": 0.01036373,
      "balance_loss_clip": 1.05396557,
      "balance_loss_mlp": 1.02854121,
      "epoch": 0.528707990140083,
      "flos": 20011257613440.0,
      "grad_norm": 1.9496307115972396,
      "language_loss": 0.741799,
      "learning_rate": 1.9112264081870804e-06,
      "loss": 0.76368219,
      "num_input_tokens_seen": 95016950,
      "step": 4397,
      "time_per_iteration": 2.489238739013672
    },
    {
      "auxiliary_loss_clip": 0.01137155,
      "auxiliary_loss_mlp": 0.01030932,
      "balance_loss_clip": 1.05250216,
      "balance_loss_mlp": 1.02244473,
      "epoch": 0.528828233030722,
      "flos": 20667956014080.0,
      "grad_norm": 2.504448705693279,
      "language_loss": 0.75749385,
      "learning_rate": 1.9104482110313843e-06,
      "loss": 0.77917469,
      "num_input_tokens_seen": 95036540,
      "step": 4398,
      "time_per_iteration": 3.306952953338623
    },
    {
      "auxiliary_loss_clip": 0.01161345,
      "auxiliary_loss_mlp": 0.01025704,
      "balance_loss_clip": 1.05063009,
      "balance_loss_mlp": 1.01820564,
      "epoch": 0.5289484759213612,
      "flos": 25192448956800.0,
      "grad_norm": 1.8535938708892494,
      "language_loss": 0.73992157,
      "learning_rate": 1.909670027460623e-06,
      "loss": 0.76179206,
      "num_input_tokens_seen": 95053840,
      "step": 4399,
      "time_per_iteration": 2.4948277473449707
    },
    {
      "auxiliary_loss_clip": 0.01162441,
      "auxiliary_loss_mlp": 0.01028153,
      "balance_loss_clip": 1.05072093,
      "balance_loss_mlp": 1.02026725,
      "epoch": 0.5290687188120002,
      "flos": 31139255715840.0,
      "grad_norm": 1.7473423374003851,
      "language_loss": 0.71797311,
      "learning_rate": 1.908891857592847e-06,
      "loss": 0.73987901,
      "num_input_tokens_seen": 95074910,
      "step": 4400,
      "time_per_iteration": 2.608015298843384
    },
    {
      "auxiliary_loss_clip": 0.01131708,
      "auxiliary_loss_mlp": 0.01026469,
      "balance_loss_clip": 1.0517652,
      "balance_loss_mlp": 1.01796985,
      "epoch": 0.5291889617026393,
      "flos": 20119851406080.0,
      "grad_norm": 2.2243282920012564,
      "language_loss": 0.90379614,
      "learning_rate": 1.9081137015461034e-06,
      "loss": 0.92537796,
      "num_input_tokens_seen": 95090985,
      "step": 4401,
      "time_per_iteration": 2.4960179328918457
    },
    {
      "auxiliary_loss_clip": 0.01116025,
      "auxiliary_loss_mlp": 0.0102671,
      "balance_loss_clip": 1.04762197,
      "balance_loss_mlp": 1.01910758,
      "epoch": 0.5293092045932785,
      "flos": 19643747610240.0,
      "grad_norm": 2.049691496636775,
      "language_loss": 0.90376449,
      "learning_rate": 1.9073355594384383e-06,
      "loss": 0.92519188,
      "num_input_tokens_seen": 95109225,
      "step": 4402,
      "time_per_iteration": 2.527320146560669
    },
    {
      "auxiliary_loss_clip": 0.01130218,
      "auxiliary_loss_mlp": 0.01033117,
      "balance_loss_clip": 1.05050611,
      "balance_loss_mlp": 1.0251714,
      "epoch": 0.5294294474839175,
      "flos": 24317736958080.0,
      "grad_norm": 1.9783832381434578,
      "language_loss": 0.8036176,
      "learning_rate": 1.906557431387895e-06,
      "loss": 0.82525098,
      "num_input_tokens_seen": 95128215,
      "step": 4403,
      "time_per_iteration": 2.5449538230895996
    },
    {
      "auxiliary_loss_clip": 0.01135048,
      "auxiliary_loss_mlp": 0.01028913,
      "balance_loss_clip": 1.0544498,
      "balance_loss_mlp": 1.02008843,
      "epoch": 0.5295496903745566,
      "flos": 18875941464960.0,
      "grad_norm": 1.9139444703641642,
      "language_loss": 0.78705049,
      "learning_rate": 1.905779317512516e-06,
      "loss": 0.80869007,
      "num_input_tokens_seen": 95145760,
      "step": 4404,
      "time_per_iteration": 2.4876649379730225
    },
    {
      "auxiliary_loss_clip": 0.01162042,
      "auxiliary_loss_mlp": 0.0102849,
      "balance_loss_clip": 1.05089378,
      "balance_loss_mlp": 1.02075028,
      "epoch": 0.5296699332651957,
      "flos": 20923101296640.0,
      "grad_norm": 3.3230698633396605,
      "language_loss": 0.80456257,
      "learning_rate": 1.9050012179303385e-06,
      "loss": 0.82646787,
      "num_input_tokens_seen": 95164270,
      "step": 4405,
      "time_per_iteration": 2.4593703746795654
    },
    {
      "auxiliary_loss_clip": 0.01164464,
      "auxiliary_loss_mlp": 0.01025555,
      "balance_loss_clip": 1.04972339,
      "balance_loss_mlp": 1.01764584,
      "epoch": 0.5297901761558348,
      "flos": 22046745525120.0,
      "grad_norm": 2.572535409535337,
      "language_loss": 0.68693566,
      "learning_rate": 1.904223132759401e-06,
      "loss": 0.70883584,
      "num_input_tokens_seen": 95182870,
      "step": 4406,
      "time_per_iteration": 2.452558755874634
    },
    {
      "auxiliary_loss_clip": 0.01166124,
      "auxiliary_loss_mlp": 0.01027852,
      "balance_loss_clip": 1.05229115,
      "balance_loss_mlp": 1.01939964,
      "epoch": 0.5299104190464738,
      "flos": 21798495653760.0,
      "grad_norm": 4.087952031227363,
      "language_loss": 0.68952179,
      "learning_rate": 1.9034450621177383e-06,
      "loss": 0.71146154,
      "num_input_tokens_seen": 95201190,
      "step": 4407,
      "time_per_iteration": 2.446545362472534
    },
    {
      "auxiliary_loss_clip": 0.01162265,
      "auxiliary_loss_mlp": 0.01038402,
      "balance_loss_clip": 1.05196393,
      "balance_loss_mlp": 1.02988458,
      "epoch": 0.530030661937113,
      "flos": 14720790119040.0,
      "grad_norm": 2.0279941268381005,
      "language_loss": 0.70698452,
      "learning_rate": 1.9026670061233824e-06,
      "loss": 0.72899115,
      "num_input_tokens_seen": 95218625,
      "step": 4408,
      "time_per_iteration": 2.41420578956604
    },
    {
      "auxiliary_loss_clip": 0.01144505,
      "auxiliary_loss_mlp": 0.01025791,
      "balance_loss_clip": 1.05098414,
      "balance_loss_mlp": 1.01783323,
      "epoch": 0.5301509048277521,
      "flos": 21251504367360.0,
      "grad_norm": 1.7368374521502858,
      "language_loss": 0.80731225,
      "learning_rate": 1.901888964894365e-06,
      "loss": 0.82901514,
      "num_input_tokens_seen": 95237665,
      "step": 4409,
      "time_per_iteration": 2.4782917499542236
    },
    {
      "auxiliary_loss_clip": 0.01178522,
      "auxiliary_loss_mlp": 0.01028883,
      "balance_loss_clip": 1.0519352,
      "balance_loss_mlp": 1.02119994,
      "epoch": 0.5302711477183911,
      "flos": 25957058791680.0,
      "grad_norm": 1.7546288430541206,
      "language_loss": 0.67922366,
      "learning_rate": 1.9011109385487134e-06,
      "loss": 0.70129764,
      "num_input_tokens_seen": 95258915,
      "step": 4410,
      "time_per_iteration": 2.444899320602417
    },
    {
      "auxiliary_loss_clip": 0.01178073,
      "auxiliary_loss_mlp": 0.01027917,
      "balance_loss_clip": 1.05082726,
      "balance_loss_mlp": 1.01964402,
      "epoch": 0.5303913906090303,
      "flos": 22273126992000.0,
      "grad_norm": 2.7118171127523176,
      "language_loss": 0.66615164,
      "learning_rate": 1.900332927204454e-06,
      "loss": 0.68821156,
      "num_input_tokens_seen": 95277365,
      "step": 4411,
      "time_per_iteration": 2.4106783866882324
    },
    {
      "auxiliary_loss_clip": 0.0115675,
      "auxiliary_loss_mlp": 0.01023546,
      "balance_loss_clip": 1.05109882,
      "balance_loss_mlp": 1.01582718,
      "epoch": 0.5305116334996693,
      "flos": 24936010784640.0,
      "grad_norm": 1.895793965911426,
      "language_loss": 0.76777911,
      "learning_rate": 1.8995549309796097e-06,
      "loss": 0.78958213,
      "num_input_tokens_seen": 95296670,
      "step": 4412,
      "time_per_iteration": 2.529984712600708
    },
    {
      "auxiliary_loss_clip": 0.01171853,
      "auxiliary_loss_mlp": 0.01028034,
      "balance_loss_clip": 1.05387521,
      "balance_loss_mlp": 1.02017188,
      "epoch": 0.5306318763903084,
      "flos": 20189338266240.0,
      "grad_norm": 2.0649350378005105,
      "language_loss": 0.76648295,
      "learning_rate": 1.8987769499922028e-06,
      "loss": 0.78848183,
      "num_input_tokens_seen": 95315640,
      "step": 4413,
      "time_per_iteration": 2.4816269874572754
    },
    {
      "auxiliary_loss_clip": 0.01162868,
      "auxiliary_loss_mlp": 0.00762269,
      "balance_loss_clip": 1.05118632,
      "balance_loss_mlp": 1.00029254,
      "epoch": 0.5307521192809476,
      "flos": 20266366982400.0,
      "grad_norm": 2.431334215031439,
      "language_loss": 0.7075935,
      "learning_rate": 1.897998984360252e-06,
      "loss": 0.72684491,
      "num_input_tokens_seen": 95334610,
      "step": 4414,
      "time_per_iteration": 2.458519220352173
    },
    {
      "auxiliary_loss_clip": 0.01147173,
      "auxiliary_loss_mlp": 0.0102515,
      "balance_loss_clip": 1.05042887,
      "balance_loss_mlp": 1.01741982,
      "epoch": 0.5308723621715866,
      "flos": 28844276976000.0,
      "grad_norm": 1.653642362517932,
      "language_loss": 0.78462231,
      "learning_rate": 1.897221034201775e-06,
      "loss": 0.80634552,
      "num_input_tokens_seen": 95358350,
      "step": 4415,
      "time_per_iteration": 3.3944060802459717
    },
    {
      "auxiliary_loss_clip": 0.01135287,
      "auxiliary_loss_mlp": 0.01027561,
      "balance_loss_clip": 1.04646444,
      "balance_loss_mlp": 1.02037251,
      "epoch": 0.5309926050622257,
      "flos": 27457766040960.0,
      "grad_norm": 1.5624555616937665,
      "language_loss": 0.6680097,
      "learning_rate": 1.8964430996347842e-06,
      "loss": 0.68963814,
      "num_input_tokens_seen": 95379900,
      "step": 4416,
      "time_per_iteration": 2.588054656982422
    },
    {
      "auxiliary_loss_clip": 0.01148879,
      "auxiliary_loss_mlp": 0.01025753,
      "balance_loss_clip": 1.04925323,
      "balance_loss_mlp": 1.01759923,
      "epoch": 0.5311128479528648,
      "flos": 20514545026560.0,
      "grad_norm": 3.7600022749501645,
      "language_loss": 0.82547688,
      "learning_rate": 1.8956651807772931e-06,
      "loss": 0.84722322,
      "num_input_tokens_seen": 95397935,
      "step": 4417,
      "time_per_iteration": 2.478142261505127
    },
    {
      "auxiliary_loss_clip": 0.01160405,
      "auxiliary_loss_mlp": 0.01022453,
      "balance_loss_clip": 1.05012894,
      "balance_loss_mlp": 1.01545537,
      "epoch": 0.5312330908435039,
      "flos": 21397660807680.0,
      "grad_norm": 2.1443934624516383,
      "language_loss": 0.83948964,
      "learning_rate": 1.8948872777473115e-06,
      "loss": 0.86131823,
      "num_input_tokens_seen": 95415890,
      "step": 4418,
      "time_per_iteration": 2.4721696376800537
    },
    {
      "auxiliary_loss_clip": 0.0114691,
      "auxiliary_loss_mlp": 0.01027225,
      "balance_loss_clip": 1.04902434,
      "balance_loss_mlp": 1.02008104,
      "epoch": 0.531353333734143,
      "flos": 24717350741760.0,
      "grad_norm": 1.8927535149723085,
      "language_loss": 0.63758755,
      "learning_rate": 1.8941093906628458e-06,
      "loss": 0.65932882,
      "num_input_tokens_seen": 95433675,
      "step": 4419,
      "time_per_iteration": 3.3488175868988037
    },
    {
      "auxiliary_loss_clip": 0.01140728,
      "auxiliary_loss_mlp": 0.01023722,
      "balance_loss_clip": 1.04614508,
      "balance_loss_mlp": 1.01649237,
      "epoch": 0.531473576624782,
      "flos": 30480689808000.0,
      "grad_norm": 1.7407172442118777,
      "language_loss": 0.70690119,
      "learning_rate": 1.893331519641902e-06,
      "loss": 0.72854573,
      "num_input_tokens_seen": 95455820,
      "step": 4420,
      "time_per_iteration": 2.550868034362793
    },
    {
      "auxiliary_loss_clip": 0.01122979,
      "auxiliary_loss_mlp": 0.01027267,
      "balance_loss_clip": 1.04337502,
      "balance_loss_mlp": 1.01942933,
      "epoch": 0.5315938195154212,
      "flos": 23002975440000.0,
      "grad_norm": 2.741893329887903,
      "language_loss": 0.7380811,
      "learning_rate": 1.8925536648024815e-06,
      "loss": 0.75958359,
      "num_input_tokens_seen": 95473240,
      "step": 4421,
      "time_per_iteration": 3.335223913192749
    },
    {
      "auxiliary_loss_clip": 0.01177398,
      "auxiliary_loss_mlp": 0.01025099,
      "balance_loss_clip": 1.05210996,
      "balance_loss_mlp": 1.01724613,
      "epoch": 0.5317140624060602,
      "flos": 22748584343040.0,
      "grad_norm": 2.3155794764712474,
      "language_loss": 0.75978935,
      "learning_rate": 1.8917758262625849e-06,
      "loss": 0.78181434,
      "num_input_tokens_seen": 95493480,
      "step": 4422,
      "time_per_iteration": 2.426776885986328
    },
    {
      "auxiliary_loss_clip": 0.01140854,
      "auxiliary_loss_mlp": 0.01029249,
      "balance_loss_clip": 1.04850554,
      "balance_loss_mlp": 1.02175641,
      "epoch": 0.5318343052966993,
      "flos": 22821087945600.0,
      "grad_norm": 1.6146765656286153,
      "language_loss": 0.8065474,
      "learning_rate": 1.8909980041402089e-06,
      "loss": 0.82824844,
      "num_input_tokens_seen": 95512075,
      "step": 4423,
      "time_per_iteration": 2.4910166263580322
    },
    {
      "auxiliary_loss_clip": 0.01157005,
      "auxiliary_loss_mlp": 0.0102723,
      "balance_loss_clip": 1.04844546,
      "balance_loss_mlp": 1.01927853,
      "epoch": 0.5319545481873384,
      "flos": 13626089274240.0,
      "grad_norm": 2.1907451519042183,
      "language_loss": 0.65487504,
      "learning_rate": 1.8902201985533494e-06,
      "loss": 0.67671734,
      "num_input_tokens_seen": 95529340,
      "step": 4424,
      "time_per_iteration": 2.4017388820648193
    },
    {
      "auxiliary_loss_clip": 0.01146654,
      "auxiliary_loss_mlp": 0.01022627,
      "balance_loss_clip": 1.04792094,
      "balance_loss_mlp": 1.0155158,
      "epoch": 0.5320747910779775,
      "flos": 22162522037760.0,
      "grad_norm": 1.8611239151710266,
      "language_loss": 0.74934757,
      "learning_rate": 1.8894424096199983e-06,
      "loss": 0.77104038,
      "num_input_tokens_seen": 95548545,
      "step": 4425,
      "time_per_iteration": 3.307497501373291
    },
    {
      "auxiliary_loss_clip": 0.01165348,
      "auxiliary_loss_mlp": 0.01026803,
      "balance_loss_clip": 1.0534761,
      "balance_loss_mlp": 1.01850557,
      "epoch": 0.5321950339686166,
      "flos": 18588081870720.0,
      "grad_norm": 1.9229733687298485,
      "language_loss": 0.85934412,
      "learning_rate": 1.8886646374581463e-06,
      "loss": 0.88126564,
      "num_input_tokens_seen": 95567770,
      "step": 4426,
      "time_per_iteration": 2.428509473800659
    },
    {
      "auxiliary_loss_clip": 0.01160504,
      "auxiliary_loss_mlp": 0.01026745,
      "balance_loss_clip": 1.04919577,
      "balance_loss_mlp": 1.01886821,
      "epoch": 0.5323152768592557,
      "flos": 22856818999680.0,
      "grad_norm": 1.8387797003995645,
      "language_loss": 0.71122217,
      "learning_rate": 1.8878868821857795e-06,
      "loss": 0.73309469,
      "num_input_tokens_seen": 95587420,
      "step": 4427,
      "time_per_iteration": 2.4547226428985596
    },
    {
      "auxiliary_loss_clip": 0.01117063,
      "auxiliary_loss_mlp": 0.01029392,
      "balance_loss_clip": 1.04383039,
      "balance_loss_mlp": 1.02111292,
      "epoch": 0.5324355197498948,
      "flos": 33948690998400.0,
      "grad_norm": 2.541021035417326,
      "language_loss": 0.751773,
      "learning_rate": 1.8871091439208838e-06,
      "loss": 0.77323759,
      "num_input_tokens_seen": 95609030,
      "step": 4428,
      "time_per_iteration": 2.6807432174682617
    },
    {
      "auxiliary_loss_clip": 0.01120159,
      "auxiliary_loss_mlp": 0.01030063,
      "balance_loss_clip": 1.04662943,
      "balance_loss_mlp": 1.02164638,
      "epoch": 0.5325557626405338,
      "flos": 23256720092160.0,
      "grad_norm": 2.148824894659783,
      "language_loss": 0.77255893,
      "learning_rate": 1.8863314227814414e-06,
      "loss": 0.79406106,
      "num_input_tokens_seen": 95627340,
      "step": 4429,
      "time_per_iteration": 2.5454797744750977
    },
    {
      "auxiliary_loss_clip": 0.01169561,
      "auxiliary_loss_mlp": 0.01027903,
      "balance_loss_clip": 1.05370152,
      "balance_loss_mlp": 1.01973116,
      "epoch": 0.532676005531173,
      "flos": 26718687797760.0,
      "grad_norm": 2.203540722026896,
      "language_loss": 0.48745137,
      "learning_rate": 1.8855537188854313e-06,
      "loss": 0.50942606,
      "num_input_tokens_seen": 95646315,
      "step": 4430,
      "time_per_iteration": 2.4973628520965576
    },
    {
      "auxiliary_loss_clip": 0.01163668,
      "auxiliary_loss_mlp": 0.01028293,
      "balance_loss_clip": 1.04846144,
      "balance_loss_mlp": 1.02065754,
      "epoch": 0.5327962484218121,
      "flos": 17894610921600.0,
      "grad_norm": 2.4094500990062464,
      "language_loss": 0.78297079,
      "learning_rate": 1.8847760323508315e-06,
      "loss": 0.80489039,
      "num_input_tokens_seen": 95665220,
      "step": 4431,
      "time_per_iteration": 2.4343512058258057
    },
    {
      "auxiliary_loss_clip": 0.01143125,
      "auxiliary_loss_mlp": 0.01026057,
      "balance_loss_clip": 1.04928517,
      "balance_loss_mlp": 1.01900911,
      "epoch": 0.5329164913124511,
      "flos": 17925385898880.0,
      "grad_norm": 1.9279859207312136,
      "language_loss": 0.7563687,
      "learning_rate": 1.883998363295616e-06,
      "loss": 0.77806056,
      "num_input_tokens_seen": 95682700,
      "step": 4432,
      "time_per_iteration": 2.4565107822418213
    },
    {
      "auxiliary_loss_clip": 0.01054434,
      "auxiliary_loss_mlp": 0.0100109,
      "balance_loss_clip": 1.01677632,
      "balance_loss_mlp": 1.00011253,
      "epoch": 0.5330367342030903,
      "flos": 57254178781440.0,
      "grad_norm": 0.87809699125139,
      "language_loss": 0.62535334,
      "learning_rate": 1.8832207118377565e-06,
      "loss": 0.64590859,
      "num_input_tokens_seen": 95738070,
      "step": 4433,
      "time_per_iteration": 2.9914746284484863
    },
    {
      "auxiliary_loss_clip": 0.01173744,
      "auxiliary_loss_mlp": 0.01024413,
      "balance_loss_clip": 1.05066645,
      "balance_loss_mlp": 1.01730776,
      "epoch": 0.5331569770937293,
      "flos": 17420518287360.0,
      "grad_norm": 2.126230536835848,
      "language_loss": 0.69381464,
      "learning_rate": 1.882443078095222e-06,
      "loss": 0.71579623,
      "num_input_tokens_seen": 95756950,
      "step": 4434,
      "time_per_iteration": 2.394221067428589
    },
    {
      "auxiliary_loss_clip": 0.01043321,
      "auxiliary_loss_mlp": 0.01001025,
      "balance_loss_clip": 1.02038908,
      "balance_loss_mlp": 1.00010157,
      "epoch": 0.5332772199843684,
      "flos": 56750783627520.0,
      "grad_norm": 1.1329709108492303,
      "language_loss": 0.6684711,
      "learning_rate": 1.8816654621859794e-06,
      "loss": 0.68891454,
      "num_input_tokens_seen": 95816615,
      "step": 4435,
      "time_per_iteration": 3.0088424682617188
    },
    {
      "auxiliary_loss_clip": 0.01174566,
      "auxiliary_loss_mlp": 0.0102574,
      "balance_loss_clip": 1.05182791,
      "balance_loss_mlp": 1.01775885,
      "epoch": 0.5333974628750076,
      "flos": 18697753071360.0,
      "grad_norm": 2.1951282833747707,
      "language_loss": 0.72302198,
      "learning_rate": 1.8808878642279915e-06,
      "loss": 0.74502504,
      "num_input_tokens_seen": 95832020,
      "step": 4436,
      "time_per_iteration": 2.408475399017334
    },
    {
      "auxiliary_loss_clip": 0.0113391,
      "auxiliary_loss_mlp": 0.01032853,
      "balance_loss_clip": 1.0433085,
      "balance_loss_mlp": 1.02476478,
      "epoch": 0.5335177057656466,
      "flos": 23805507058560.0,
      "grad_norm": 2.454058446815523,
      "language_loss": 0.6479131,
      "learning_rate": 1.8801102843392209e-06,
      "loss": 0.6695807,
      "num_input_tokens_seen": 95851425,
      "step": 4437,
      "time_per_iteration": 2.5382144451141357
    },
    {
      "auxiliary_loss_clip": 0.01132168,
      "auxiliary_loss_mlp": 0.01027486,
      "balance_loss_clip": 1.0454185,
      "balance_loss_mlp": 1.02004147,
      "epoch": 0.5336379486562857,
      "flos": 25078683605760.0,
      "grad_norm": 1.611145090353668,
      "language_loss": 0.8530699,
      "learning_rate": 1.8793327226376238e-06,
      "loss": 0.87466645,
      "num_input_tokens_seen": 95870745,
      "step": 4438,
      "time_per_iteration": 2.5607376098632812
    },
    {
      "auxiliary_loss_clip": 0.01155704,
      "auxiliary_loss_mlp": 0.0102699,
      "balance_loss_clip": 1.04951477,
      "balance_loss_mlp": 1.0193429,
      "epoch": 0.5337581915469248,
      "flos": 21396691140480.0,
      "grad_norm": 1.7876418179343254,
      "language_loss": 0.80082142,
      "learning_rate": 1.8785551792411569e-06,
      "loss": 0.82264841,
      "num_input_tokens_seen": 95889755,
      "step": 4439,
      "time_per_iteration": 2.4819371700286865
    },
    {
      "auxiliary_loss_clip": 0.01147436,
      "auxiliary_loss_mlp": 0.01028686,
      "balance_loss_clip": 1.04863691,
      "balance_loss_mlp": 1.0216918,
      "epoch": 0.5338784344375639,
      "flos": 14865905064960.0,
      "grad_norm": 2.0018118718720026,
      "language_loss": 0.82300633,
      "learning_rate": 1.8777776542677733e-06,
      "loss": 0.84476757,
      "num_input_tokens_seen": 95907805,
      "step": 4440,
      "time_per_iteration": 2.4573745727539062
    },
    {
      "auxiliary_loss_clip": 0.01132546,
      "auxiliary_loss_mlp": 0.01023086,
      "balance_loss_clip": 1.0441972,
      "balance_loss_mlp": 1.01518857,
      "epoch": 0.5339986773282029,
      "flos": 20813501923200.0,
      "grad_norm": 1.9381061106172475,
      "language_loss": 0.73208892,
      "learning_rate": 1.8770001478354216e-06,
      "loss": 0.7536453,
      "num_input_tokens_seen": 95927480,
      "step": 4441,
      "time_per_iteration": 2.5188307762145996
    },
    {
      "auxiliary_loss_clip": 0.01157611,
      "auxiliary_loss_mlp": 0.0103235,
      "balance_loss_clip": 1.04856658,
      "balance_loss_mlp": 1.02410674,
      "epoch": 0.5341189202188421,
      "flos": 17969089772160.0,
      "grad_norm": 2.334046273550115,
      "language_loss": 0.84375846,
      "learning_rate": 1.8762226600620504e-06,
      "loss": 0.86565804,
      "num_input_tokens_seen": 95946095,
      "step": 4442,
      "time_per_iteration": 3.258622407913208
    },
    {
      "auxiliary_loss_clip": 0.0115465,
      "auxiliary_loss_mlp": 0.01026939,
      "balance_loss_clip": 1.04857564,
      "balance_loss_mlp": 1.01852918,
      "epoch": 0.5342391631094812,
      "flos": 11031866328960.0,
      "grad_norm": 3.895016297064507,
      "language_loss": 0.58954144,
      "learning_rate": 1.8754451910656031e-06,
      "loss": 0.61135733,
      "num_input_tokens_seen": 95959995,
      "step": 4443,
      "time_per_iteration": 2.4188201427459717
    },
    {
      "auxiliary_loss_clip": 0.01128019,
      "auxiliary_loss_mlp": 0.0102669,
      "balance_loss_clip": 1.04714465,
      "balance_loss_mlp": 1.0186522,
      "epoch": 0.5343594060001202,
      "flos": 15339135772800.0,
      "grad_norm": 2.0982498329638704,
      "language_loss": 0.82721788,
      "learning_rate": 1.8746677409640212e-06,
      "loss": 0.84876502,
      "num_input_tokens_seen": 95977095,
      "step": 4444,
      "time_per_iteration": 2.5219223499298096
    },
    {
      "auxiliary_loss_clip": 0.01167921,
      "auxiliary_loss_mlp": 0.01028401,
      "balance_loss_clip": 1.05392075,
      "balance_loss_mlp": 1.02075338,
      "epoch": 0.5344796488907594,
      "flos": 26900898514560.0,
      "grad_norm": 1.7665490761804004,
      "language_loss": 0.84337252,
      "learning_rate": 1.8738903098752432e-06,
      "loss": 0.86533576,
      "num_input_tokens_seen": 95996225,
      "step": 4445,
      "time_per_iteration": 2.4985456466674805
    },
    {
      "auxiliary_loss_clip": 0.01148807,
      "auxiliary_loss_mlp": 0.01030856,
      "balance_loss_clip": 1.04945087,
      "balance_loss_mlp": 1.02342033,
      "epoch": 0.5345998917813984,
      "flos": 25411216740480.0,
      "grad_norm": 2.271288353522816,
      "language_loss": 0.73147124,
      "learning_rate": 1.8731128979172052e-06,
      "loss": 0.75326788,
      "num_input_tokens_seen": 96015425,
      "step": 4446,
      "time_per_iteration": 3.371224880218506
    },
    {
      "auxiliary_loss_clip": 0.01145833,
      "auxiliary_loss_mlp": 0.01022238,
      "balance_loss_clip": 1.04989493,
      "balance_loss_mlp": 1.01518083,
      "epoch": 0.5347201346720375,
      "flos": 32853379622400.0,
      "grad_norm": 2.1882611576014437,
      "language_loss": 0.66941226,
      "learning_rate": 1.8723355052078394e-06,
      "loss": 0.69109297,
      "num_input_tokens_seen": 96035460,
      "step": 4447,
      "time_per_iteration": 2.618471384048462
    },
    {
      "auxiliary_loss_clip": 0.01159026,
      "auxiliary_loss_mlp": 0.01035384,
      "balance_loss_clip": 1.04825974,
      "balance_loss_mlp": 1.02736092,
      "epoch": 0.5348403775626767,
      "flos": 17967940536960.0,
      "grad_norm": 2.674220123911979,
      "language_loss": 0.77289617,
      "learning_rate": 1.8715581318650765e-06,
      "loss": 0.79484028,
      "num_input_tokens_seen": 96054515,
      "step": 4448,
      "time_per_iteration": 3.2570197582244873
    },
    {
      "auxiliary_loss_clip": 0.011453,
      "auxiliary_loss_mlp": 0.01029161,
      "balance_loss_clip": 1.05004001,
      "balance_loss_mlp": 1.02024376,
      "epoch": 0.5349606204533157,
      "flos": 17603339535360.0,
      "grad_norm": 2.2044857888588596,
      "language_loss": 0.81117606,
      "learning_rate": 1.8707807780068422e-06,
      "loss": 0.83292067,
      "num_input_tokens_seen": 96072330,
      "step": 4449,
      "time_per_iteration": 2.5061838626861572
    },
    {
      "auxiliary_loss_clip": 0.01144826,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 1.04815948,
      "balance_loss_mlp": 1.01901841,
      "epoch": 0.5350808633439548,
      "flos": 29167831710720.0,
      "grad_norm": 2.551763531400763,
      "language_loss": 0.66170347,
      "learning_rate": 1.8700034437510611e-06,
      "loss": 0.68341517,
      "num_input_tokens_seen": 96092425,
      "step": 4450,
      "time_per_iteration": 2.5877599716186523
    },
    {
      "auxiliary_loss_clip": 0.01125631,
      "auxiliary_loss_mlp": 0.01025652,
      "balance_loss_clip": 1.0466001,
      "balance_loss_mlp": 1.01777864,
      "epoch": 0.5352011062345938,
      "flos": 19499997381120.0,
      "grad_norm": 2.2407233072706485,
      "language_loss": 0.81731796,
      "learning_rate": 1.8692261292156549e-06,
      "loss": 0.83883083,
      "num_input_tokens_seen": 96111660,
      "step": 4451,
      "time_per_iteration": 2.520409345626831
    },
    {
      "auxiliary_loss_clip": 0.01176956,
      "auxiliary_loss_mlp": 0.01024901,
      "balance_loss_clip": 1.054739,
      "balance_loss_mlp": 1.01735544,
      "epoch": 0.535321349125233,
      "flos": 23477642691840.0,
      "grad_norm": 1.9449492563437474,
      "language_loss": 0.81255102,
      "learning_rate": 1.8684488345185401e-06,
      "loss": 0.83456957,
      "num_input_tokens_seen": 96131835,
      "step": 4452,
      "time_per_iteration": 3.1671388149261475
    },
    {
      "auxiliary_loss_clip": 0.01178818,
      "auxiliary_loss_mlp": 0.0102924,
      "balance_loss_clip": 1.05379868,
      "balance_loss_mlp": 1.02142,
      "epoch": 0.535441592015872,
      "flos": 20478059786880.0,
      "grad_norm": 2.289697788667961,
      "language_loss": 0.78501272,
      "learning_rate": 1.8676715597776332e-06,
      "loss": 0.80709332,
      "num_input_tokens_seen": 96150180,
      "step": 4453,
      "time_per_iteration": 2.398439645767212
    },
    {
      "auxiliary_loss_clip": 0.01110919,
      "auxiliary_loss_mlp": 0.0102319,
      "balance_loss_clip": 1.04223621,
      "balance_loss_mlp": 1.01565957,
      "epoch": 0.5355618349065111,
      "flos": 19573147428480.0,
      "grad_norm": 1.807117462478234,
      "language_loss": 0.76476395,
      "learning_rate": 1.8668943051108455e-06,
      "loss": 0.78610504,
      "num_input_tokens_seen": 96167485,
      "step": 4454,
      "time_per_iteration": 2.528853416442871
    },
    {
      "auxiliary_loss_clip": 0.01146749,
      "auxiliary_loss_mlp": 0.01030506,
      "balance_loss_clip": 1.04812694,
      "balance_loss_mlp": 1.0226028,
      "epoch": 0.5356820777971503,
      "flos": 24024633978240.0,
      "grad_norm": 1.8669990840015993,
      "language_loss": 0.76142764,
      "learning_rate": 1.8661170706360856e-06,
      "loss": 0.7832002,
      "num_input_tokens_seen": 96186650,
      "step": 4455,
      "time_per_iteration": 2.507431745529175
    },
    {
      "auxiliary_loss_clip": 0.0116157,
      "auxiliary_loss_mlp": 0.01022897,
      "balance_loss_clip": 1.0519228,
      "balance_loss_mlp": 1.01577687,
      "epoch": 0.5358023206877893,
      "flos": 20884676722560.0,
      "grad_norm": 1.6291379145055414,
      "language_loss": 0.81508356,
      "learning_rate": 1.8653398564712594e-06,
      "loss": 0.83692825,
      "num_input_tokens_seen": 96205595,
      "step": 4456,
      "time_per_iteration": 2.446989059448242
    },
    {
      "auxiliary_loss_clip": 0.01160343,
      "auxiliary_loss_mlp": 0.01024168,
      "balance_loss_clip": 1.05177546,
      "balance_loss_mlp": 1.01643705,
      "epoch": 0.5359225635784284,
      "flos": 22418996123520.0,
      "grad_norm": 1.5592992000963437,
      "language_loss": 0.82007861,
      "learning_rate": 1.8645626627342704e-06,
      "loss": 0.84192371,
      "num_input_tokens_seen": 96226360,
      "step": 4457,
      "time_per_iteration": 2.4673218727111816
    },
    {
      "auxiliary_loss_clip": 0.01165079,
      "auxiliary_loss_mlp": 0.01028229,
      "balance_loss_clip": 1.05072236,
      "balance_loss_mlp": 1.02073061,
      "epoch": 0.5360428064690675,
      "flos": 24097784025600.0,
      "grad_norm": 2.2760354984773654,
      "language_loss": 0.80978495,
      "learning_rate": 1.8637854895430172e-06,
      "loss": 0.83171803,
      "num_input_tokens_seen": 96245625,
      "step": 4458,
      "time_per_iteration": 2.476638078689575
    },
    {
      "auxiliary_loss_clip": 0.01127204,
      "auxiliary_loss_mlp": 0.0102944,
      "balance_loss_clip": 1.04690289,
      "balance_loss_mlp": 1.02123857,
      "epoch": 0.5361630493597066,
      "flos": 21434505183360.0,
      "grad_norm": 2.137333711046165,
      "language_loss": 0.69852883,
      "learning_rate": 1.8630083370153978e-06,
      "loss": 0.72009528,
      "num_input_tokens_seen": 96265265,
      "step": 4459,
      "time_per_iteration": 2.5222668647766113
    },
    {
      "auxiliary_loss_clip": 0.01028763,
      "auxiliary_loss_mlp": 0.01001179,
      "balance_loss_clip": 1.01792812,
      "balance_loss_mlp": 1.00026083,
      "epoch": 0.5362832922503457,
      "flos": 68888696520960.0,
      "grad_norm": 0.7515503345002652,
      "language_loss": 0.55431956,
      "learning_rate": 1.8622312052693041e-06,
      "loss": 0.57461894,
      "num_input_tokens_seen": 96326445,
      "step": 4460,
      "time_per_iteration": 3.224353075027466
    },
    {
      "auxiliary_loss_clip": 0.01152905,
      "auxiliary_loss_mlp": 0.01026892,
      "balance_loss_clip": 1.04578114,
      "balance_loss_mlp": 1.01929271,
      "epoch": 0.5364035351409848,
      "flos": 9793702563840.0,
      "grad_norm": 2.3747290463549957,
      "language_loss": 0.72064066,
      "learning_rate": 1.8614540944226267e-06,
      "loss": 0.74243867,
      "num_input_tokens_seen": 96343115,
      "step": 4461,
      "time_per_iteration": 2.4199695587158203
    },
    {
      "auxiliary_loss_clip": 0.01143205,
      "auxiliary_loss_mlp": 0.01024452,
      "balance_loss_clip": 1.04959953,
      "balance_loss_mlp": 1.01750493,
      "epoch": 0.5365237780316239,
      "flos": 23290080848640.0,
      "grad_norm": 1.8957466065967739,
      "language_loss": 0.67709768,
      "learning_rate": 1.8606770045932537e-06,
      "loss": 0.69877422,
      "num_input_tokens_seen": 96362230,
      "step": 4462,
      "time_per_iteration": 2.5026862621307373
    },
    {
      "auxiliary_loss_clip": 0.01126294,
      "auxiliary_loss_mlp": 0.01029348,
      "balance_loss_clip": 1.04264879,
      "balance_loss_mlp": 1.02093208,
      "epoch": 0.5366440209222629,
      "flos": 26578133879040.0,
      "grad_norm": 2.385155724961322,
      "language_loss": 0.81893027,
      "learning_rate": 1.859899935899068e-06,
      "loss": 0.84048671,
      "num_input_tokens_seen": 96382085,
      "step": 4463,
      "time_per_iteration": 2.5726125240325928
    },
    {
      "auxiliary_loss_clip": 0.01148911,
      "auxiliary_loss_mlp": 0.01027171,
      "balance_loss_clip": 1.05319619,
      "balance_loss_mlp": 1.01915979,
      "epoch": 0.5367642638129021,
      "flos": 19608052469760.0,
      "grad_norm": 1.5633399524107081,
      "language_loss": 0.7887969,
      "learning_rate": 1.8591228884579506e-06,
      "loss": 0.81055772,
      "num_input_tokens_seen": 96400580,
      "step": 4464,
      "time_per_iteration": 2.474151849746704
    },
    {
      "auxiliary_loss_clip": 0.01137231,
      "auxiliary_loss_mlp": 0.01026211,
      "balance_loss_clip": 1.04786599,
      "balance_loss_mlp": 1.01870084,
      "epoch": 0.5368845067035412,
      "flos": 23915214172800.0,
      "grad_norm": 2.001505398609328,
      "language_loss": 0.82025689,
      "learning_rate": 1.8583458623877795e-06,
      "loss": 0.84189129,
      "num_input_tokens_seen": 96419680,
      "step": 4465,
      "time_per_iteration": 2.533148765563965
    },
    {
      "auxiliary_loss_clip": 0.01163519,
      "auxiliary_loss_mlp": 0.0102476,
      "balance_loss_clip": 1.05115592,
      "balance_loss_mlp": 1.01730335,
      "epoch": 0.5370047495941802,
      "flos": 16873131951360.0,
      "grad_norm": 1.7100307378037336,
      "language_loss": 0.74166769,
      "learning_rate": 1.8575688578064281e-06,
      "loss": 0.76355052,
      "num_input_tokens_seen": 96437805,
      "step": 4466,
      "time_per_iteration": 2.419611692428589
    },
    {
      "auxiliary_loss_clip": 0.0116479,
      "auxiliary_loss_mlp": 0.01027861,
      "balance_loss_clip": 1.05205464,
      "balance_loss_mlp": 1.02023792,
      "epoch": 0.5371249924848194,
      "flos": 20740926493440.0,
      "grad_norm": 1.6992476666578011,
      "language_loss": 0.76564705,
      "learning_rate": 1.8567918748317674e-06,
      "loss": 0.78757358,
      "num_input_tokens_seen": 96457155,
      "step": 4467,
      "time_per_iteration": 2.457407236099243
    },
    {
      "auxiliary_loss_clip": 0.01131867,
      "auxiliary_loss_mlp": 0.01027422,
      "balance_loss_clip": 1.04425859,
      "balance_loss_mlp": 1.0196259,
      "epoch": 0.5372452353754584,
      "flos": 17968120104960.0,
      "grad_norm": 2.164692405242359,
      "language_loss": 0.82604158,
      "learning_rate": 1.8560149135816659e-06,
      "loss": 0.84763443,
      "num_input_tokens_seen": 96473990,
      "step": 4468,
      "time_per_iteration": 3.3103084564208984
    },
    {
      "auxiliary_loss_clip": 0.01156866,
      "auxiliary_loss_mlp": 0.01021639,
      "balance_loss_clip": 1.04751825,
      "balance_loss_mlp": 1.01446533,
      "epoch": 0.5373654782660975,
      "flos": 15377021642880.0,
      "grad_norm": 2.1594839537498522,
      "language_loss": 0.84159189,
      "learning_rate": 1.8552379741739873e-06,
      "loss": 0.86337698,
      "num_input_tokens_seen": 96491335,
      "step": 4469,
      "time_per_iteration": 2.421666383743286
    },
    {
      "auxiliary_loss_clip": 0.01045487,
      "auxiliary_loss_mlp": 0.00752661,
      "balance_loss_clip": 1.01706469,
      "balance_loss_mlp": 0.99971449,
      "epoch": 0.5374857211567367,
      "flos": 69000091574400.0,
      "grad_norm": 0.8942068885640814,
      "language_loss": 0.55684155,
      "learning_rate": 1.8544610567265935e-06,
      "loss": 0.57482308,
      "num_input_tokens_seen": 96545275,
      "step": 4470,
      "time_per_iteration": 3.03814959526062
    },
    {
      "auxiliary_loss_clip": 0.0114831,
      "auxiliary_loss_mlp": 0.00761958,
      "balance_loss_clip": 1.05115497,
      "balance_loss_mlp": 1.00026214,
      "epoch": 0.5376059640473757,
      "flos": 15085355207040.0,
      "grad_norm": 1.9626685817636675,
      "language_loss": 0.83195674,
      "learning_rate": 1.853684161357341e-06,
      "loss": 0.85105944,
      "num_input_tokens_seen": 96562935,
      "step": 4471,
      "time_per_iteration": 2.464038133621216
    },
    {
      "auxiliary_loss_clip": 0.01159627,
      "auxiliary_loss_mlp": 0.0076238,
      "balance_loss_clip": 1.05127382,
      "balance_loss_mlp": 1.00029433,
      "epoch": 0.5377262069380148,
      "flos": 19792597570560.0,
      "grad_norm": 1.7383531331988016,
      "language_loss": 0.76809514,
      "learning_rate": 1.852907288184085e-06,
      "loss": 0.78731519,
      "num_input_tokens_seen": 96581820,
      "step": 4472,
      "time_per_iteration": 2.4673213958740234
    },
    {
      "auxiliary_loss_clip": 0.01123999,
      "auxiliary_loss_mlp": 0.01027439,
      "balance_loss_clip": 1.04639482,
      "balance_loss_mlp": 1.01884377,
      "epoch": 0.5378464498286539,
      "flos": 30003077640960.0,
      "grad_norm": 1.8400786012806205,
      "language_loss": 0.70284921,
      "learning_rate": 1.8521304373246762e-06,
      "loss": 0.72436363,
      "num_input_tokens_seen": 96602865,
      "step": 4473,
      "time_per_iteration": 3.4835758209228516
    },
    {
      "auxiliary_loss_clip": 0.01165097,
      "auxiliary_loss_mlp": 0.01025631,
      "balance_loss_clip": 1.04973698,
      "balance_loss_mlp": 1.01779938,
      "epoch": 0.537966692719293,
      "flos": 21251217058560.0,
      "grad_norm": 2.506134031203014,
      "language_loss": 0.88833678,
      "learning_rate": 1.8513536088969626e-06,
      "loss": 0.91024411,
      "num_input_tokens_seen": 96620530,
      "step": 4474,
      "time_per_iteration": 3.2373931407928467
    },
    {
      "auxiliary_loss_clip": 0.01164577,
      "auxiliary_loss_mlp": 0.0103712,
      "balance_loss_clip": 1.05295444,
      "balance_loss_mlp": 1.02890062,
      "epoch": 0.538086935609932,
      "flos": 21543170803200.0,
      "grad_norm": 2.1503063923741546,
      "language_loss": 0.80407178,
      "learning_rate": 1.8505768030187884e-06,
      "loss": 0.82608879,
      "num_input_tokens_seen": 96640660,
      "step": 4475,
      "time_per_iteration": 2.4650754928588867
    },
    {
      "auxiliary_loss_clip": 0.01144412,
      "auxiliary_loss_mlp": 0.01025949,
      "balance_loss_clip": 1.05049706,
      "balance_loss_mlp": 1.01882017,
      "epoch": 0.5382071785005712,
      "flos": 22747219626240.0,
      "grad_norm": 1.7119117056191293,
      "language_loss": 0.8012085,
      "learning_rate": 1.849800019807995e-06,
      "loss": 0.8229121,
      "num_input_tokens_seen": 96661885,
      "step": 4476,
      "time_per_iteration": 2.526031970977783
    },
    {
      "auxiliary_loss_clip": 0.0113173,
      "auxiliary_loss_mlp": 0.0102614,
      "balance_loss_clip": 1.04729128,
      "balance_loss_mlp": 1.01858521,
      "epoch": 0.5383274213912103,
      "flos": 24934574240640.0,
      "grad_norm": 2.210722727540462,
      "language_loss": 0.70869958,
      "learning_rate": 1.8490232593824186e-06,
      "loss": 0.73027825,
      "num_input_tokens_seen": 96678340,
      "step": 4477,
      "time_per_iteration": 2.5267958641052246
    },
    {
      "auxiliary_loss_clip": 0.01147608,
      "auxiliary_loss_mlp": 0.01026257,
      "balance_loss_clip": 1.05252624,
      "balance_loss_mlp": 1.01952422,
      "epoch": 0.5384476642818493,
      "flos": 22310186849280.0,
      "grad_norm": 1.763550013622752,
      "language_loss": 0.84696341,
      "learning_rate": 1.8482465218598935e-06,
      "loss": 0.86870199,
      "num_input_tokens_seen": 96698285,
      "step": 4478,
      "time_per_iteration": 3.2263917922973633
    },
    {
      "auxiliary_loss_clip": 0.01133589,
      "auxiliary_loss_mlp": 0.01027434,
      "balance_loss_clip": 1.04629946,
      "balance_loss_mlp": 1.01925039,
      "epoch": 0.5385679071724885,
      "flos": 22711021695360.0,
      "grad_norm": 1.8017459716605533,
      "language_loss": 0.83167791,
      "learning_rate": 1.8474698073582508e-06,
      "loss": 0.85328817,
      "num_input_tokens_seen": 96719655,
      "step": 4479,
      "time_per_iteration": 2.5472662448883057
    },
    {
      "auxiliary_loss_clip": 0.01138837,
      "auxiliary_loss_mlp": 0.01021095,
      "balance_loss_clip": 1.04722977,
      "balance_loss_mlp": 1.01362062,
      "epoch": 0.5386881500631275,
      "flos": 15953746412160.0,
      "grad_norm": 1.9907662895863352,
      "language_loss": 0.87314087,
      "learning_rate": 1.8466931159953166e-06,
      "loss": 0.89474022,
      "num_input_tokens_seen": 96736290,
      "step": 4480,
      "time_per_iteration": 2.486788034439087
    },
    {
      "auxiliary_loss_clip": 0.01153111,
      "auxiliary_loss_mlp": 0.01028763,
      "balance_loss_clip": 1.05212665,
      "balance_loss_mlp": 1.02114248,
      "epoch": 0.5388083929537666,
      "flos": 24060041809920.0,
      "grad_norm": 2.394991686298086,
      "language_loss": 0.84316683,
      "learning_rate": 1.8459164478889158e-06,
      "loss": 0.86498547,
      "num_input_tokens_seen": 96757685,
      "step": 4481,
      "time_per_iteration": 2.5128936767578125
    },
    {
      "auxiliary_loss_clip": 0.01127108,
      "auxiliary_loss_mlp": 0.01023799,
      "balance_loss_clip": 1.04517388,
      "balance_loss_mlp": 1.01647949,
      "epoch": 0.5389286358444056,
      "flos": 22236893147520.0,
      "grad_norm": 1.875004365145341,
      "language_loss": 0.76047671,
      "learning_rate": 1.8451398031568663e-06,
      "loss": 0.78198576,
      "num_input_tokens_seen": 96777310,
      "step": 4482,
      "time_per_iteration": 2.516167640686035
    },
    {
      "auxiliary_loss_clip": 0.01134465,
      "auxiliary_loss_mlp": 0.01024904,
      "balance_loss_clip": 1.04816866,
      "balance_loss_mlp": 1.01709616,
      "epoch": 0.5390488787350448,
      "flos": 24281718595200.0,
      "grad_norm": 1.7031201848368782,
      "language_loss": 0.74754083,
      "learning_rate": 1.844363181916986e-06,
      "loss": 0.76913452,
      "num_input_tokens_seen": 96798035,
      "step": 4483,
      "time_per_iteration": 2.5668463706970215
    },
    {
      "auxiliary_loss_clip": 0.01159778,
      "auxiliary_loss_mlp": 0.01029984,
      "balance_loss_clip": 1.04962683,
      "balance_loss_mlp": 1.02264643,
      "epoch": 0.5391691216256839,
      "flos": 16581393688320.0,
      "grad_norm": 2.250537525079556,
      "language_loss": 0.83371311,
      "learning_rate": 1.8435865842870868e-06,
      "loss": 0.85561067,
      "num_input_tokens_seen": 96815975,
      "step": 4484,
      "time_per_iteration": 2.437711477279663
    },
    {
      "auxiliary_loss_clip": 0.01138163,
      "auxiliary_loss_mlp": 0.0076231,
      "balance_loss_clip": 1.0449276,
      "balance_loss_mlp": 1.00037479,
      "epoch": 0.5392893645163229,
      "flos": 23330049707520.0,
      "grad_norm": 2.0537061908818086,
      "language_loss": 0.71873277,
      "learning_rate": 1.8428100103849787e-06,
      "loss": 0.73773748,
      "num_input_tokens_seen": 96835770,
      "step": 4485,
      "time_per_iteration": 2.5080618858337402
    },
    {
      "auxiliary_loss_clip": 0.01150014,
      "auxiliary_loss_mlp": 0.01031514,
      "balance_loss_clip": 1.05330825,
      "balance_loss_mlp": 1.02337241,
      "epoch": 0.5394096074069621,
      "flos": 15669801400320.0,
      "grad_norm": 2.4248632676515145,
      "language_loss": 0.73307145,
      "learning_rate": 1.842033460328467e-06,
      "loss": 0.75488675,
      "num_input_tokens_seen": 96854490,
      "step": 4486,
      "time_per_iteration": 2.455778121948242
    },
    {
      "auxiliary_loss_clip": 0.01150211,
      "auxiliary_loss_mlp": 0.00761969,
      "balance_loss_clip": 1.04753304,
      "balance_loss_mlp": 1.0003593,
      "epoch": 0.5395298502976011,
      "flos": 22893447893760.0,
      "grad_norm": 1.6631617719100145,
      "language_loss": 0.75173771,
      "learning_rate": 1.8412569342353541e-06,
      "loss": 0.77085948,
      "num_input_tokens_seen": 96874645,
      "step": 4487,
      "time_per_iteration": 2.5174357891082764
    },
    {
      "auxiliary_loss_clip": 0.01153241,
      "auxiliary_loss_mlp": 0.01032649,
      "balance_loss_clip": 1.05108953,
      "balance_loss_mlp": 1.02436388,
      "epoch": 0.5396500931882402,
      "flos": 23842135952640.0,
      "grad_norm": 1.9005491259605878,
      "language_loss": 0.84796524,
      "learning_rate": 1.840480432223438e-06,
      "loss": 0.86982417,
      "num_input_tokens_seen": 96893650,
      "step": 4488,
      "time_per_iteration": 2.4916694164276123
    },
    {
      "auxiliary_loss_clip": 0.01148741,
      "auxiliary_loss_mlp": 0.01030293,
      "balance_loss_clip": 1.04751098,
      "balance_loss_mlp": 1.02270222,
      "epoch": 0.5397703360788794,
      "flos": 26322988596480.0,
      "grad_norm": 2.205160208525085,
      "language_loss": 0.77600622,
      "learning_rate": 1.8397039544105131e-06,
      "loss": 0.79779655,
      "num_input_tokens_seen": 96912735,
      "step": 4489,
      "time_per_iteration": 2.534141778945923
    },
    {
      "auxiliary_loss_clip": 0.01142902,
      "auxiliary_loss_mlp": 0.01024235,
      "balance_loss_clip": 1.04639983,
      "balance_loss_mlp": 1.01648653,
      "epoch": 0.5398905789695184,
      "flos": 21214588164480.0,
      "grad_norm": 2.154275094202878,
      "language_loss": 0.69819635,
      "learning_rate": 1.8389275009143711e-06,
      "loss": 0.71986771,
      "num_input_tokens_seen": 96932475,
      "step": 4490,
      "time_per_iteration": 2.4854838848114014
    },
    {
      "auxiliary_loss_clip": 0.01172929,
      "auxiliary_loss_mlp": 0.0102558,
      "balance_loss_clip": 1.05064297,
      "balance_loss_mlp": 1.01840913,
      "epoch": 0.5400108218601575,
      "flos": 25080335631360.0,
      "grad_norm": 1.8663062078000634,
      "language_loss": 0.73665971,
      "learning_rate": 1.8381510718527988e-06,
      "loss": 0.75864476,
      "num_input_tokens_seen": 96952085,
      "step": 4491,
      "time_per_iteration": 2.4466052055358887
    },
    {
      "auxiliary_loss_clip": 0.01151174,
      "auxiliary_loss_mlp": 0.01029152,
      "balance_loss_clip": 1.04742694,
      "balance_loss_mlp": 1.02110505,
      "epoch": 0.5401310647507966,
      "flos": 26357498588160.0,
      "grad_norm": 2.5472658261863463,
      "language_loss": 0.63529658,
      "learning_rate": 1.8373746673435812e-06,
      "loss": 0.65709984,
      "num_input_tokens_seen": 96973110,
      "step": 4492,
      "time_per_iteration": 2.532409191131592
    },
    {
      "auxiliary_loss_clip": 0.01179352,
      "auxiliary_loss_mlp": 0.0102842,
      "balance_loss_clip": 1.05430782,
      "balance_loss_mlp": 1.02051091,
      "epoch": 0.5402513076414357,
      "flos": 27855332749440.0,
      "grad_norm": 1.661301080479555,
      "language_loss": 0.79172289,
      "learning_rate": 1.8365982875044964e-06,
      "loss": 0.81380063,
      "num_input_tokens_seen": 96993420,
      "step": 4493,
      "time_per_iteration": 2.4797379970550537
    },
    {
      "auxiliary_loss_clip": 0.01169397,
      "auxiliary_loss_mlp": 0.00762604,
      "balance_loss_clip": 1.05331743,
      "balance_loss_mlp": 1.00028503,
      "epoch": 0.5403715505320748,
      "flos": 22893771116160.0,
      "grad_norm": 1.9877258192457343,
      "language_loss": 0.7570402,
      "learning_rate": 1.8358219324533217e-06,
      "loss": 0.77636015,
      "num_input_tokens_seen": 97013685,
      "step": 4494,
      "time_per_iteration": 2.471734046936035
    },
    {
      "auxiliary_loss_clip": 0.01143205,
      "auxiliary_loss_mlp": 0.01025245,
      "balance_loss_clip": 1.04836488,
      "balance_loss_mlp": 1.01842606,
      "epoch": 0.5404917934227139,
      "flos": 30224143895040.0,
      "grad_norm": 1.875463548352243,
      "language_loss": 0.70485508,
      "learning_rate": 1.8350456023078292e-06,
      "loss": 0.72653961,
      "num_input_tokens_seen": 97036060,
      "step": 4495,
      "time_per_iteration": 3.3996193408966064
    },
    {
      "auxiliary_loss_clip": 0.01180751,
      "auxiliary_loss_mlp": 0.01033635,
      "balance_loss_clip": 1.05342209,
      "balance_loss_mlp": 1.02522516,
      "epoch": 0.540612036313353,
      "flos": 19938502615680.0,
      "grad_norm": 2.780571036390765,
      "language_loss": 0.78138185,
      "learning_rate": 1.8342692971857874e-06,
      "loss": 0.80352569,
      "num_input_tokens_seen": 97055260,
      "step": 4496,
      "time_per_iteration": 2.4275875091552734
    },
    {
      "auxiliary_loss_clip": 0.01147921,
      "auxiliary_loss_mlp": 0.01028973,
      "balance_loss_clip": 1.05072308,
      "balance_loss_mlp": 1.02148688,
      "epoch": 0.540732279203992,
      "flos": 24279599692800.0,
      "grad_norm": 2.1974390425763413,
      "language_loss": 0.71137136,
      "learning_rate": 1.833493017204962e-06,
      "loss": 0.73314023,
      "num_input_tokens_seen": 97075365,
      "step": 4497,
      "time_per_iteration": 2.558467149734497
    },
    {
      "auxiliary_loss_clip": 0.01176532,
      "auxiliary_loss_mlp": 0.01033103,
      "balance_loss_clip": 1.05226088,
      "balance_loss_mlp": 1.02559316,
      "epoch": 0.5408525220946312,
      "flos": 20193216935040.0,
      "grad_norm": 1.8664053892363894,
      "language_loss": 0.78216189,
      "learning_rate": 1.8327167624831134e-06,
      "loss": 0.80425823,
      "num_input_tokens_seen": 97093095,
      "step": 4498,
      "time_per_iteration": 2.4242374897003174
    },
    {
      "auxiliary_loss_clip": 0.01174873,
      "auxiliary_loss_mlp": 0.01027145,
      "balance_loss_clip": 1.05251908,
      "balance_loss_mlp": 1.01994157,
      "epoch": 0.5409727649852702,
      "flos": 24134448833280.0,
      "grad_norm": 1.5840676542609582,
      "language_loss": 0.71054977,
      "learning_rate": 1.831940533137999e-06,
      "loss": 0.73256993,
      "num_input_tokens_seen": 97112000,
      "step": 4499,
      "time_per_iteration": 3.3257219791412354
    },
    {
      "auxiliary_loss_clip": 0.01161919,
      "auxiliary_loss_mlp": 0.01026878,
      "balance_loss_clip": 1.05461526,
      "balance_loss_mlp": 1.01951051,
      "epoch": 0.5410930078759093,
      "flos": 23912700220800.0,
      "grad_norm": 1.6962871266940485,
      "language_loss": 0.72297829,
      "learning_rate": 1.8311643292873718e-06,
      "loss": 0.74486625,
      "num_input_tokens_seen": 97130820,
      "step": 4500,
      "time_per_iteration": 2.4935522079467773
    },
    {
      "auxiliary_loss_clip": 0.01158953,
      "auxiliary_loss_mlp": 0.01027909,
      "balance_loss_clip": 1.05154514,
      "balance_loss_mlp": 1.02086663,
      "epoch": 0.5412132507665485,
      "flos": 21105132445440.0,
      "grad_norm": 2.0953335547982097,
      "language_loss": 0.87982357,
      "learning_rate": 1.8303881510489818e-06,
      "loss": 0.90169221,
      "num_input_tokens_seen": 97149210,
      "step": 4501,
      "time_per_iteration": 3.255511522293091
    },
    {
      "auxiliary_loss_clip": 0.01150309,
      "auxiliary_loss_mlp": 0.01027774,
      "balance_loss_clip": 1.05197513,
      "balance_loss_mlp": 1.01922679,
      "epoch": 0.5413334936571875,
      "flos": 30227340205440.0,
      "grad_norm": 2.464181921881129,
      "language_loss": 0.69331312,
      "learning_rate": 1.829611998540574e-06,
      "loss": 0.71509397,
      "num_input_tokens_seen": 97170415,
      "step": 4502,
      "time_per_iteration": 2.5475172996520996
    },
    {
      "auxiliary_loss_clip": 0.01163828,
      "auxiliary_loss_mlp": 0.0076218,
      "balance_loss_clip": 1.05030084,
      "balance_loss_mlp": 1.00030172,
      "epoch": 0.5414537365478266,
      "flos": 24279635606400.0,
      "grad_norm": 2.1929211885137927,
      "language_loss": 0.79975057,
      "learning_rate": 1.8288358718798914e-06,
      "loss": 0.81901073,
      "num_input_tokens_seen": 97189605,
      "step": 4503,
      "time_per_iteration": 2.4653518199920654
    },
    {
      "auxiliary_loss_clip": 0.01155891,
      "auxiliary_loss_mlp": 0.00761946,
      "balance_loss_clip": 1.04989314,
      "balance_loss_mlp": 1.00035596,
      "epoch": 0.5415739794384657,
      "flos": 16654543735680.0,
      "grad_norm": 1.7906156978222616,
      "language_loss": 0.72438109,
      "learning_rate": 1.8280597711846703e-06,
      "loss": 0.74355948,
      "num_input_tokens_seen": 97207845,
      "step": 4504,
      "time_per_iteration": 2.4137439727783203
    },
    {
      "auxiliary_loss_clip": 0.01161612,
      "auxiliary_loss_mlp": 0.01022528,
      "balance_loss_clip": 1.05430222,
      "balance_loss_mlp": 1.01501155,
      "epoch": 0.5416942223291048,
      "flos": 23185724860800.0,
      "grad_norm": 1.7553121924385462,
      "language_loss": 0.83786285,
      "learning_rate": 1.8272836965726455e-06,
      "loss": 0.8597042,
      "num_input_tokens_seen": 97226780,
      "step": 4505,
      "time_per_iteration": 3.207233190536499
    },
    {
      "auxiliary_loss_clip": 0.01104327,
      "auxiliary_loss_mlp": 0.01027141,
      "balance_loss_clip": 1.04199433,
      "balance_loss_mlp": 1.01915944,
      "epoch": 0.5418144652197439,
      "flos": 20303247271680.0,
      "grad_norm": 1.7384649819730214,
      "language_loss": 0.78313279,
      "learning_rate": 1.8265076481615461e-06,
      "loss": 0.80444747,
      "num_input_tokens_seen": 97246695,
      "step": 4506,
      "time_per_iteration": 2.5814051628112793
    },
    {
      "auxiliary_loss_clip": 0.01147569,
      "auxiliary_loss_mlp": 0.01030564,
      "balance_loss_clip": 1.05069005,
      "balance_loss_mlp": 1.02223718,
      "epoch": 0.541934708110383,
      "flos": 12458633431680.0,
      "grad_norm": 2.161994067558727,
      "language_loss": 0.87442434,
      "learning_rate": 1.8257316260690987e-06,
      "loss": 0.89620566,
      "num_input_tokens_seen": 97264480,
      "step": 4507,
      "time_per_iteration": 2.5089099407196045
    },
    {
      "auxiliary_loss_clip": 0.01161769,
      "auxiliary_loss_mlp": 0.01017635,
      "balance_loss_clip": 1.05010569,
      "balance_loss_mlp": 1.01073325,
      "epoch": 0.5420549510010221,
      "flos": 21253802837760.0,
      "grad_norm": 1.4342178975487165,
      "language_loss": 0.75699228,
      "learning_rate": 1.8249556304130254e-06,
      "loss": 0.7787863,
      "num_input_tokens_seen": 97285760,
      "step": 4508,
      "time_per_iteration": 2.461242198944092
    },
    {
      "auxiliary_loss_clip": 0.01138354,
      "auxiliary_loss_mlp": 0.01029662,
      "balance_loss_clip": 1.04738355,
      "balance_loss_mlp": 1.02181256,
      "epoch": 0.5421751938916611,
      "flos": 29490524519040.0,
      "grad_norm": 1.9113721024560573,
      "language_loss": 0.68916261,
      "learning_rate": 1.824179661311044e-06,
      "loss": 0.71084273,
      "num_input_tokens_seen": 97304510,
      "step": 4509,
      "time_per_iteration": 2.5242104530334473
    },
    {
      "auxiliary_loss_clip": 0.01116441,
      "auxiliary_loss_mlp": 0.01024462,
      "balance_loss_clip": 1.04052949,
      "balance_loss_mlp": 1.01673722,
      "epoch": 0.5422954367823003,
      "flos": 18734238311040.0,
      "grad_norm": 1.9054684177628225,
      "language_loss": 0.79972827,
      "learning_rate": 1.823403718880868e-06,
      "loss": 0.82113731,
      "num_input_tokens_seen": 97323270,
      "step": 4510,
      "time_per_iteration": 2.5528883934020996
    },
    {
      "auxiliary_loss_clip": 0.01146336,
      "auxiliary_loss_mlp": 0.01029674,
      "balance_loss_clip": 1.04513681,
      "balance_loss_mlp": 1.02189541,
      "epoch": 0.5424156796729394,
      "flos": 39969006940800.0,
      "grad_norm": 1.584981514043345,
      "language_loss": 0.66644168,
      "learning_rate": 1.822627803240207e-06,
      "loss": 0.68820179,
      "num_input_tokens_seen": 97345600,
      "step": 4511,
      "time_per_iteration": 2.646287441253662
    },
    {
      "auxiliary_loss_clip": 0.01136932,
      "auxiliary_loss_mlp": 0.01029103,
      "balance_loss_clip": 1.04874229,
      "balance_loss_mlp": 1.02182531,
      "epoch": 0.5425359225635784,
      "flos": 11546538353280.0,
      "grad_norm": 2.1479674754870053,
      "language_loss": 0.84785694,
      "learning_rate": 1.8218519145067675e-06,
      "loss": 0.86951733,
      "num_input_tokens_seen": 97361220,
      "step": 4512,
      "time_per_iteration": 2.4993834495544434
    },
    {
      "auxiliary_loss_clip": 0.01126343,
      "auxiliary_loss_mlp": 0.01028071,
      "balance_loss_clip": 1.04443717,
      "balance_loss_mlp": 1.02036405,
      "epoch": 0.5426561654542175,
      "flos": 20229702174720.0,
      "grad_norm": 2.1856135700902257,
      "language_loss": 0.89349431,
      "learning_rate": 1.8210760527982508e-06,
      "loss": 0.91503847,
      "num_input_tokens_seen": 97381505,
      "step": 4513,
      "time_per_iteration": 2.519304037094116
    },
    {
      "auxiliary_loss_clip": 0.01149375,
      "auxiliary_loss_mlp": 0.00762314,
      "balance_loss_clip": 1.05110645,
      "balance_loss_mlp": 1.00032353,
      "epoch": 0.5427764083448566,
      "flos": 21871681614720.0,
      "grad_norm": 1.7386745694831183,
      "language_loss": 0.75077903,
      "learning_rate": 1.8203002182323552e-06,
      "loss": 0.76989591,
      "num_input_tokens_seen": 97399060,
      "step": 4514,
      "time_per_iteration": 2.5139713287353516
    },
    {
      "auxiliary_loss_clip": 0.01152836,
      "auxiliary_loss_mlp": 0.01025624,
      "balance_loss_clip": 1.05245519,
      "balance_loss_mlp": 1.01754737,
      "epoch": 0.5428966512354957,
      "flos": 19640946349440.0,
      "grad_norm": 1.9764111876447796,
      "language_loss": 0.7580114,
      "learning_rate": 1.819524410926773e-06,
      "loss": 0.779796,
      "num_input_tokens_seen": 97416740,
      "step": 4515,
      "time_per_iteration": 2.479501247406006
    },
    {
      "auxiliary_loss_clip": 0.01103346,
      "auxiliary_loss_mlp": 0.01027456,
      "balance_loss_clip": 1.04651427,
      "balance_loss_mlp": 1.01951098,
      "epoch": 0.5430168941261347,
      "flos": 22382187661440.0,
      "grad_norm": 1.4678912406432258,
      "language_loss": 0.77089119,
      "learning_rate": 1.8187486309991944e-06,
      "loss": 0.79219925,
      "num_input_tokens_seen": 97437620,
      "step": 4516,
      "time_per_iteration": 2.607738733291626
    },
    {
      "auxiliary_loss_clip": 0.01166328,
      "auxiliary_loss_mlp": 0.01029861,
      "balance_loss_clip": 1.0521723,
      "balance_loss_mlp": 1.0227294,
      "epoch": 0.5431371370167739,
      "flos": 18764187275520.0,
      "grad_norm": 1.649220096398923,
      "language_loss": 0.7728231,
      "learning_rate": 1.817972878567304e-06,
      "loss": 0.7947849,
      "num_input_tokens_seen": 97456275,
      "step": 4517,
      "time_per_iteration": 2.4406261444091797
    },
    {
      "auxiliary_loss_clip": 0.01151926,
      "auxiliary_loss_mlp": 0.01027294,
      "balance_loss_clip": 1.04838395,
      "balance_loss_mlp": 1.02008462,
      "epoch": 0.543257379907413,
      "flos": 18806023641600.0,
      "grad_norm": 1.7383614917474701,
      "language_loss": 0.76066166,
      "learning_rate": 1.8171971537487834e-06,
      "loss": 0.78245389,
      "num_input_tokens_seen": 97474925,
      "step": 4518,
      "time_per_iteration": 2.49815034866333
    },
    {
      "auxiliary_loss_clip": 0.01174344,
      "auxiliary_loss_mlp": 0.01031816,
      "balance_loss_clip": 1.05001843,
      "balance_loss_mlp": 1.02399635,
      "epoch": 0.543377622798052,
      "flos": 17493381025920.0,
      "grad_norm": 1.830820324531856,
      "language_loss": 0.80420929,
      "learning_rate": 1.8164214566613093e-06,
      "loss": 0.82627088,
      "num_input_tokens_seen": 97493550,
      "step": 4519,
      "time_per_iteration": 2.411087989807129
    },
    {
      "auxiliary_loss_clip": 0.01172077,
      "auxiliary_loss_mlp": 0.01023134,
      "balance_loss_clip": 1.04954863,
      "balance_loss_mlp": 1.01589847,
      "epoch": 0.5434978656886912,
      "flos": 18989311766400.0,
      "grad_norm": 4.114227920719771,
      "language_loss": 0.65828776,
      "learning_rate": 1.8156457874225547e-06,
      "loss": 0.68023992,
      "num_input_tokens_seen": 97512010,
      "step": 4520,
      "time_per_iteration": 2.4073808193206787
    },
    {
      "auxiliary_loss_clip": 0.01139216,
      "auxiliary_loss_mlp": 0.01023683,
      "balance_loss_clip": 1.04899454,
      "balance_loss_mlp": 1.01642895,
      "epoch": 0.5436181085793302,
      "flos": 17274936464640.0,
      "grad_norm": 2.035172728444253,
      "language_loss": 0.80638015,
      "learning_rate": 1.814870146150187e-06,
      "loss": 0.82800913,
      "num_input_tokens_seen": 97530120,
      "step": 4521,
      "time_per_iteration": 2.4470407962799072
    },
    {
      "auxiliary_loss_clip": 0.01152884,
      "auxiliary_loss_mlp": 0.0103139,
      "balance_loss_clip": 1.04701042,
      "balance_loss_mlp": 1.02381146,
      "epoch": 0.5437383514699693,
      "flos": 19098587917440.0,
      "grad_norm": 1.91983004723685,
      "language_loss": 0.78594959,
      "learning_rate": 1.814094532961871e-06,
      "loss": 0.80779231,
      "num_input_tokens_seen": 97548695,
      "step": 4522,
      "time_per_iteration": 3.2551491260528564
    },
    {
      "auxiliary_loss_clip": 0.01120395,
      "auxiliary_loss_mlp": 0.01028786,
      "balance_loss_clip": 1.04434252,
      "balance_loss_mlp": 1.02104342,
      "epoch": 0.5438585943606085,
      "flos": 22602715211520.0,
      "grad_norm": 1.846615871248621,
      "language_loss": 0.83428198,
      "learning_rate": 1.8133189479752666e-06,
      "loss": 0.85577381,
      "num_input_tokens_seen": 97567625,
      "step": 4523,
      "time_per_iteration": 2.554713249206543
    },
    {
      "auxiliary_loss_clip": 0.01139024,
      "auxiliary_loss_mlp": 0.01027661,
      "balance_loss_clip": 1.04557085,
      "balance_loss_mlp": 1.02073824,
      "epoch": 0.5439788372512475,
      "flos": 21798495653760.0,
      "grad_norm": 1.8881763630042188,
      "language_loss": 0.81947517,
      "learning_rate": 1.8125433913080292e-06,
      "loss": 0.84114206,
      "num_input_tokens_seen": 97585325,
      "step": 4524,
      "time_per_iteration": 2.479999303817749
    },
    {
      "auxiliary_loss_clip": 0.01089927,
      "auxiliary_loss_mlp": 0.01025363,
      "balance_loss_clip": 1.0459249,
      "balance_loss_mlp": 1.01899123,
      "epoch": 0.5440990801418866,
      "flos": 16399362539520.0,
      "grad_norm": 2.0509675596091728,
      "language_loss": 0.82522714,
      "learning_rate": 1.811767863077811e-06,
      "loss": 0.84638011,
      "num_input_tokens_seen": 97604275,
      "step": 4525,
      "time_per_iteration": 2.579888105392456
    },
    {
      "auxiliary_loss_clip": 0.01097783,
      "auxiliary_loss_mlp": 0.01027814,
      "balance_loss_clip": 1.04573405,
      "balance_loss_mlp": 1.02042603,
      "epoch": 0.5442193230325257,
      "flos": 21615638492160.0,
      "grad_norm": 1.6406692432042242,
      "language_loss": 0.78299081,
      "learning_rate": 1.8109923634022577e-06,
      "loss": 0.80424678,
      "num_input_tokens_seen": 97624300,
      "step": 4526,
      "time_per_iteration": 3.4092373847961426
    },
    {
      "auxiliary_loss_clip": 0.01177091,
      "auxiliary_loss_mlp": 0.01025861,
      "balance_loss_clip": 1.05138004,
      "balance_loss_mlp": 1.0180887,
      "epoch": 0.5443395659231648,
      "flos": 15481198062720.0,
      "grad_norm": 1.9315214886879715,
      "language_loss": 0.86531496,
      "learning_rate": 1.8102168923990128e-06,
      "loss": 0.88734448,
      "num_input_tokens_seen": 97637845,
      "step": 4527,
      "time_per_iteration": 2.3992555141448975
    },
    {
      "auxiliary_loss_clip": 0.01165053,
      "auxiliary_loss_mlp": 0.00761583,
      "balance_loss_clip": 1.05297613,
      "balance_loss_mlp": 1.00039887,
      "epoch": 0.5444598088138038,
      "flos": 18770436241920.0,
      "grad_norm": 1.792593614094289,
      "language_loss": 0.79970706,
      "learning_rate": 1.809441450185714e-06,
      "loss": 0.81897336,
      "num_input_tokens_seen": 97656330,
      "step": 4528,
      "time_per_iteration": 3.2790322303771973
    },
    {
      "auxiliary_loss_clip": 0.011512,
      "auxiliary_loss_mlp": 0.01024985,
      "balance_loss_clip": 1.04636526,
      "balance_loss_mlp": 1.01722479,
      "epoch": 0.544580051704443,
      "flos": 21142335957120.0,
      "grad_norm": 2.0519548152882843,
      "language_loss": 0.73393506,
      "learning_rate": 1.8086660368799958e-06,
      "loss": 0.75569689,
      "num_input_tokens_seen": 97674380,
      "step": 4529,
      "time_per_iteration": 2.4725565910339355
    },
    {
      "auxiliary_loss_clip": 0.01150644,
      "auxiliary_loss_mlp": 0.01020466,
      "balance_loss_clip": 1.05085957,
      "balance_loss_mlp": 1.01263428,
      "epoch": 0.5447002945950821,
      "flos": 32491508054400.0,
      "grad_norm": 1.7249744067724055,
      "language_loss": 0.77068698,
      "learning_rate": 1.807890652599488e-06,
      "loss": 0.7923981,
      "num_input_tokens_seen": 97698765,
      "step": 4530,
      "time_per_iteration": 2.585057258605957
    },
    {
      "auxiliary_loss_clip": 0.01172829,
      "auxiliary_loss_mlp": 0.0102648,
      "balance_loss_clip": 1.05118966,
      "balance_loss_mlp": 1.01974797,
      "epoch": 0.5448205374857211,
      "flos": 11798307757440.0,
      "grad_norm": 4.200980973443889,
      "language_loss": 0.82388544,
      "learning_rate": 1.8071152974618156e-06,
      "loss": 0.84587854,
      "num_input_tokens_seen": 97716565,
      "step": 4531,
      "time_per_iteration": 3.148620128631592
    },
    {
      "auxiliary_loss_clip": 0.01134952,
      "auxiliary_loss_mlp": 0.00761862,
      "balance_loss_clip": 1.04592037,
      "balance_loss_mlp": 1.00027609,
      "epoch": 0.5449407803763603,
      "flos": 24133766474880.0,
      "grad_norm": 2.3330085471769504,
      "language_loss": 0.78289199,
      "learning_rate": 1.806339971584599e-06,
      "loss": 0.80186015,
      "num_input_tokens_seen": 97733225,
      "step": 4532,
      "time_per_iteration": 2.5186805725097656
    },
    {
      "auxiliary_loss_clip": 0.01174376,
      "auxiliary_loss_mlp": 0.01025538,
      "balance_loss_clip": 1.05073059,
      "balance_loss_mlp": 1.01780176,
      "epoch": 0.5450610232669993,
      "flos": 23258551685760.0,
      "grad_norm": 2.000358025561131,
      "language_loss": 0.8523401,
      "learning_rate": 1.8055646750854546e-06,
      "loss": 0.87433922,
      "num_input_tokens_seen": 97752735,
      "step": 4533,
      "time_per_iteration": 2.430197238922119
    },
    {
      "auxiliary_loss_clip": 0.01149936,
      "auxiliary_loss_mlp": 0.01023404,
      "balance_loss_clip": 1.04878426,
      "balance_loss_mlp": 1.01578069,
      "epoch": 0.5451812661576384,
      "flos": 17785083375360.0,
      "grad_norm": 2.186783014702608,
      "language_loss": 0.81599176,
      "learning_rate": 1.8047894080819945e-06,
      "loss": 0.83772516,
      "num_input_tokens_seen": 97769985,
      "step": 4534,
      "time_per_iteration": 2.4424550533294678
    },
    {
      "auxiliary_loss_clip": 0.01077564,
      "auxiliary_loss_mlp": 0.01000203,
      "balance_loss_clip": 1.01713634,
      "balance_loss_mlp": 0.99932688,
      "epoch": 0.5453015090482776,
      "flos": 71062586513280.0,
      "grad_norm": 0.7222235935203467,
      "language_loss": 0.63152587,
      "learning_rate": 1.8040141706918258e-06,
      "loss": 0.65230346,
      "num_input_tokens_seen": 97831225,
      "step": 4535,
      "time_per_iteration": 3.1208040714263916
    },
    {
      "auxiliary_loss_clip": 0.01148813,
      "auxiliary_loss_mlp": 0.0102608,
      "balance_loss_clip": 1.05027533,
      "balance_loss_mlp": 1.01842642,
      "epoch": 0.5454217519389166,
      "flos": 25552201622400.0,
      "grad_norm": 1.7454166229967965,
      "language_loss": 0.76542628,
      "learning_rate": 1.8032389630325525e-06,
      "loss": 0.78717518,
      "num_input_tokens_seen": 97849975,
      "step": 4536,
      "time_per_iteration": 2.5153892040252686
    },
    {
      "auxiliary_loss_clip": 0.01144809,
      "auxiliary_loss_mlp": 0.01029455,
      "balance_loss_clip": 1.0441432,
      "balance_loss_mlp": 1.02153361,
      "epoch": 0.5455419948295557,
      "flos": 23658345037440.0,
      "grad_norm": 1.6372953379951058,
      "language_loss": 0.75886774,
      "learning_rate": 1.8024637852217707e-06,
      "loss": 0.78061038,
      "num_input_tokens_seen": 97869700,
      "step": 4537,
      "time_per_iteration": 2.5058796405792236
    },
    {
      "auxiliary_loss_clip": 0.01146537,
      "auxiliary_loss_mlp": 0.01033976,
      "balance_loss_clip": 1.04838705,
      "balance_loss_mlp": 1.02599454,
      "epoch": 0.5456622377201948,
      "flos": 23403989854080.0,
      "grad_norm": 1.8888518967027725,
      "language_loss": 0.84811908,
      "learning_rate": 1.8016886373770766e-06,
      "loss": 0.86992419,
      "num_input_tokens_seen": 97888215,
      "step": 4538,
      "time_per_iteration": 2.4896507263183594
    },
    {
      "auxiliary_loss_clip": 0.01147538,
      "auxiliary_loss_mlp": 0.0102518,
      "balance_loss_clip": 1.04890561,
      "balance_loss_mlp": 1.01766419,
      "epoch": 0.5457824806108339,
      "flos": 23988040997760.0,
      "grad_norm": 1.70619246750523,
      "language_loss": 0.78872037,
      "learning_rate": 1.8009135196160579e-06,
      "loss": 0.81044763,
      "num_input_tokens_seen": 97907090,
      "step": 4539,
      "time_per_iteration": 2.5665969848632812
    },
    {
      "auxiliary_loss_clip": 0.01129304,
      "auxiliary_loss_mlp": 0.01027329,
      "balance_loss_clip": 1.04558575,
      "balance_loss_mlp": 1.02077579,
      "epoch": 0.545902723501473,
      "flos": 22565870835840.0,
      "grad_norm": 1.6616023518629373,
      "language_loss": 0.84290826,
      "learning_rate": 1.8001384320563e-06,
      "loss": 0.86447465,
      "num_input_tokens_seen": 97927345,
      "step": 4540,
      "time_per_iteration": 2.5505895614624023
    },
    {
      "auxiliary_loss_clip": 0.01076303,
      "auxiliary_loss_mlp": 0.01000217,
      "balance_loss_clip": 1.01617897,
      "balance_loss_mlp": 0.99933493,
      "epoch": 0.5460229663921121,
      "flos": 55198399685760.0,
      "grad_norm": 0.7749301543654749,
      "language_loss": 0.57799071,
      "learning_rate": 1.7993633748153833e-06,
      "loss": 0.5987559,
      "num_input_tokens_seen": 97981950,
      "step": 4541,
      "time_per_iteration": 2.912912368774414
    },
    {
      "auxiliary_loss_clip": 0.01165002,
      "auxiliary_loss_mlp": 0.01028674,
      "balance_loss_clip": 1.05021369,
      "balance_loss_mlp": 1.02117836,
      "epoch": 0.5461432092827512,
      "flos": 15413866018560.0,
      "grad_norm": 2.0416428692083866,
      "language_loss": 0.73028022,
      "learning_rate": 1.7985883480108834e-06,
      "loss": 0.75221699,
      "num_input_tokens_seen": 97999585,
      "step": 4542,
      "time_per_iteration": 2.424006700515747
    },
    {
      "auxiliary_loss_clip": 0.01156192,
      "auxiliary_loss_mlp": 0.01030804,
      "balance_loss_clip": 1.04839957,
      "balance_loss_mlp": 1.02288282,
      "epoch": 0.5462634521733902,
      "flos": 24024921287040.0,
      "grad_norm": 1.7501600149856733,
      "language_loss": 0.71994466,
      "learning_rate": 1.797813351760371e-06,
      "loss": 0.74181461,
      "num_input_tokens_seen": 98021290,
      "step": 4543,
      "time_per_iteration": 2.481827735900879
    },
    {
      "auxiliary_loss_clip": 0.01176667,
      "auxiliary_loss_mlp": 0.0102296,
      "balance_loss_clip": 1.052508,
      "balance_loss_mlp": 1.01519907,
      "epoch": 0.5463836950640293,
      "flos": 22820944291200.0,
      "grad_norm": 1.836872192055815,
      "language_loss": 0.77999276,
      "learning_rate": 1.7970383861814116e-06,
      "loss": 0.80198902,
      "num_input_tokens_seen": 98041060,
      "step": 4544,
      "time_per_iteration": 2.439911127090454
    },
    {
      "auxiliary_loss_clip": 0.01160459,
      "auxiliary_loss_mlp": 0.01025248,
      "balance_loss_clip": 1.051036,
      "balance_loss_mlp": 1.01744556,
      "epoch": 0.5465039379546685,
      "flos": 20448290390400.0,
      "grad_norm": 1.8993314336524234,
      "language_loss": 0.74120528,
      "learning_rate": 1.7962634513915684e-06,
      "loss": 0.76306236,
      "num_input_tokens_seen": 98058410,
      "step": 4545,
      "time_per_iteration": 2.43542742729187
    },
    {
      "auxiliary_loss_clip": 0.01173234,
      "auxiliary_loss_mlp": 0.01021611,
      "balance_loss_clip": 1.0507493,
      "balance_loss_mlp": 1.01440763,
      "epoch": 0.5466241808453075,
      "flos": 17343310003200.0,
      "grad_norm": 1.8499934561082187,
      "language_loss": 0.79394221,
      "learning_rate": 1.7954885475083969e-06,
      "loss": 0.81589067,
      "num_input_tokens_seen": 98076080,
      "step": 4546,
      "time_per_iteration": 2.386650800704956
    },
    {
      "auxiliary_loss_clip": 0.01177076,
      "auxiliary_loss_mlp": 0.01030274,
      "balance_loss_clip": 1.0522691,
      "balance_loss_mlp": 1.02276933,
      "epoch": 0.5467444237359466,
      "flos": 21617039122560.0,
      "grad_norm": 2.683068920328305,
      "language_loss": 0.7231003,
      "learning_rate": 1.7947136746494513e-06,
      "loss": 0.74517381,
      "num_input_tokens_seen": 98096995,
      "step": 4547,
      "time_per_iteration": 2.4507923126220703
    },
    {
      "auxiliary_loss_clip": 0.01160442,
      "auxiliary_loss_mlp": 0.01028466,
      "balance_loss_clip": 1.05037403,
      "balance_loss_mlp": 1.02111673,
      "epoch": 0.5468646666265857,
      "flos": 24170467196160.0,
      "grad_norm": 2.0244713422737535,
      "language_loss": 0.87999213,
      "learning_rate": 1.793938832932277e-06,
      "loss": 0.90188122,
      "num_input_tokens_seen": 98115105,
      "step": 4548,
      "time_per_iteration": 2.543168544769287
    },
    {
      "auxiliary_loss_clip": 0.01175335,
      "auxiliary_loss_mlp": 0.01022658,
      "balance_loss_clip": 1.0509218,
      "balance_loss_mlp": 1.01533258,
      "epoch": 0.5469849095172248,
      "flos": 27527001505920.0,
      "grad_norm": 3.9813891389050253,
      "language_loss": 0.70256078,
      "learning_rate": 1.7931640224744185e-06,
      "loss": 0.72454071,
      "num_input_tokens_seen": 98135655,
      "step": 4549,
      "time_per_iteration": 3.3618717193603516
    },
    {
      "auxiliary_loss_clip": 0.0111812,
      "auxiliary_loss_mlp": 0.01024883,
      "balance_loss_clip": 1.04001212,
      "balance_loss_mlp": 1.01751029,
      "epoch": 0.5471051524078638,
      "flos": 27964680727680.0,
      "grad_norm": 1.6906578686327298,
      "language_loss": 0.73609936,
      "learning_rate": 1.7923892433934127e-06,
      "loss": 0.75752944,
      "num_input_tokens_seen": 98156730,
      "step": 4550,
      "time_per_iteration": 2.6062722206115723
    },
    {
      "auxiliary_loss_clip": 0.01149645,
      "auxiliary_loss_mlp": 0.00762469,
      "balance_loss_clip": 1.04947627,
      "balance_loss_mlp": 1.00027156,
      "epoch": 0.547225395298503,
      "flos": 18150510389760.0,
      "grad_norm": 1.7988293326444136,
      "language_loss": 0.78776646,
      "learning_rate": 1.7916144958067939e-06,
      "loss": 0.80688763,
      "num_input_tokens_seen": 98174590,
      "step": 4551,
      "time_per_iteration": 2.458024263381958
    },
    {
      "auxiliary_loss_clip": 0.01162338,
      "auxiliary_loss_mlp": 0.0102254,
      "balance_loss_clip": 1.04953206,
      "balance_loss_mlp": 1.01510155,
      "epoch": 0.5473456381891421,
      "flos": 21361498790400.0,
      "grad_norm": 1.6765500906810593,
      "language_loss": 0.78673708,
      "learning_rate": 1.7908397798320905e-06,
      "loss": 0.80858582,
      "num_input_tokens_seen": 98194325,
      "step": 4552,
      "time_per_iteration": 2.4699718952178955
    },
    {
      "auxiliary_loss_clip": 0.01160948,
      "auxiliary_loss_mlp": 0.00762508,
      "balance_loss_clip": 1.04991162,
      "balance_loss_mlp": 1.00033486,
      "epoch": 0.5474658810797811,
      "flos": 19932145908480.0,
      "grad_norm": 1.8051822545327347,
      "language_loss": 0.74956799,
      "learning_rate": 1.7900650955868265e-06,
      "loss": 0.76880252,
      "num_input_tokens_seen": 98213970,
      "step": 4553,
      "time_per_iteration": 3.324789524078369
    },
    {
      "auxiliary_loss_clip": 0.01160909,
      "auxiliary_loss_mlp": 0.0076161,
      "balance_loss_clip": 1.05158138,
      "balance_loss_mlp": 1.00024807,
      "epoch": 0.5475861239704203,
      "flos": 50476217264640.0,
      "grad_norm": 1.6981592132020786,
      "language_loss": 0.76457,
      "learning_rate": 1.7892904431885202e-06,
      "loss": 0.78379524,
      "num_input_tokens_seen": 98241145,
      "step": 4554,
      "time_per_iteration": 2.726081371307373
    },
    {
      "auxiliary_loss_clip": 0.01118066,
      "auxiliary_loss_mlp": 0.01027191,
      "balance_loss_clip": 1.04255903,
      "balance_loss_mlp": 1.02022946,
      "epoch": 0.5477063668610593,
      "flos": 20705123612160.0,
      "grad_norm": 2.6497862896537736,
      "language_loss": 0.75188208,
      "learning_rate": 1.788515822754686e-06,
      "loss": 0.77333462,
      "num_input_tokens_seen": 98261565,
      "step": 4555,
      "time_per_iteration": 3.524815082550049
    },
    {
      "auxiliary_loss_clip": 0.01133472,
      "auxiliary_loss_mlp": 0.01029459,
      "balance_loss_clip": 1.04474831,
      "balance_loss_mlp": 1.02174652,
      "epoch": 0.5478266097516984,
      "flos": 19609740408960.0,
      "grad_norm": 2.713399239387335,
      "language_loss": 0.78099608,
      "learning_rate": 1.7877412344028335e-06,
      "loss": 0.80262542,
      "num_input_tokens_seen": 98281370,
      "step": 4556,
      "time_per_iteration": 2.532320737838745
    },
    {
      "auxiliary_loss_clip": 0.01162166,
      "auxiliary_loss_mlp": 0.01021516,
      "balance_loss_clip": 1.04914927,
      "balance_loss_mlp": 1.0140655,
      "epoch": 0.5479468526423376,
      "flos": 12896599962240.0,
      "grad_norm": 2.202349931979879,
      "language_loss": 0.77485198,
      "learning_rate": 1.7869666782504668e-06,
      "loss": 0.7966888,
      "num_input_tokens_seen": 98297950,
      "step": 4557,
      "time_per_iteration": 2.463698387145996
    },
    {
      "auxiliary_loss_clip": 0.01132633,
      "auxiliary_loss_mlp": 0.01024082,
      "balance_loss_clip": 1.04283547,
      "balance_loss_mlp": 1.01650059,
      "epoch": 0.5480670955329766,
      "flos": 18588800142720.0,
      "grad_norm": 2.424999626467018,
      "language_loss": 0.68952286,
      "learning_rate": 1.7861921544150867e-06,
      "loss": 0.71109009,
      "num_input_tokens_seen": 98316800,
      "step": 4558,
      "time_per_iteration": 3.2716052532196045
    },
    {
      "auxiliary_loss_clip": 0.01091073,
      "auxiliary_loss_mlp": 0.00761889,
      "balance_loss_clip": 1.04247427,
      "balance_loss_mlp": 1.00020099,
      "epoch": 0.5481873384236157,
      "flos": 15954608338560.0,
      "grad_norm": 1.8777549783472427,
      "language_loss": 0.76459134,
      "learning_rate": 1.7854176630141856e-06,
      "loss": 0.78312099,
      "num_input_tokens_seen": 98333935,
      "step": 4559,
      "time_per_iteration": 2.6099140644073486
    },
    {
      "auxiliary_loss_clip": 0.01179481,
      "auxiliary_loss_mlp": 0.01036978,
      "balance_loss_clip": 1.05334163,
      "balance_loss_mlp": 1.02925944,
      "epoch": 0.5483075813142548,
      "flos": 22783812606720.0,
      "grad_norm": 2.117852096102336,
      "language_loss": 0.84697604,
      "learning_rate": 1.784643204165255e-06,
      "loss": 0.86914062,
      "num_input_tokens_seen": 98353255,
      "step": 4560,
      "time_per_iteration": 2.4737913608551025
    },
    {
      "auxiliary_loss_clip": 0.01155709,
      "auxiliary_loss_mlp": 0.01024309,
      "balance_loss_clip": 1.05070949,
      "balance_loss_mlp": 1.01701951,
      "epoch": 0.5484278242048939,
      "flos": 19317212046720.0,
      "grad_norm": 1.929608792232688,
      "language_loss": 0.77659905,
      "learning_rate": 1.7838687779857783e-06,
      "loss": 0.79839921,
      "num_input_tokens_seen": 98371130,
      "step": 4561,
      "time_per_iteration": 2.4686620235443115
    },
    {
      "auxiliary_loss_clip": 0.01138915,
      "auxiliary_loss_mlp": 0.01026644,
      "balance_loss_clip": 1.04524457,
      "balance_loss_mlp": 1.01859772,
      "epoch": 0.5485480670955329,
      "flos": 22816024128000.0,
      "grad_norm": 2.525157956433148,
      "language_loss": 0.63625896,
      "learning_rate": 1.7830943845932366e-06,
      "loss": 0.65791452,
      "num_input_tokens_seen": 98390455,
      "step": 4562,
      "time_per_iteration": 2.511744976043701
    },
    {
      "auxiliary_loss_clip": 0.01148393,
      "auxiliary_loss_mlp": 0.01027185,
      "balance_loss_clip": 1.04811239,
      "balance_loss_mlp": 1.02007735,
      "epoch": 0.5486683099861721,
      "flos": 22671304231680.0,
      "grad_norm": 1.7376921800225773,
      "language_loss": 0.7511009,
      "learning_rate": 1.7823200241051044e-06,
      "loss": 0.77285671,
      "num_input_tokens_seen": 98409370,
      "step": 4563,
      "time_per_iteration": 2.5301826000213623
    },
    {
      "auxiliary_loss_clip": 0.01174728,
      "auxiliary_loss_mlp": 0.01022712,
      "balance_loss_clip": 1.0509963,
      "balance_loss_mlp": 1.01518416,
      "epoch": 0.5487885528768112,
      "flos": 23149383275520.0,
      "grad_norm": 1.8530807796113584,
      "language_loss": 0.80451554,
      "learning_rate": 1.7815456966388513e-06,
      "loss": 0.82648993,
      "num_input_tokens_seen": 98428465,
      "step": 4564,
      "time_per_iteration": 2.4201343059539795
    },
    {
      "auxiliary_loss_clip": 0.01133188,
      "auxiliary_loss_mlp": 0.01030926,
      "balance_loss_clip": 1.04530263,
      "balance_loss_mlp": 1.02349341,
      "epoch": 0.5489087957674502,
      "flos": 22053928245120.0,
      "grad_norm": 2.242278662268403,
      "language_loss": 0.80838966,
      "learning_rate": 1.780771402311943e-06,
      "loss": 0.8300308,
      "num_input_tokens_seen": 98447300,
      "step": 4565,
      "time_per_iteration": 2.5529279708862305
    },
    {
      "auxiliary_loss_clip": 0.01145576,
      "auxiliary_loss_mlp": 0.01029005,
      "balance_loss_clip": 1.04783845,
      "balance_loss_mlp": 1.02119112,
      "epoch": 0.5490290386580894,
      "flos": 24315977191680.0,
      "grad_norm": 1.9985236507139452,
      "language_loss": 0.78857285,
      "learning_rate": 1.7799971412418374e-06,
      "loss": 0.81031871,
      "num_input_tokens_seen": 98468695,
      "step": 4566,
      "time_per_iteration": 2.515193462371826
    },
    {
      "auxiliary_loss_clip": 0.01135284,
      "auxiliary_loss_mlp": 0.01025986,
      "balance_loss_clip": 1.04826117,
      "balance_loss_mlp": 1.01838374,
      "epoch": 0.5491492815487284,
      "flos": 18294942977280.0,
      "grad_norm": 11.723893213270623,
      "language_loss": 0.73885345,
      "learning_rate": 1.7792229135459918e-06,
      "loss": 0.76046616,
      "num_input_tokens_seen": 98485345,
      "step": 4567,
      "time_per_iteration": 2.4846317768096924
    },
    {
      "auxiliary_loss_clip": 0.01045923,
      "auxiliary_loss_mlp": 0.01020286,
      "balance_loss_clip": 1.03527236,
      "balance_loss_mlp": 1.01878965,
      "epoch": 0.5492695244393675,
      "flos": 64550257050240.0,
      "grad_norm": 0.7975063242764259,
      "language_loss": 0.61580998,
      "learning_rate": 1.7784487193418538e-06,
      "loss": 0.63647211,
      "num_input_tokens_seen": 98543195,
      "step": 4568,
      "time_per_iteration": 3.056511878967285
    },
    {
      "auxiliary_loss_clip": 0.01117544,
      "auxiliary_loss_mlp": 0.01025708,
      "balance_loss_clip": 1.04118085,
      "balance_loss_mlp": 1.01760793,
      "epoch": 0.5493897673300067,
      "flos": 17379579761280.0,
      "grad_norm": 1.977251591286577,
      "language_loss": 0.61347437,
      "learning_rate": 1.7776745587468698e-06,
      "loss": 0.63490689,
      "num_input_tokens_seen": 98560620,
      "step": 4569,
      "time_per_iteration": 2.5322492122650146
    },
    {
      "auxiliary_loss_clip": 0.01172484,
      "auxiliary_loss_mlp": 0.01029719,
      "balance_loss_clip": 1.04877591,
      "balance_loss_mlp": 1.02223301,
      "epoch": 0.5495100102206457,
      "flos": 19901765980800.0,
      "grad_norm": 2.280690154085733,
      "language_loss": 0.81421161,
      "learning_rate": 1.7769004318784776e-06,
      "loss": 0.83623374,
      "num_input_tokens_seen": 98578265,
      "step": 4570,
      "time_per_iteration": 2.4282124042510986
    },
    {
      "auxiliary_loss_clip": 0.01162038,
      "auxiliary_loss_mlp": 0.01022539,
      "balance_loss_clip": 1.04976487,
      "balance_loss_mlp": 1.01530933,
      "epoch": 0.5496302531112848,
      "flos": 16727190992640.0,
      "grad_norm": 2.3230471837682507,
      "language_loss": 0.80471182,
      "learning_rate": 1.776126338854113e-06,
      "loss": 0.82655752,
      "num_input_tokens_seen": 98596055,
      "step": 4571,
      "time_per_iteration": 2.436657667160034
    },
    {
      "auxiliary_loss_clip": 0.01158951,
      "auxiliary_loss_mlp": 0.01026041,
      "balance_loss_clip": 1.05287945,
      "balance_loss_mlp": 1.01882625,
      "epoch": 0.5497504960019239,
      "flos": 24572343536640.0,
      "grad_norm": 1.6746441438583883,
      "language_loss": 0.84460211,
      "learning_rate": 1.7753522797912044e-06,
      "loss": 0.86645204,
      "num_input_tokens_seen": 98616140,
      "step": 4572,
      "time_per_iteration": 2.495299816131592
    },
    {
      "auxiliary_loss_clip": 0.01152074,
      "auxiliary_loss_mlp": 0.01024062,
      "balance_loss_clip": 1.04634762,
      "balance_loss_mlp": 1.01652193,
      "epoch": 0.549870738892563,
      "flos": 15450494912640.0,
      "grad_norm": 2.2534041290342057,
      "language_loss": 0.69912559,
      "learning_rate": 1.7745782548071765e-06,
      "loss": 0.72088695,
      "num_input_tokens_seen": 98633035,
      "step": 4573,
      "time_per_iteration": 2.463361978530884
    },
    {
      "auxiliary_loss_clip": 0.01131782,
      "auxiliary_loss_mlp": 0.01031766,
      "balance_loss_clip": 1.0534482,
      "balance_loss_mlp": 1.02428865,
      "epoch": 0.549990981783202,
      "flos": 21069114082560.0,
      "grad_norm": 1.6027613517811543,
      "language_loss": 0.74147856,
      "learning_rate": 1.7738042640194482e-06,
      "loss": 0.76311398,
      "num_input_tokens_seen": 98652700,
      "step": 4574,
      "time_per_iteration": 2.513716220855713
    },
    {
      "auxiliary_loss_clip": 0.01174342,
      "auxiliary_loss_mlp": 0.0102484,
      "balance_loss_clip": 1.05057776,
      "balance_loss_mlp": 1.01723456,
      "epoch": 0.5501112246738411,
      "flos": 21395901041280.0,
      "grad_norm": 4.283970291102364,
      "language_loss": 0.70605361,
      "learning_rate": 1.7730303075454335e-06,
      "loss": 0.72804546,
      "num_input_tokens_seen": 98671590,
      "step": 4575,
      "time_per_iteration": 2.417705535888672
    },
    {
      "auxiliary_loss_clip": 0.0113337,
      "auxiliary_loss_mlp": 0.01026899,
      "balance_loss_clip": 1.04462266,
      "balance_loss_mlp": 1.01873326,
      "epoch": 0.5502314675644803,
      "flos": 17456931699840.0,
      "grad_norm": 1.8406541595085508,
      "language_loss": 0.85018128,
      "learning_rate": 1.7722563855025402e-06,
      "loss": 0.87178397,
      "num_input_tokens_seen": 98689620,
      "step": 4576,
      "time_per_iteration": 3.239882707595825
    },
    {
      "auxiliary_loss_clip": 0.01146196,
      "auxiliary_loss_mlp": 0.01020372,
      "balance_loss_clip": 1.0441941,
      "balance_loss_mlp": 1.01271927,
      "epoch": 0.5503517104551193,
      "flos": 24310410583680.0,
      "grad_norm": 2.5152415230034992,
      "language_loss": 0.70804691,
      "learning_rate": 1.7714824980081721e-06,
      "loss": 0.72971261,
      "num_input_tokens_seen": 98708915,
      "step": 4577,
      "time_per_iteration": 2.530142068862915
    },
    {
      "auxiliary_loss_clip": 0.01158352,
      "auxiliary_loss_mlp": 0.01021886,
      "balance_loss_clip": 1.05130887,
      "balance_loss_mlp": 1.01495147,
      "epoch": 0.5504719533457584,
      "flos": 22419427086720.0,
      "grad_norm": 1.6876752239705388,
      "language_loss": 0.73926985,
      "learning_rate": 1.7707086451797276e-06,
      "loss": 0.76107216,
      "num_input_tokens_seen": 98729790,
      "step": 4578,
      "time_per_iteration": 2.4867489337921143
    },
    {
      "auxiliary_loss_clip": 0.01041873,
      "auxiliary_loss_mlp": 0.0100182,
      "balance_loss_clip": 1.01362205,
      "balance_loss_mlp": 1.00084198,
      "epoch": 0.5505921962363975,
      "flos": 67294155968640.0,
      "grad_norm": 0.6992956562704757,
      "language_loss": 0.52346671,
      "learning_rate": 1.7699348271345993e-06,
      "loss": 0.54390365,
      "num_input_tokens_seen": 98792415,
      "step": 4579,
      "time_per_iteration": 3.89481520652771
    },
    {
      "auxiliary_loss_clip": 0.01037737,
      "auxiliary_loss_mlp": 0.01003438,
      "balance_loss_clip": 1.01361394,
      "balance_loss_mlp": 1.00232971,
      "epoch": 0.5507124391270366,
      "flos": 45685125578880.0,
      "grad_norm": 3.1794472675558203,
      "language_loss": 0.54430765,
      "learning_rate": 1.7691610439901753e-06,
      "loss": 0.56471938,
      "num_input_tokens_seen": 98855350,
      "step": 4580,
      "time_per_iteration": 3.1726150512695312
    },
    {
      "auxiliary_loss_clip": 0.01163625,
      "auxiliary_loss_mlp": 0.01026128,
      "balance_loss_clip": 1.05114615,
      "balance_loss_mlp": 1.01914787,
      "epoch": 0.5508326820176757,
      "flos": 22273845264000.0,
      "grad_norm": 1.795442319083522,
      "language_loss": 0.75315332,
      "learning_rate": 1.7683872958638367e-06,
      "loss": 0.77505088,
      "num_input_tokens_seen": 98874230,
      "step": 4581,
      "time_per_iteration": 2.5027122497558594
    },
    {
      "auxiliary_loss_clip": 0.01142542,
      "auxiliary_loss_mlp": 0.01025634,
      "balance_loss_clip": 1.04526246,
      "balance_loss_mlp": 1.01792121,
      "epoch": 0.5509529249083148,
      "flos": 20012442762240.0,
      "grad_norm": 2.3341089321278923,
      "language_loss": 0.84708309,
      "learning_rate": 1.7676135828729614e-06,
      "loss": 0.86876488,
      "num_input_tokens_seen": 98893940,
      "step": 4582,
      "time_per_iteration": 3.3754777908325195
    },
    {
      "auxiliary_loss_clip": 0.0116154,
      "auxiliary_loss_mlp": 0.01026191,
      "balance_loss_clip": 1.05121076,
      "balance_loss_mlp": 1.01863003,
      "epoch": 0.5510731677989539,
      "flos": 21834801325440.0,
      "grad_norm": 2.546083563910847,
      "language_loss": 0.82716769,
      "learning_rate": 1.7668399051349205e-06,
      "loss": 0.84904498,
      "num_input_tokens_seen": 98913620,
      "step": 4583,
      "time_per_iteration": 2.481414794921875
    },
    {
      "auxiliary_loss_clip": 0.01129364,
      "auxiliary_loss_mlp": 0.01021433,
      "balance_loss_clip": 1.04554474,
      "balance_loss_mlp": 1.01402414,
      "epoch": 0.5511934106895929,
      "flos": 21467901853440.0,
      "grad_norm": 1.9774319854423994,
      "language_loss": 0.83366269,
      "learning_rate": 1.766066262767081e-06,
      "loss": 0.85517067,
      "num_input_tokens_seen": 98931460,
      "step": 4584,
      "time_per_iteration": 2.5133705139160156
    },
    {
      "auxiliary_loss_clip": 0.01143608,
      "auxiliary_loss_mlp": 0.01023765,
      "balance_loss_clip": 1.05011714,
      "balance_loss_mlp": 1.01636803,
      "epoch": 0.5513136535802321,
      "flos": 21068934514560.0,
      "grad_norm": 2.11975912492017,
      "language_loss": 0.77296448,
      "learning_rate": 1.765292655886803e-06,
      "loss": 0.79463822,
      "num_input_tokens_seen": 98950105,
      "step": 4585,
      "time_per_iteration": 3.266927719116211
    },
    {
      "auxiliary_loss_clip": 0.01137984,
      "auxiliary_loss_mlp": 0.01027223,
      "balance_loss_clip": 1.04614711,
      "balance_loss_mlp": 1.01968932,
      "epoch": 0.5514338964708712,
      "flos": 27815004754560.0,
      "grad_norm": 2.316459250751171,
      "language_loss": 0.70491791,
      "learning_rate": 1.764519084611443e-06,
      "loss": 0.72657001,
      "num_input_tokens_seen": 98970560,
      "step": 4586,
      "time_per_iteration": 2.554762840270996
    },
    {
      "auxiliary_loss_clip": 0.01144834,
      "auxiliary_loss_mlp": 0.01025652,
      "balance_loss_clip": 1.04485667,
      "balance_loss_mlp": 1.01716995,
      "epoch": 0.5515541393615102,
      "flos": 21908525990400.0,
      "grad_norm": 1.9169869126700863,
      "language_loss": 0.77876776,
      "learning_rate": 1.7637455490583505e-06,
      "loss": 0.80047262,
      "num_input_tokens_seen": 98989885,
      "step": 4587,
      "time_per_iteration": 2.5103631019592285
    },
    {
      "auxiliary_loss_clip": 0.01160474,
      "auxiliary_loss_mlp": 0.0102577,
      "balance_loss_clip": 1.05094504,
      "balance_loss_mlp": 1.01883531,
      "epoch": 0.5516743822521494,
      "flos": 20485422074880.0,
      "grad_norm": 2.1088689234778375,
      "language_loss": 0.77264249,
      "learning_rate": 1.7629720493448701e-06,
      "loss": 0.794505,
      "num_input_tokens_seen": 99007180,
      "step": 4588,
      "time_per_iteration": 2.4755096435546875
    },
    {
      "auxiliary_loss_clip": 0.01155645,
      "auxiliary_loss_mlp": 0.01029743,
      "balance_loss_clip": 1.04966128,
      "balance_loss_mlp": 1.02235818,
      "epoch": 0.5517946251427884,
      "flos": 14940383915520.0,
      "grad_norm": 2.221206655272309,
      "language_loss": 0.85483432,
      "learning_rate": 1.7621985855883418e-06,
      "loss": 0.87668824,
      "num_input_tokens_seen": 99023880,
      "step": 4589,
      "time_per_iteration": 2.4562203884124756
    },
    {
      "auxiliary_loss_clip": 0.01141417,
      "auxiliary_loss_mlp": 0.01022181,
      "balance_loss_clip": 1.0477711,
      "balance_loss_mlp": 1.0145998,
      "epoch": 0.5519148680334275,
      "flos": 18404865573120.0,
      "grad_norm": 1.9633910655858247,
      "language_loss": 0.72875792,
      "learning_rate": 1.7614251579060983e-06,
      "loss": 0.75039393,
      "num_input_tokens_seen": 99042475,
      "step": 4590,
      "time_per_iteration": 2.4668638706207275
    },
    {
      "auxiliary_loss_clip": 0.01135538,
      "auxiliary_loss_mlp": 0.01025429,
      "balance_loss_clip": 1.04711854,
      "balance_loss_mlp": 1.01762688,
      "epoch": 0.5520351109240667,
      "flos": 25113337251840.0,
      "grad_norm": 1.7528165984326978,
      "language_loss": 0.84657156,
      "learning_rate": 1.76065176641547e-06,
      "loss": 0.86818123,
      "num_input_tokens_seen": 99065185,
      "step": 4591,
      "time_per_iteration": 2.5800955295562744
    },
    {
      "auxiliary_loss_clip": 0.01159643,
      "auxiliary_loss_mlp": 0.01022772,
      "balance_loss_clip": 1.04646087,
      "balance_loss_mlp": 1.01492786,
      "epoch": 0.5521553538147057,
      "flos": 21069545045760.0,
      "grad_norm": 1.8201871631045887,
      "language_loss": 0.78194022,
      "learning_rate": 1.759878411233777e-06,
      "loss": 0.80376434,
      "num_input_tokens_seen": 99083645,
      "step": 4592,
      "time_per_iteration": 2.4513750076293945
    },
    {
      "auxiliary_loss_clip": 0.01158278,
      "auxiliary_loss_mlp": 0.01023502,
      "balance_loss_clip": 1.04787374,
      "balance_loss_mlp": 1.0156281,
      "epoch": 0.5522755967053448,
      "flos": 18879999701760.0,
      "grad_norm": 2.23111348106347,
      "language_loss": 0.75911099,
      "learning_rate": 1.7591050924783388e-06,
      "loss": 0.78092873,
      "num_input_tokens_seen": 99100835,
      "step": 4593,
      "time_per_iteration": 2.435822010040283
    },
    {
      "auxiliary_loss_clip": 0.01035245,
      "auxiliary_loss_mlp": 0.0100228,
      "balance_loss_clip": 1.01448512,
      "balance_loss_mlp": 1.0012188,
      "epoch": 0.5523958395959839,
      "flos": 64675622494080.0,
      "grad_norm": 0.8409316010093588,
      "language_loss": 0.57956284,
      "learning_rate": 1.7583318102664661e-06,
      "loss": 0.59993804,
      "num_input_tokens_seen": 99168400,
      "step": 4594,
      "time_per_iteration": 3.162726879119873
    },
    {
      "auxiliary_loss_clip": 0.01163231,
      "auxiliary_loss_mlp": 0.01027378,
      "balance_loss_clip": 1.04684472,
      "balance_loss_mlp": 1.01982021,
      "epoch": 0.552516082486623,
      "flos": 10889732211840.0,
      "grad_norm": 2.0834189121283617,
      "language_loss": 0.7913065,
      "learning_rate": 1.757558564715466e-06,
      "loss": 0.81321263,
      "num_input_tokens_seen": 99186475,
      "step": 4595,
      "time_per_iteration": 2.448741912841797
    },
    {
      "auxiliary_loss_clip": 0.01161826,
      "auxiliary_loss_mlp": 0.01026472,
      "balance_loss_clip": 1.04710138,
      "balance_loss_mlp": 1.01874757,
      "epoch": 0.552636325377262,
      "flos": 22199797376640.0,
      "grad_norm": 2.5948709360976223,
      "language_loss": 0.74100387,
      "learning_rate": 1.7567853559426386e-06,
      "loss": 0.76288682,
      "num_input_tokens_seen": 99203525,
      "step": 4596,
      "time_per_iteration": 2.4705755710601807
    },
    {
      "auxiliary_loss_clip": 0.01162587,
      "auxiliary_loss_mlp": 0.0102786,
      "balance_loss_clip": 1.04906225,
      "balance_loss_mlp": 1.02059984,
      "epoch": 0.5527565682679012,
      "flos": 23988184652160.0,
      "grad_norm": 1.954694638151164,
      "language_loss": 0.75518489,
      "learning_rate": 1.7560121840652797e-06,
      "loss": 0.77708936,
      "num_input_tokens_seen": 99222910,
      "step": 4597,
      "time_per_iteration": 2.4726083278656006
    },
    {
      "auxiliary_loss_clip": 0.01123301,
      "auxiliary_loss_mlp": 0.01021177,
      "balance_loss_clip": 1.04606032,
      "balance_loss_mlp": 1.01351833,
      "epoch": 0.5528768111585403,
      "flos": 19719267955200.0,
      "grad_norm": 1.7970635243716697,
      "language_loss": 0.69439077,
      "learning_rate": 1.7552390492006782e-06,
      "loss": 0.71583557,
      "num_input_tokens_seen": 99241230,
      "step": 4598,
      "time_per_iteration": 2.5285274982452393
    },
    {
      "auxiliary_loss_clip": 0.0112505,
      "auxiliary_loss_mlp": 0.00761991,
      "balance_loss_clip": 1.04282832,
      "balance_loss_mlp": 1.00033128,
      "epoch": 0.5529970540491793,
      "flos": 26215975002240.0,
      "grad_norm": 2.841268767851113,
      "language_loss": 0.65259719,
      "learning_rate": 1.7544659514661184e-06,
      "loss": 0.67146754,
      "num_input_tokens_seen": 99264320,
      "step": 4599,
      "time_per_iteration": 2.6457877159118652
    },
    {
      "auxiliary_loss_clip": 0.01142168,
      "auxiliary_loss_mlp": 0.01022629,
      "balance_loss_clip": 1.04485798,
      "balance_loss_mlp": 1.01534522,
      "epoch": 0.5531172969398185,
      "flos": 24425971614720.0,
      "grad_norm": 2.0676723586161323,
      "language_loss": 0.79505628,
      "learning_rate": 1.7536928909788786e-06,
      "loss": 0.81670427,
      "num_input_tokens_seen": 99283625,
      "step": 4600,
      "time_per_iteration": 2.519928455352783
    },
    {
      "auxiliary_loss_clip": 0.01038012,
      "auxiliary_loss_mlp": 0.01002878,
      "balance_loss_clip": 1.01439595,
      "balance_loss_mlp": 1.00188303,
      "epoch": 0.5532375398304575,
      "flos": 64907316195840.0,
      "grad_norm": 0.8867001722156724,
      "language_loss": 0.61995786,
      "learning_rate": 1.752919867856231e-06,
      "loss": 0.64036679,
      "num_input_tokens_seen": 99335270,
      "step": 4601,
      "time_per_iteration": 2.9552173614501953
    },
    {
      "auxiliary_loss_clip": 0.01137935,
      "auxiliary_loss_mlp": 0.01024868,
      "balance_loss_clip": 1.04410577,
      "balance_loss_mlp": 1.0175426,
      "epoch": 0.5533577827210966,
      "flos": 19683105937920.0,
      "grad_norm": 1.7535870211720843,
      "language_loss": 0.78933799,
      "learning_rate": 1.7521468822154436e-06,
      "loss": 0.81096601,
      "num_input_tokens_seen": 99354185,
      "step": 4602,
      "time_per_iteration": 2.4831244945526123
    },
    {
      "auxiliary_loss_clip": 0.01141231,
      "auxiliary_loss_mlp": 0.01026506,
      "balance_loss_clip": 1.04819393,
      "balance_loss_mlp": 1.01973748,
      "epoch": 0.5534780256117358,
      "flos": 32306496076800.0,
      "grad_norm": 1.82088858066226,
      "language_loss": 0.74731058,
      "learning_rate": 1.751373934173777e-06,
      "loss": 0.76898801,
      "num_input_tokens_seen": 99376930,
      "step": 4603,
      "time_per_iteration": 3.4246315956115723
    },
    {
      "auxiliary_loss_clip": 0.01175575,
      "auxiliary_loss_mlp": 0.01025871,
      "balance_loss_clip": 1.04968834,
      "balance_loss_mlp": 1.01816392,
      "epoch": 0.5535982685023748,
      "flos": 23222425582080.0,
      "grad_norm": 1.5675869527216764,
      "language_loss": 0.73273396,
      "learning_rate": 1.750601023848487e-06,
      "loss": 0.75474846,
      "num_input_tokens_seen": 99397655,
      "step": 4604,
      "time_per_iteration": 2.4424405097961426
    },
    {
      "auxiliary_loss_clip": 0.01174596,
      "auxiliary_loss_mlp": 0.00761752,
      "balance_loss_clip": 1.05228043,
      "balance_loss_mlp": 1.00026774,
      "epoch": 0.5537185113930139,
      "flos": 24352534258560.0,
      "grad_norm": 2.1650850015714878,
      "language_loss": 0.73828447,
      "learning_rate": 1.749828151356823e-06,
      "loss": 0.75764793,
      "num_input_tokens_seen": 99417850,
      "step": 4605,
      "time_per_iteration": 2.438225269317627
    },
    {
      "auxiliary_loss_clip": 0.0114486,
      "auxiliary_loss_mlp": 0.01029615,
      "balance_loss_clip": 1.04667068,
      "balance_loss_mlp": 1.02261186,
      "epoch": 0.553838754283653,
      "flos": 23549068886400.0,
      "grad_norm": 1.6494187455371543,
      "language_loss": 0.75877815,
      "learning_rate": 1.7490553168160297e-06,
      "loss": 0.78052294,
      "num_input_tokens_seen": 99438920,
      "step": 4606,
      "time_per_iteration": 3.3813464641571045
    },
    {
      "auxiliary_loss_clip": 0.01144062,
      "auxiliary_loss_mlp": 0.0102847,
      "balance_loss_clip": 1.0461936,
      "balance_loss_mlp": 1.02014923,
      "epoch": 0.5539589971742921,
      "flos": 17275044205440.0,
      "grad_norm": 1.8498956039126726,
      "language_loss": 0.76338255,
      "learning_rate": 1.748282520343345e-06,
      "loss": 0.78510791,
      "num_input_tokens_seen": 99457950,
      "step": 4607,
      "time_per_iteration": 2.4512076377868652
    },
    {
      "auxiliary_loss_clip": 0.01169421,
      "auxiliary_loss_mlp": 0.01022544,
      "balance_loss_clip": 1.05077076,
      "balance_loss_mlp": 1.01465845,
      "epoch": 0.5540792400649311,
      "flos": 27564169104000.0,
      "grad_norm": 1.968105000801093,
      "language_loss": 0.78958642,
      "learning_rate": 1.7475097620560023e-06,
      "loss": 0.81150603,
      "num_input_tokens_seen": 99478015,
      "step": 4608,
      "time_per_iteration": 3.328918933868408
    },
    {
      "auxiliary_loss_clip": 0.01174592,
      "auxiliary_loss_mlp": 0.01022635,
      "balance_loss_clip": 1.05170345,
      "balance_loss_mlp": 1.01548803,
      "epoch": 0.5541994829555702,
      "flos": 23878657105920.0,
      "grad_norm": 1.8116425445954205,
      "language_loss": 0.71257412,
      "learning_rate": 1.746737042071228e-06,
      "loss": 0.73454636,
      "num_input_tokens_seen": 99496520,
      "step": 4609,
      "time_per_iteration": 2.435483694076538
    },
    {
      "auxiliary_loss_clip": 0.01142669,
      "auxiliary_loss_mlp": 0.01022868,
      "balance_loss_clip": 1.0479964,
      "balance_loss_mlp": 1.01517344,
      "epoch": 0.5543197258462094,
      "flos": 20115721342080.0,
      "grad_norm": 5.749539878932833,
      "language_loss": 0.79233378,
      "learning_rate": 1.7459643605062424e-06,
      "loss": 0.81398916,
      "num_input_tokens_seen": 99513780,
      "step": 4610,
      "time_per_iteration": 2.4782073497772217
    },
    {
      "auxiliary_loss_clip": 0.01115797,
      "auxiliary_loss_mlp": 0.01022765,
      "balance_loss_clip": 1.0455904,
      "balance_loss_mlp": 1.01502824,
      "epoch": 0.5544399687368484,
      "flos": 20916565021440.0,
      "grad_norm": 1.6165793758894886,
      "language_loss": 0.80664867,
      "learning_rate": 1.745191717478262e-06,
      "loss": 0.82803428,
      "num_input_tokens_seen": 99532360,
      "step": 4611,
      "time_per_iteration": 2.5607051849365234
    },
    {
      "auxiliary_loss_clip": 0.01143397,
      "auxiliary_loss_mlp": 0.01031002,
      "balance_loss_clip": 1.0495584,
      "balance_loss_mlp": 1.02327096,
      "epoch": 0.5545602116274875,
      "flos": 25518661297920.0,
      "grad_norm": 1.9027095472513167,
      "language_loss": 0.79611576,
      "learning_rate": 1.7444191131044948e-06,
      "loss": 0.81785971,
      "num_input_tokens_seen": 99552635,
      "step": 4612,
      "time_per_iteration": 3.260999917984009
    },
    {
      "auxiliary_loss_clip": 0.01146826,
      "auxiliary_loss_mlp": 0.01028122,
      "balance_loss_clip": 1.04976153,
      "balance_loss_mlp": 1.01980758,
      "epoch": 0.5546804545181266,
      "flos": 20995568985600.0,
      "grad_norm": 1.7553601877609761,
      "language_loss": 0.73065877,
      "learning_rate": 1.7436465475021456e-06,
      "loss": 0.75240827,
      "num_input_tokens_seen": 99572685,
      "step": 4613,
      "time_per_iteration": 2.50041127204895
    },
    {
      "auxiliary_loss_clip": 0.01123572,
      "auxiliary_loss_mlp": 0.01023986,
      "balance_loss_clip": 1.04623449,
      "balance_loss_mlp": 1.01608241,
      "epoch": 0.5548006974087657,
      "flos": 26833638297600.0,
      "grad_norm": 1.8490591092707873,
      "language_loss": 0.71849304,
      "learning_rate": 1.7428740207884111e-06,
      "loss": 0.73996866,
      "num_input_tokens_seen": 99593565,
      "step": 4614,
      "time_per_iteration": 2.595573663711548
    },
    {
      "auxiliary_loss_clip": 0.01119956,
      "auxiliary_loss_mlp": 0.01026869,
      "balance_loss_clip": 1.04569817,
      "balance_loss_mlp": 1.01921582,
      "epoch": 0.5549209402994048,
      "flos": 33656414031360.0,
      "grad_norm": 1.7744247723909636,
      "language_loss": 0.61129081,
      "learning_rate": 1.7421015330804833e-06,
      "loss": 0.63275909,
      "num_input_tokens_seen": 99613485,
      "step": 4615,
      "time_per_iteration": 2.650078296661377
    },
    {
      "auxiliary_loss_clip": 0.01174529,
      "auxiliary_loss_mlp": 0.01027293,
      "balance_loss_clip": 1.0511719,
      "balance_loss_mlp": 1.01981258,
      "epoch": 0.5550411831900439,
      "flos": 23769524609280.0,
      "grad_norm": 1.7454185104715776,
      "language_loss": 0.72400331,
      "learning_rate": 1.7413290844955475e-06,
      "loss": 0.74602151,
      "num_input_tokens_seen": 99633515,
      "step": 4616,
      "time_per_iteration": 2.481126070022583
    },
    {
      "auxiliary_loss_clip": 0.01155513,
      "auxiliary_loss_mlp": 0.01027665,
      "balance_loss_clip": 1.05171251,
      "balance_loss_mlp": 1.02044129,
      "epoch": 0.555161426080683,
      "flos": 21651189978240.0,
      "grad_norm": 1.8683043543788287,
      "language_loss": 0.78068757,
      "learning_rate": 1.7405566751507843e-06,
      "loss": 0.80251938,
      "num_input_tokens_seen": 99651560,
      "step": 4617,
      "time_per_iteration": 2.460554838180542
    },
    {
      "auxiliary_loss_clip": 0.01131125,
      "auxiliary_loss_mlp": 0.01032996,
      "balance_loss_clip": 1.04484153,
      "balance_loss_mlp": 1.02599287,
      "epoch": 0.555281668971322,
      "flos": 49563116605440.0,
      "grad_norm": 1.451190297533999,
      "language_loss": 0.67608857,
      "learning_rate": 1.7397843051633668e-06,
      "loss": 0.69772977,
      "num_input_tokens_seen": 99674255,
      "step": 4618,
      "time_per_iteration": 2.8118202686309814
    },
    {
      "auxiliary_loss_clip": 0.01155112,
      "auxiliary_loss_mlp": 0.01027301,
      "balance_loss_clip": 1.04858947,
      "balance_loss_mlp": 1.01961231,
      "epoch": 0.5554019118619612,
      "flos": 20741608851840.0,
      "grad_norm": 1.6997013659868998,
      "language_loss": 0.71361554,
      "learning_rate": 1.739011974650464e-06,
      "loss": 0.73543972,
      "num_input_tokens_seen": 99693585,
      "step": 4619,
      "time_per_iteration": 2.458878517150879
    },
    {
      "auxiliary_loss_clip": 0.01122592,
      "auxiliary_loss_mlp": 0.01027172,
      "balance_loss_clip": 1.04540467,
      "balance_loss_mlp": 1.01944089,
      "epoch": 0.5555221547526003,
      "flos": 25483217552640.0,
      "grad_norm": 3.1959499183109434,
      "language_loss": 0.76647395,
      "learning_rate": 1.7382396837292365e-06,
      "loss": 0.78797162,
      "num_input_tokens_seen": 99714045,
      "step": 4620,
      "time_per_iteration": 2.6123523712158203
    },
    {
      "auxiliary_loss_clip": 0.01175055,
      "auxiliary_loss_mlp": 0.01022097,
      "balance_loss_clip": 1.05240893,
      "balance_loss_mlp": 1.01437187,
      "epoch": 0.5556423976432393,
      "flos": 21762513204480.0,
      "grad_norm": 1.8926166248520258,
      "language_loss": 0.73393691,
      "learning_rate": 1.737467432516841e-06,
      "loss": 0.75590849,
      "num_input_tokens_seen": 99734145,
      "step": 4621,
      "time_per_iteration": 2.434607982635498
    },
    {
      "auxiliary_loss_clip": 0.01144136,
      "auxiliary_loss_mlp": 0.01024403,
      "balance_loss_clip": 1.04441214,
      "balance_loss_mlp": 1.0169735,
      "epoch": 0.5557626405338785,
      "flos": 24900171989760.0,
      "grad_norm": 2.30868781223006,
      "language_loss": 0.74419034,
      "learning_rate": 1.7366952211304274e-06,
      "loss": 0.7658757,
      "num_input_tokens_seen": 99751990,
      "step": 4622,
      "time_per_iteration": 2.510425329208374
    },
    {
      "auxiliary_loss_clip": 0.01139126,
      "auxiliary_loss_mlp": 0.01025484,
      "balance_loss_clip": 1.04648924,
      "balance_loss_mlp": 1.01780653,
      "epoch": 0.5558828834245175,
      "flos": 18697501676160.0,
      "grad_norm": 2.0231673981919944,
      "language_loss": 0.83958733,
      "learning_rate": 1.735923049687139e-06,
      "loss": 0.86123341,
      "num_input_tokens_seen": 99768565,
      "step": 4623,
      "time_per_iteration": 2.466840982437134
    },
    {
      "auxiliary_loss_clip": 0.01141118,
      "auxiliary_loss_mlp": 0.01025308,
      "balance_loss_clip": 1.04664481,
      "balance_loss_mlp": 1.01798511,
      "epoch": 0.5560031263151566,
      "flos": 27272179445760.0,
      "grad_norm": 1.4503057712483887,
      "language_loss": 0.73727715,
      "learning_rate": 1.7351509183041144e-06,
      "loss": 0.75894141,
      "num_input_tokens_seen": 99788895,
      "step": 4624,
      "time_per_iteration": 2.5360922813415527
    },
    {
      "auxiliary_loss_clip": 0.01176778,
      "auxiliary_loss_mlp": 0.01023867,
      "balance_loss_clip": 1.05155659,
      "balance_loss_mlp": 1.01664829,
      "epoch": 0.5561233692057957,
      "flos": 23403738458880.0,
      "grad_norm": 1.6291736817877638,
      "language_loss": 0.71514523,
      "learning_rate": 1.7343788270984852e-06,
      "loss": 0.73715168,
      "num_input_tokens_seen": 99808035,
      "step": 4625,
      "time_per_iteration": 2.441911458969116
    },
    {
      "auxiliary_loss_clip": 0.01143773,
      "auxiliary_loss_mlp": 0.01023015,
      "balance_loss_clip": 1.0490911,
      "balance_loss_mlp": 1.01510882,
      "epoch": 0.5562436120964348,
      "flos": 37670867804160.0,
      "grad_norm": 1.8309422499479109,
      "language_loss": 0.74689662,
      "learning_rate": 1.7336067761873764e-06,
      "loss": 0.76856446,
      "num_input_tokens_seen": 99830460,
      "step": 4626,
      "time_per_iteration": 2.6328115463256836
    },
    {
      "auxiliary_loss_clip": 0.01169926,
      "auxiliary_loss_mlp": 0.01029226,
      "balance_loss_clip": 1.05162418,
      "balance_loss_mlp": 1.02109575,
      "epoch": 0.5563638549870739,
      "flos": 25155245445120.0,
      "grad_norm": 1.9123073967380806,
      "language_loss": 0.76350218,
      "learning_rate": 1.7328347656879076e-06,
      "loss": 0.78549373,
      "num_input_tokens_seen": 99850320,
      "step": 4627,
      "time_per_iteration": 2.504725694656372
    },
    {
      "auxiliary_loss_clip": 0.01132678,
      "auxiliary_loss_mlp": 0.01021946,
      "balance_loss_clip": 1.04657364,
      "balance_loss_mlp": 1.01391721,
      "epoch": 0.556484097877713,
      "flos": 13581810783360.0,
      "grad_norm": 2.270634774073174,
      "language_loss": 0.67934072,
      "learning_rate": 1.7320627957171927e-06,
      "loss": 0.70088696,
      "num_input_tokens_seen": 99864980,
      "step": 4628,
      "time_per_iteration": 2.4655189514160156
    },
    {
      "auxiliary_loss_clip": 0.01176321,
      "auxiliary_loss_mlp": 0.01024437,
      "balance_loss_clip": 1.05267274,
      "balance_loss_mlp": 1.01743913,
      "epoch": 0.5566043407683521,
      "flos": 24681368292480.0,
      "grad_norm": 4.885454474018997,
      "language_loss": 0.8158983,
      "learning_rate": 1.7312908663923382e-06,
      "loss": 0.83790582,
      "num_input_tokens_seen": 99881155,
      "step": 4629,
      "time_per_iteration": 2.449221134185791
    },
    {
      "auxiliary_loss_clip": 0.01155254,
      "auxiliary_loss_mlp": 0.0102506,
      "balance_loss_clip": 1.04774356,
      "balance_loss_mlp": 1.01722527,
      "epoch": 0.5567245836589911,
      "flos": 20588161950720.0,
      "grad_norm": 2.6351390872886213,
      "language_loss": 0.67132974,
      "learning_rate": 1.7305189778304463e-06,
      "loss": 0.69313288,
      "num_input_tokens_seen": 99899330,
      "step": 4630,
      "time_per_iteration": 3.199492931365967
    },
    {
      "auxiliary_loss_clip": 0.01149082,
      "auxiliary_loss_mlp": 0.01028249,
      "balance_loss_clip": 1.05212951,
      "balance_loss_mlp": 1.02062833,
      "epoch": 0.5568448265496303,
      "flos": 20704189858560.0,
      "grad_norm": 1.7926391479239319,
      "language_loss": 0.79410577,
      "learning_rate": 1.729747130148611e-06,
      "loss": 0.81587911,
      "num_input_tokens_seen": 99918525,
      "step": 4631,
      "time_per_iteration": 2.484035015106201
    },
    {
      "auxiliary_loss_clip": 0.01138729,
      "auxiliary_loss_mlp": 0.01026581,
      "balance_loss_clip": 1.0478282,
      "balance_loss_mlp": 1.01826048,
      "epoch": 0.5569650694402694,
      "flos": 25302910256640.0,
      "grad_norm": 1.8188745188738866,
      "language_loss": 0.76582205,
      "learning_rate": 1.7289753234639208e-06,
      "loss": 0.78747523,
      "num_input_tokens_seen": 99937500,
      "step": 4632,
      "time_per_iteration": 2.560176372528076
    },
    {
      "auxiliary_loss_clip": 0.01167092,
      "auxiliary_loss_mlp": 0.01026494,
      "balance_loss_clip": 1.05225813,
      "balance_loss_mlp": 1.01866245,
      "epoch": 0.5570853123309084,
      "flos": 19712623939200.0,
      "grad_norm": 3.534945863279641,
      "language_loss": 0.76477939,
      "learning_rate": 1.7282035578934592e-06,
      "loss": 0.78671527,
      "num_input_tokens_seen": 99955665,
      "step": 4633,
      "time_per_iteration": 3.321467161178589
    },
    {
      "auxiliary_loss_clip": 0.01140989,
      "auxiliary_loss_mlp": 0.01036001,
      "balance_loss_clip": 1.05089867,
      "balance_loss_mlp": 1.02868712,
      "epoch": 0.5572055552215476,
      "flos": 16108091153280.0,
      "grad_norm": 1.8059749404313463,
      "language_loss": 0.78944975,
      "learning_rate": 1.727431833554301e-06,
      "loss": 0.81121969,
      "num_input_tokens_seen": 99974140,
      "step": 4634,
      "time_per_iteration": 2.521886110305786
    },
    {
      "auxiliary_loss_clip": 0.01114021,
      "auxiliary_loss_mlp": 0.01024146,
      "balance_loss_clip": 1.04610968,
      "balance_loss_mlp": 1.01646328,
      "epoch": 0.5573257981121866,
      "flos": 17128815937920.0,
      "grad_norm": 1.791886050918992,
      "language_loss": 0.77460384,
      "learning_rate": 1.7266601505635175e-06,
      "loss": 0.79598558,
      "num_input_tokens_seen": 99991480,
      "step": 4635,
      "time_per_iteration": 3.4280803203582764
    },
    {
      "auxiliary_loss_clip": 0.01162694,
      "auxiliary_loss_mlp": 0.01026877,
      "balance_loss_clip": 1.05179167,
      "balance_loss_mlp": 1.01921773,
      "epoch": 0.5574460410028257,
      "flos": 18807029222400.0,
      "grad_norm": 2.4125204917444436,
      "language_loss": 0.75997543,
      "learning_rate": 1.7258885090381717e-06,
      "loss": 0.7818712,
      "num_input_tokens_seen": 100009520,
      "step": 4636,
      "time_per_iteration": 2.451374053955078
    },
    {
      "auxiliary_loss_clip": 0.01150358,
      "auxiliary_loss_mlp": 0.01028099,
      "balance_loss_clip": 1.04871297,
      "balance_loss_mlp": 1.02099407,
      "epoch": 0.5575662838934649,
      "flos": 29642678530560.0,
      "grad_norm": 1.8321429816259176,
      "language_loss": 0.7845186,
      "learning_rate": 1.7251169090953213e-06,
      "loss": 0.80630314,
      "num_input_tokens_seen": 100029995,
      "step": 4637,
      "time_per_iteration": 2.5829689502716064
    },
    {
      "auxiliary_loss_clip": 0.01161407,
      "auxiliary_loss_mlp": 0.01024314,
      "balance_loss_clip": 1.04985428,
      "balance_loss_mlp": 1.01630306,
      "epoch": 0.5576865267841039,
      "flos": 22054466949120.0,
      "grad_norm": 2.5592453917876905,
      "language_loss": 0.75951588,
      "learning_rate": 1.7243453508520168e-06,
      "loss": 0.78137308,
      "num_input_tokens_seen": 100046980,
      "step": 4638,
      "time_per_iteration": 3.157209634780884
    },
    {
      "auxiliary_loss_clip": 0.01147754,
      "auxiliary_loss_mlp": 0.01027883,
      "balance_loss_clip": 1.04773486,
      "balance_loss_mlp": 1.02013993,
      "epoch": 0.557806769674743,
      "flos": 17196040241280.0,
      "grad_norm": 1.8866592588179323,
      "language_loss": 0.8457588,
      "learning_rate": 1.7235738344253038e-06,
      "loss": 0.86751521,
      "num_input_tokens_seen": 100060610,
      "step": 4639,
      "time_per_iteration": 2.4657938480377197
    },
    {
      "auxiliary_loss_clip": 0.01164245,
      "auxiliary_loss_mlp": 0.01028033,
      "balance_loss_clip": 1.055089,
      "balance_loss_mlp": 1.01990926,
      "epoch": 0.557927012565382,
      "flos": 24712717887360.0,
      "grad_norm": 3.1801102619615538,
      "language_loss": 0.82909626,
      "learning_rate": 1.72280235993222e-06,
      "loss": 0.85101902,
      "num_input_tokens_seen": 100078915,
      "step": 4640,
      "time_per_iteration": 2.508254051208496
    },
    {
      "auxiliary_loss_clip": 0.01157918,
      "auxiliary_loss_mlp": 0.00762758,
      "balance_loss_clip": 1.04921222,
      "balance_loss_mlp": 1.00029778,
      "epoch": 0.5580472554560212,
      "flos": 16983090460800.0,
      "grad_norm": 2.2492420488409355,
      "language_loss": 0.69526637,
      "learning_rate": 1.722030927489798e-06,
      "loss": 0.71447319,
      "num_input_tokens_seen": 100096195,
      "step": 4641,
      "time_per_iteration": 2.440819501876831
    },
    {
      "auxiliary_loss_clip": 0.011362,
      "auxiliary_loss_mlp": 0.01022272,
      "balance_loss_clip": 1.05016398,
      "balance_loss_mlp": 1.01445508,
      "epoch": 0.5581674983466602,
      "flos": 23509100027520.0,
      "grad_norm": 1.6734723263534574,
      "language_loss": 0.74127352,
      "learning_rate": 1.7212595372150634e-06,
      "loss": 0.76285827,
      "num_input_tokens_seen": 100116175,
      "step": 4642,
      "time_per_iteration": 2.5422399044036865
    },
    {
      "auxiliary_loss_clip": 0.01178055,
      "auxiliary_loss_mlp": 0.01025011,
      "balance_loss_clip": 1.05357552,
      "balance_loss_mlp": 1.01781082,
      "epoch": 0.5582877412372993,
      "flos": 13480291969920.0,
      "grad_norm": 2.3068966582922714,
      "language_loss": 0.72841978,
      "learning_rate": 1.720488189225035e-06,
      "loss": 0.75045049,
      "num_input_tokens_seen": 100133875,
      "step": 4643,
      "time_per_iteration": 2.3839259147644043
    },
    {
      "auxiliary_loss_clip": 0.01163627,
      "auxiliary_loss_mlp": 0.010264,
      "balance_loss_clip": 1.05009913,
      "balance_loss_mlp": 1.01868665,
      "epoch": 0.5584079841279385,
      "flos": 21903605827200.0,
      "grad_norm": 2.1320329244769236,
      "language_loss": 0.79387081,
      "learning_rate": 1.7197168836367265e-06,
      "loss": 0.8157711,
      "num_input_tokens_seen": 100150685,
      "step": 4644,
      "time_per_iteration": 2.4428772926330566
    },
    {
      "auxiliary_loss_clip": 0.01157057,
      "auxiliary_loss_mlp": 0.00762141,
      "balance_loss_clip": 1.04720819,
      "balance_loss_mlp": 1.00025892,
      "epoch": 0.5585282270185775,
      "flos": 18843550375680.0,
      "grad_norm": 1.8273082612237137,
      "language_loss": 0.81958282,
      "learning_rate": 1.7189456205671433e-06,
      "loss": 0.8387748,
      "num_input_tokens_seen": 100169530,
      "step": 4645,
      "time_per_iteration": 2.4554295539855957
    },
    {
      "auxiliary_loss_clip": 0.01170389,
      "auxiliary_loss_mlp": 0.01026292,
      "balance_loss_clip": 1.05173147,
      "balance_loss_mlp": 1.01820397,
      "epoch": 0.5586484699092166,
      "flos": 21868449390720.0,
      "grad_norm": 1.7662071525388225,
      "language_loss": 0.82310963,
      "learning_rate": 1.7181744001332866e-06,
      "loss": 0.84507644,
      "num_input_tokens_seen": 100188140,
      "step": 4646,
      "time_per_iteration": 2.4654436111450195
    },
    {
      "auxiliary_loss_clip": 0.0117644,
      "auxiliary_loss_mlp": 0.01025342,
      "balance_loss_clip": 1.05408692,
      "balance_loss_mlp": 1.01763546,
      "epoch": 0.5587687127998557,
      "flos": 22893232412160.0,
      "grad_norm": 1.7128976657847303,
      "language_loss": 0.63498676,
      "learning_rate": 1.7174032224521493e-06,
      "loss": 0.65700454,
      "num_input_tokens_seen": 100206850,
      "step": 4647,
      "time_per_iteration": 2.4171664714813232
    },
    {
      "auxiliary_loss_clip": 0.01161559,
      "auxiliary_loss_mlp": 0.01026541,
      "balance_loss_clip": 1.05078363,
      "balance_loss_mlp": 1.01918876,
      "epoch": 0.5588889556904948,
      "flos": 20303067703680.0,
      "grad_norm": 1.649069331894175,
      "language_loss": 0.69753414,
      "learning_rate": 1.7166320876407184e-06,
      "loss": 0.71941519,
      "num_input_tokens_seen": 100226270,
      "step": 4648,
      "time_per_iteration": 2.5370969772338867
    },
    {
      "auxiliary_loss_clip": 0.01179676,
      "auxiliary_loss_mlp": 0.00762517,
      "balance_loss_clip": 1.0536679,
      "balance_loss_mlp": 1.00035954,
      "epoch": 0.5590091985811338,
      "flos": 16472153450880.0,
      "grad_norm": 2.00652214296276,
      "language_loss": 0.67454857,
      "learning_rate": 1.7158609958159742e-06,
      "loss": 0.6939705,
      "num_input_tokens_seen": 100243675,
      "step": 4649,
      "time_per_iteration": 2.410348653793335
    },
    {
      "auxiliary_loss_clip": 0.01118583,
      "auxiliary_loss_mlp": 0.01032325,
      "balance_loss_clip": 1.04787612,
      "balance_loss_mlp": 1.02455831,
      "epoch": 0.559129441471773,
      "flos": 14532186781440.0,
      "grad_norm": 5.473230571034753,
      "language_loss": 0.78052616,
      "learning_rate": 1.7150899470948911e-06,
      "loss": 0.80203521,
      "num_input_tokens_seen": 100258940,
      "step": 4650,
      "time_per_iteration": 2.5439555644989014
    },
    {
      "auxiliary_loss_clip": 0.01056703,
      "auxiliary_loss_mlp": 0.01001233,
      "balance_loss_clip": 1.0197506,
      "balance_loss_mlp": 1.00019574,
      "epoch": 0.5592496843624121,
      "flos": 60521009852160.0,
      "grad_norm": 0.7966790626454702,
      "language_loss": 0.56700563,
      "learning_rate": 1.7143189415944365e-06,
      "loss": 0.58758503,
      "num_input_tokens_seen": 100323400,
      "step": 4651,
      "time_per_iteration": 3.0973634719848633
    },
    {
      "auxiliary_loss_clip": 0.01162009,
      "auxiliary_loss_mlp": 0.01025623,
      "balance_loss_clip": 1.05181646,
      "balance_loss_mlp": 1.01738596,
      "epoch": 0.5593699272530511,
      "flos": 20886256920960.0,
      "grad_norm": 1.632918942786101,
      "language_loss": 0.7624889,
      "learning_rate": 1.7135479794315714e-06,
      "loss": 0.78436518,
      "num_input_tokens_seen": 100340355,
      "step": 4652,
      "time_per_iteration": 2.449916362762451
    },
    {
      "auxiliary_loss_clip": 0.01132379,
      "auxiliary_loss_mlp": 0.01022603,
      "balance_loss_clip": 1.04853582,
      "balance_loss_mlp": 1.01524162,
      "epoch": 0.5594901701436903,
      "flos": 12896743616640.0,
      "grad_norm": 1.918269431886607,
      "language_loss": 0.79292631,
      "learning_rate": 1.7127770607232502e-06,
      "loss": 0.81447613,
      "num_input_tokens_seen": 100358900,
      "step": 4653,
      "time_per_iteration": 2.5069708824157715
    },
    {
      "auxiliary_loss_clip": 0.01141167,
      "auxiliary_loss_mlp": 0.01024676,
      "balance_loss_clip": 1.04892516,
      "balance_loss_mlp": 1.01671898,
      "epoch": 0.5596104130343293,
      "flos": 23112107936640.0,
      "grad_norm": 20.5625506750041,
      "language_loss": 0.79580224,
      "learning_rate": 1.7120061855864204e-06,
      "loss": 0.81746072,
      "num_input_tokens_seen": 100378910,
      "step": 4654,
      "time_per_iteration": 2.54665470123291
    },
    {
      "auxiliary_loss_clip": 0.01164816,
      "auxiliary_loss_mlp": 0.01028303,
      "balance_loss_clip": 1.05463767,
      "balance_loss_mlp": 1.02052474,
      "epoch": 0.5597306559249684,
      "flos": 25957812977280.0,
      "grad_norm": 5.0736947359311095,
      "language_loss": 0.70961249,
      "learning_rate": 1.7112353541380233e-06,
      "loss": 0.73154372,
      "num_input_tokens_seen": 100398770,
      "step": 4655,
      "time_per_iteration": 2.4963390827178955
    },
    {
      "auxiliary_loss_clip": 0.0115177,
      "auxiliary_loss_mlp": 0.01030571,
      "balance_loss_clip": 1.05194855,
      "balance_loss_mlp": 1.02184474,
      "epoch": 0.5598508988156076,
      "flos": 22492289825280.0,
      "grad_norm": 1.487328339271208,
      "language_loss": 0.72375858,
      "learning_rate": 1.7104645664949931e-06,
      "loss": 0.74558198,
      "num_input_tokens_seen": 100421240,
      "step": 4656,
      "time_per_iteration": 3.320931911468506
    },
    {
      "auxiliary_loss_clip": 0.01151403,
      "auxiliary_loss_mlp": 0.01027767,
      "balance_loss_clip": 1.04784155,
      "balance_loss_mlp": 1.01947641,
      "epoch": 0.5599711417062466,
      "flos": 23112538899840.0,
      "grad_norm": 2.0556188410887994,
      "language_loss": 0.71332437,
      "learning_rate": 1.7096938227742584e-06,
      "loss": 0.73511606,
      "num_input_tokens_seen": 100442370,
      "step": 4657,
      "time_per_iteration": 2.521167039871216
    },
    {
      "auxiliary_loss_clip": 0.01178348,
      "auxiliary_loss_mlp": 0.01026276,
      "balance_loss_clip": 1.05337214,
      "balance_loss_mlp": 1.01842594,
      "epoch": 0.5600913845968857,
      "flos": 22339345714560.0,
      "grad_norm": 1.8397435219302765,
      "language_loss": 0.84366721,
      "learning_rate": 1.70892312309274e-06,
      "loss": 0.86571342,
      "num_input_tokens_seen": 100460260,
      "step": 4658,
      "time_per_iteration": 2.433173894882202
    },
    {
      "auxiliary_loss_clip": 0.01147959,
      "auxiliary_loss_mlp": 0.0102594,
      "balance_loss_clip": 1.04343128,
      "balance_loss_mlp": 1.01782179,
      "epoch": 0.5602116274875248,
      "flos": 17633791290240.0,
      "grad_norm": 2.975392959743797,
      "language_loss": 0.68112212,
      "learning_rate": 1.7081524675673523e-06,
      "loss": 0.70286107,
      "num_input_tokens_seen": 100475750,
      "step": 4659,
      "time_per_iteration": 2.4482812881469727
    },
    {
      "auxiliary_loss_clip": 0.01056263,
      "auxiliary_loss_mlp": 0.0100064,
      "balance_loss_clip": 1.01519918,
      "balance_loss_mlp": 0.99962026,
      "epoch": 0.5603318703781639,
      "flos": 70115945529600.0,
      "grad_norm": 0.7713027304735945,
      "language_loss": 0.5960052,
      "learning_rate": 1.7073818563150026e-06,
      "loss": 0.61657417,
      "num_input_tokens_seen": 100537830,
      "step": 4660,
      "time_per_iteration": 3.9954657554626465
    },
    {
      "auxiliary_loss_clip": 0.01159389,
      "auxiliary_loss_mlp": 0.01025404,
      "balance_loss_clip": 1.04952502,
      "balance_loss_mlp": 1.01765525,
      "epoch": 0.560452113268803,
      "flos": 18545850455040.0,
      "grad_norm": 2.1787942141850234,
      "language_loss": 0.86644322,
      "learning_rate": 1.7066112894525935e-06,
      "loss": 0.88829112,
      "num_input_tokens_seen": 100555910,
      "step": 4661,
      "time_per_iteration": 2.4842474460601807
    },
    {
      "auxiliary_loss_clip": 0.01141045,
      "auxiliary_loss_mlp": 0.01031546,
      "balance_loss_clip": 1.04810047,
      "balance_loss_mlp": 1.02371418,
      "epoch": 0.5605723561594421,
      "flos": 25264665250560.0,
      "grad_norm": 1.6296382630935142,
      "language_loss": 0.72843611,
      "learning_rate": 1.7058407670970177e-06,
      "loss": 0.75016201,
      "num_input_tokens_seen": 100577385,
      "step": 4662,
      "time_per_iteration": 3.357377529144287
    },
    {
      "auxiliary_loss_clip": 0.01167646,
      "auxiliary_loss_mlp": 0.01029488,
      "balance_loss_clip": 1.05021274,
      "balance_loss_mlp": 1.02133441,
      "epoch": 0.5606925990500812,
      "flos": 20594949621120.0,
      "grad_norm": 1.6277637897738084,
      "language_loss": 0.60955441,
      "learning_rate": 1.7050702893651643e-06,
      "loss": 0.63152575,
      "num_input_tokens_seen": 100596965,
      "step": 4663,
      "time_per_iteration": 2.4452908039093018
    },
    {
      "auxiliary_loss_clip": 0.01163357,
      "auxiliary_loss_mlp": 0.01026493,
      "balance_loss_clip": 1.05200791,
      "balance_loss_mlp": 1.01842904,
      "epoch": 0.5608128419407202,
      "flos": 35006044677120.0,
      "grad_norm": 2.5419460593517735,
      "language_loss": 0.76049721,
      "learning_rate": 1.7042998563739134e-06,
      "loss": 0.78239572,
      "num_input_tokens_seen": 100615315,
      "step": 4664,
      "time_per_iteration": 2.560274362564087
    },
    {
      "auxiliary_loss_clip": 0.01155389,
      "auxiliary_loss_mlp": 0.0103225,
      "balance_loss_clip": 1.04751158,
      "balance_loss_mlp": 1.0241617,
      "epoch": 0.5609330848313594,
      "flos": 24639819235200.0,
      "grad_norm": 2.256913655105567,
      "language_loss": 0.71309018,
      "learning_rate": 1.703529468240139e-06,
      "loss": 0.73496658,
      "num_input_tokens_seen": 100634185,
      "step": 4665,
      "time_per_iteration": 3.2564074993133545
    },
    {
      "auxiliary_loss_clip": 0.01142823,
      "auxiliary_loss_mlp": 0.01027203,
      "balance_loss_clip": 1.05012894,
      "balance_loss_mlp": 1.01922441,
      "epoch": 0.5610533277219985,
      "flos": 18762894385920.0,
      "grad_norm": 2.372843300588213,
      "language_loss": 0.74257457,
      "learning_rate": 1.7027591250807088e-06,
      "loss": 0.76427484,
      "num_input_tokens_seen": 100651360,
      "step": 4666,
      "time_per_iteration": 2.4598982334136963
    },
    {
      "auxiliary_loss_clip": 0.01181383,
      "auxiliary_loss_mlp": 0.01026381,
      "balance_loss_clip": 1.05544686,
      "balance_loss_mlp": 1.01853704,
      "epoch": 0.5611735706126375,
      "flos": 15012384727680.0,
      "grad_norm": 2.4330320735330004,
      "language_loss": 0.84826428,
      "learning_rate": 1.7019888270124825e-06,
      "loss": 0.8703419,
      "num_input_tokens_seen": 100668525,
      "step": 4667,
      "time_per_iteration": 2.3941941261291504
    },
    {
      "auxiliary_loss_clip": 0.01169649,
      "auxiliary_loss_mlp": 0.01029658,
      "balance_loss_clip": 1.05386961,
      "balance_loss_mlp": 1.02099109,
      "epoch": 0.5612938135032767,
      "flos": 16468167041280.0,
      "grad_norm": 1.7602220885835493,
      "language_loss": 0.82137924,
      "learning_rate": 1.7012185741523147e-06,
      "loss": 0.84337234,
      "num_input_tokens_seen": 100684850,
      "step": 4668,
      "time_per_iteration": 2.4194586277008057
    },
    {
      "auxiliary_loss_clip": 0.01180237,
      "auxiliary_loss_mlp": 0.0102903,
      "balance_loss_clip": 1.05491328,
      "balance_loss_mlp": 1.02130532,
      "epoch": 0.5614140563939157,
      "flos": 25666433850240.0,
      "grad_norm": 2.8924875264821504,
      "language_loss": 0.62510872,
      "learning_rate": 1.7004483666170514e-06,
      "loss": 0.64720142,
      "num_input_tokens_seen": 100705345,
      "step": 4669,
      "time_per_iteration": 2.4603211879730225
    },
    {
      "auxiliary_loss_clip": 0.01163957,
      "auxiliary_loss_mlp": 0.01026964,
      "balance_loss_clip": 1.05115509,
      "balance_loss_mlp": 1.01964474,
      "epoch": 0.5615342992845548,
      "flos": 24717566223360.0,
      "grad_norm": 2.130757627893791,
      "language_loss": 0.80699134,
      "learning_rate": 1.699678204523533e-06,
      "loss": 0.82890058,
      "num_input_tokens_seen": 100725210,
      "step": 4670,
      "time_per_iteration": 2.473994731903076
    },
    {
      "auxiliary_loss_clip": 0.01154442,
      "auxiliary_loss_mlp": 0.01029491,
      "balance_loss_clip": 1.05346835,
      "balance_loss_mlp": 1.02069354,
      "epoch": 0.5616545421751938,
      "flos": 22015934634240.0,
      "grad_norm": 3.787960458819434,
      "language_loss": 0.69138598,
      "learning_rate": 1.6989080879885918e-06,
      "loss": 0.71322536,
      "num_input_tokens_seen": 100743070,
      "step": 4671,
      "time_per_iteration": 2.504103660583496
    },
    {
      "auxiliary_loss_clip": 0.01042694,
      "auxiliary_loss_mlp": 0.0100182,
      "balance_loss_clip": 1.01499963,
      "balance_loss_mlp": 1.00078893,
      "epoch": 0.561774785065833,
      "flos": 53760358690560.0,
      "grad_norm": 0.9010901182194945,
      "language_loss": 0.61043864,
      "learning_rate": 1.6981380171290544e-06,
      "loss": 0.63088381,
      "num_input_tokens_seen": 100804095,
      "step": 4672,
      "time_per_iteration": 3.06972074508667
    },
    {
      "auxiliary_loss_clip": 0.01145017,
      "auxiliary_loss_mlp": 0.01028078,
      "balance_loss_clip": 1.04662192,
      "balance_loss_mlp": 1.02015662,
      "epoch": 0.5618950279564721,
      "flos": 19750007018880.0,
      "grad_norm": 1.9095044973685826,
      "language_loss": 0.74349374,
      "learning_rate": 1.6973679920617396e-06,
      "loss": 0.76522464,
      "num_input_tokens_seen": 100821630,
      "step": 4673,
      "time_per_iteration": 2.4905998706817627
    },
    {
      "auxiliary_loss_clip": 0.01147745,
      "auxiliary_loss_mlp": 0.01027616,
      "balance_loss_clip": 1.05116308,
      "balance_loss_mlp": 1.01955128,
      "epoch": 0.5620152708471111,
      "flos": 16800592435200.0,
      "grad_norm": 2.422210034430206,
      "language_loss": 0.85206586,
      "learning_rate": 1.6965980129034603e-06,
      "loss": 0.87381947,
      "num_input_tokens_seen": 100839015,
      "step": 4674,
      "time_per_iteration": 2.469625473022461
    },
    {
      "auxiliary_loss_clip": 0.01152973,
      "auxiliary_loss_mlp": 0.01025876,
      "balance_loss_clip": 1.0539403,
      "balance_loss_mlp": 1.01809764,
      "epoch": 0.5621355137377503,
      "flos": 26797799502720.0,
      "grad_norm": 1.5154359158780253,
      "language_loss": 0.76440513,
      "learning_rate": 1.6958280797710209e-06,
      "loss": 0.78619361,
      "num_input_tokens_seen": 100860940,
      "step": 4675,
      "time_per_iteration": 2.5471930503845215
    },
    {
      "auxiliary_loss_clip": 0.01053591,
      "auxiliary_loss_mlp": 0.01002585,
      "balance_loss_clip": 1.01535988,
      "balance_loss_mlp": 1.00156581,
      "epoch": 0.5622557566283893,
      "flos": 61207046686080.0,
      "grad_norm": 0.7154377444825417,
      "language_loss": 0.54814363,
      "learning_rate": 1.6950581927812198e-06,
      "loss": 0.56870538,
      "num_input_tokens_seen": 100920510,
      "step": 4676,
      "time_per_iteration": 2.943524122238159
    },
    {
      "auxiliary_loss_clip": 0.01164336,
      "auxiliary_loss_mlp": 0.01026777,
      "balance_loss_clip": 1.05168009,
      "balance_loss_mlp": 1.01879597,
      "epoch": 0.5623759995190284,
      "flos": 26468534505600.0,
      "grad_norm": 2.952573418583003,
      "language_loss": 0.79496479,
      "learning_rate": 1.6942883520508486e-06,
      "loss": 0.81687587,
      "num_input_tokens_seen": 100939245,
      "step": 4677,
      "time_per_iteration": 2.500335931777954
    },
    {
      "auxiliary_loss_clip": 0.01164734,
      "auxiliary_loss_mlp": 0.01025996,
      "balance_loss_clip": 1.05082297,
      "balance_loss_mlp": 1.01802731,
      "epoch": 0.5624962424096676,
      "flos": 19390900798080.0,
      "grad_norm": 2.0054358670043393,
      "language_loss": 0.77308154,
      "learning_rate": 1.693518557696691e-06,
      "loss": 0.79498887,
      "num_input_tokens_seen": 100958385,
      "step": 4678,
      "time_per_iteration": 2.434480667114258
    },
    {
      "auxiliary_loss_clip": 0.01159946,
      "auxiliary_loss_mlp": 0.01024886,
      "balance_loss_clip": 1.04851234,
      "balance_loss_mlp": 1.01710153,
      "epoch": 0.5626164853003066,
      "flos": 20667345482880.0,
      "grad_norm": 2.010738097474918,
      "language_loss": 0.88986361,
      "learning_rate": 1.6927488098355252e-06,
      "loss": 0.91171193,
      "num_input_tokens_seen": 100976015,
      "step": 4679,
      "time_per_iteration": 2.4582018852233887
    },
    {
      "auxiliary_loss_clip": 0.01038544,
      "auxiliary_loss_mlp": 0.01003137,
      "balance_loss_clip": 1.01413584,
      "balance_loss_mlp": 1.00193274,
      "epoch": 0.5627367281909457,
      "flos": 62766071665920.0,
      "grad_norm": 0.9094488422671037,
      "language_loss": 0.63175243,
      "learning_rate": 1.6919791085841201e-06,
      "loss": 0.65216923,
      "num_input_tokens_seen": 101033425,
      "step": 4680,
      "time_per_iteration": 3.0931854248046875
    },
    {
      "auxiliary_loss_clip": 0.0115805,
      "auxiliary_loss_mlp": 0.01030213,
      "balance_loss_clip": 1.04772329,
      "balance_loss_mlp": 1.0214572,
      "epoch": 0.5628569710815848,
      "flos": 12787144243200.0,
      "grad_norm": 2.2988181829739522,
      "language_loss": 0.78827047,
      "learning_rate": 1.6912094540592396e-06,
      "loss": 0.81015313,
      "num_input_tokens_seen": 101048945,
      "step": 4681,
      "time_per_iteration": 2.4543240070343018
    },
    {
      "auxiliary_loss_clip": 0.01162639,
      "auxiliary_loss_mlp": 0.01028973,
      "balance_loss_clip": 1.05067933,
      "balance_loss_mlp": 1.02122474,
      "epoch": 0.5629772139722239,
      "flos": 13762082165760.0,
      "grad_norm": 5.605968390907182,
      "language_loss": 0.81280792,
      "learning_rate": 1.6904398463776393e-06,
      "loss": 0.83472401,
      "num_input_tokens_seen": 101062745,
      "step": 4682,
      "time_per_iteration": 2.3975889682769775
    },
    {
      "auxiliary_loss_clip": 0.0116499,
      "auxiliary_loss_mlp": 0.01025936,
      "balance_loss_clip": 1.0495671,
      "balance_loss_mlp": 1.01804996,
      "epoch": 0.5630974568628629,
      "flos": 21467830026240.0,
      "grad_norm": 1.7068709996142475,
      "language_loss": 0.72629416,
      "learning_rate": 1.6896702856560683e-06,
      "loss": 0.7482034,
      "num_input_tokens_seen": 101081840,
      "step": 4683,
      "time_per_iteration": 3.284876585006714
    },
    {
      "auxiliary_loss_clip": 0.01133642,
      "auxiliary_loss_mlp": 0.01025367,
      "balance_loss_clip": 1.04509115,
      "balance_loss_mlp": 1.01748109,
      "epoch": 0.5632176997535021,
      "flos": 14245907385600.0,
      "grad_norm": 2.825214411476421,
      "language_loss": 0.69354212,
      "learning_rate": 1.6889007720112677e-06,
      "loss": 0.71513218,
      "num_input_tokens_seen": 101099585,
      "step": 4684,
      "time_per_iteration": 2.4920742511749268
    },
    {
      "auxiliary_loss_clip": 0.01167496,
      "auxiliary_loss_mlp": 0.01026046,
      "balance_loss_clip": 1.05389929,
      "balance_loss_mlp": 1.01875997,
      "epoch": 0.5633379426441412,
      "flos": 20812244947200.0,
      "grad_norm": 1.5469558476558942,
      "language_loss": 0.77391911,
      "learning_rate": 1.6881313055599734e-06,
      "loss": 0.79585457,
      "num_input_tokens_seen": 101119515,
      "step": 4685,
      "time_per_iteration": 2.456150531768799
    },
    {
      "auxiliary_loss_clip": 0.01136625,
      "auxiliary_loss_mlp": 0.01023181,
      "balance_loss_clip": 1.04566884,
      "balance_loss_mlp": 1.01486015,
      "epoch": 0.5634581855347802,
      "flos": 22600883617920.0,
      "grad_norm": 2.4028321410814883,
      "language_loss": 0.82260096,
      "learning_rate": 1.6873618864189117e-06,
      "loss": 0.84419894,
      "num_input_tokens_seen": 101135285,
      "step": 4686,
      "time_per_iteration": 3.324130058288574
    },
    {
      "auxiliary_loss_clip": 0.01163568,
      "auxiliary_loss_mlp": 0.01033185,
      "balance_loss_clip": 1.05002165,
      "balance_loss_mlp": 1.02490616,
      "epoch": 0.5635784284254194,
      "flos": 21506972872320.0,
      "grad_norm": 2.2035073149545683,
      "language_loss": 0.7750845,
      "learning_rate": 1.686592514704803e-06,
      "loss": 0.79705203,
      "num_input_tokens_seen": 101152680,
      "step": 4687,
      "time_per_iteration": 2.4573662281036377
    },
    {
      "auxiliary_loss_clip": 0.01149164,
      "auxiliary_loss_mlp": 0.01026419,
      "balance_loss_clip": 1.05315936,
      "balance_loss_mlp": 1.01918936,
      "epoch": 0.5636986713160584,
      "flos": 19827466698240.0,
      "grad_norm": 2.2018676842780556,
      "language_loss": 0.70977378,
      "learning_rate": 1.685823190534361e-06,
      "loss": 0.73152965,
      "num_input_tokens_seen": 101170920,
      "step": 4688,
      "time_per_iteration": 2.476706027984619
    },
    {
      "auxiliary_loss_clip": 0.01181615,
      "auxiliary_loss_mlp": 0.01024859,
      "balance_loss_clip": 1.05358446,
      "balance_loss_mlp": 1.016204,
      "epoch": 0.5638189142066975,
      "flos": 19792453916160.0,
      "grad_norm": 1.8715929235791309,
      "language_loss": 0.83937359,
      "learning_rate": 1.6850539140242907e-06,
      "loss": 0.86143827,
      "num_input_tokens_seen": 101190180,
      "step": 4689,
      "time_per_iteration": 3.2551026344299316
    },
    {
      "auxiliary_loss_clip": 0.01167918,
      "auxiliary_loss_mlp": 0.0103107,
      "balance_loss_clip": 1.05150843,
      "balance_loss_mlp": 1.0233779,
      "epoch": 0.5639391570973367,
      "flos": 22893771116160.0,
      "grad_norm": 1.882163472555844,
      "language_loss": 0.81961226,
      "learning_rate": 1.684284685291292e-06,
      "loss": 0.84160215,
      "num_input_tokens_seen": 101211825,
      "step": 4690,
      "time_per_iteration": 2.471123695373535
    },
    {
      "auxiliary_loss_clip": 0.01179296,
      "auxiliary_loss_mlp": 0.01031148,
      "balance_loss_clip": 1.05327809,
      "balance_loss_mlp": 1.02306604,
      "epoch": 0.5640593999879757,
      "flos": 23727077712000.0,
      "grad_norm": 1.8887818603786286,
      "language_loss": 0.81538939,
      "learning_rate": 1.683515504452055e-06,
      "loss": 0.83749378,
      "num_input_tokens_seen": 101229200,
      "step": 4691,
      "time_per_iteration": 3.263162612915039
    },
    {
      "auxiliary_loss_clip": 0.0112631,
      "auxiliary_loss_mlp": 0.01034121,
      "balance_loss_clip": 1.04475379,
      "balance_loss_mlp": 1.02512097,
      "epoch": 0.5641796428786148,
      "flos": 22710123855360.0,
      "grad_norm": 2.11691359539167,
      "language_loss": 0.66213882,
      "learning_rate": 1.6827463716232648e-06,
      "loss": 0.68374312,
      "num_input_tokens_seen": 101249860,
      "step": 4692,
      "time_per_iteration": 2.557039260864258
    },
    {
      "auxiliary_loss_clip": 0.01162309,
      "auxiliary_loss_mlp": 0.00762565,
      "balance_loss_clip": 1.05038953,
      "balance_loss_mlp": 1.00035405,
      "epoch": 0.5642998857692539,
      "flos": 19791987039360.0,
      "grad_norm": 2.8169886462108717,
      "language_loss": 0.75555813,
      "learning_rate": 1.6819772869215972e-06,
      "loss": 0.77480686,
      "num_input_tokens_seen": 101268940,
      "step": 4693,
      "time_per_iteration": 2.4532949924468994
    },
    {
      "auxiliary_loss_clip": 0.01155663,
      "auxiliary_loss_mlp": 0.01028177,
      "balance_loss_clip": 1.05228376,
      "balance_loss_mlp": 1.02092314,
      "epoch": 0.564420128659893,
      "flos": 23185904428800.0,
      "grad_norm": 1.8697942062197417,
      "language_loss": 0.82250357,
      "learning_rate": 1.6812082504637228e-06,
      "loss": 0.84434199,
      "num_input_tokens_seen": 101290260,
      "step": 4694,
      "time_per_iteration": 2.5473644733428955
    },
    {
      "auxiliary_loss_clip": 0.01161255,
      "auxiliary_loss_mlp": 0.01022309,
      "balance_loss_clip": 1.05277026,
      "balance_loss_mlp": 1.0144887,
      "epoch": 0.564540371550532,
      "flos": 23258264376960.0,
      "grad_norm": 1.6775129219736873,
      "language_loss": 0.74346316,
      "learning_rate": 1.6804392623663025e-06,
      "loss": 0.76529878,
      "num_input_tokens_seen": 101311465,
      "step": 4695,
      "time_per_iteration": 2.5016591548919678
    },
    {
      "auxiliary_loss_clip": 0.01156502,
      "auxiliary_loss_mlp": 0.01025042,
      "balance_loss_clip": 1.04980719,
      "balance_loss_mlp": 1.01694155,
      "epoch": 0.5646606144411712,
      "flos": 25010058672000.0,
      "grad_norm": 1.940493550732572,
      "language_loss": 0.78128088,
      "learning_rate": 1.6796703227459935e-06,
      "loss": 0.80309623,
      "num_input_tokens_seen": 101329420,
      "step": 4696,
      "time_per_iteration": 2.481167793273926
    },
    {
      "auxiliary_loss_clip": 0.01112191,
      "auxiliary_loss_mlp": 0.01024885,
      "balance_loss_clip": 1.04430425,
      "balance_loss_mlp": 1.01689839,
      "epoch": 0.5647808573318103,
      "flos": 36539645806080.0,
      "grad_norm": 1.7537009572840152,
      "language_loss": 0.7605378,
      "learning_rate": 1.6789014317194407e-06,
      "loss": 0.78190857,
      "num_input_tokens_seen": 101350900,
      "step": 4697,
      "time_per_iteration": 2.7260093688964844
    },
    {
      "auxiliary_loss_clip": 0.01159502,
      "auxiliary_loss_mlp": 0.01028812,
      "balance_loss_clip": 1.05382454,
      "balance_loss_mlp": 1.02039647,
      "epoch": 0.5649011002224493,
      "flos": 22528451842560.0,
      "grad_norm": 2.605412069969427,
      "language_loss": 0.72936368,
      "learning_rate": 1.6781325894032853e-06,
      "loss": 0.75124681,
      "num_input_tokens_seen": 101369860,
      "step": 4698,
      "time_per_iteration": 2.546757221221924
    },
    {
      "auxiliary_loss_clip": 0.01148021,
      "auxiliary_loss_mlp": 0.01032979,
      "balance_loss_clip": 1.05302024,
      "balance_loss_mlp": 1.02506351,
      "epoch": 0.5650213431130885,
      "flos": 18515147304960.0,
      "grad_norm": 1.9566633712524366,
      "language_loss": 0.92016459,
      "learning_rate": 1.6773637959141608e-06,
      "loss": 0.94197464,
      "num_input_tokens_seen": 101386835,
      "step": 4699,
      "time_per_iteration": 2.4633591175079346
    },
    {
      "auxiliary_loss_clip": 0.01139715,
      "auxiliary_loss_mlp": 0.01028008,
      "balance_loss_clip": 1.04763365,
      "balance_loss_mlp": 1.02000892,
      "epoch": 0.5651415860037275,
      "flos": 17526310819200.0,
      "grad_norm": 2.056013661694584,
      "language_loss": 0.66345894,
      "learning_rate": 1.6765950513686915e-06,
      "loss": 0.68513614,
      "num_input_tokens_seen": 101404945,
      "step": 4700,
      "time_per_iteration": 2.4759459495544434
    },
    {
      "auxiliary_loss_clip": 0.01121685,
      "auxiliary_loss_mlp": 0.01033106,
      "balance_loss_clip": 1.04298306,
      "balance_loss_mlp": 1.0247252,
      "epoch": 0.5652618288943666,
      "flos": 25520026014720.0,
      "grad_norm": 2.1644489644946385,
      "language_loss": 0.76034021,
      "learning_rate": 1.675826355883496e-06,
      "loss": 0.78188813,
      "num_input_tokens_seen": 101424160,
      "step": 4701,
      "time_per_iteration": 2.634275197982788
    },
    {
      "auxiliary_loss_clip": 0.01144623,
      "auxiliary_loss_mlp": 0.01031596,
      "balance_loss_clip": 1.05011082,
      "balance_loss_mlp": 1.02342439,
      "epoch": 0.5653820717850057,
      "flos": 19683105937920.0,
      "grad_norm": 2.4807865185467373,
      "language_loss": 0.78902924,
      "learning_rate": 1.6750577095751848e-06,
      "loss": 0.81079137,
      "num_input_tokens_seen": 101443270,
      "step": 4702,
      "time_per_iteration": 2.4767277240753174
    },
    {
      "auxiliary_loss_clip": 0.01175368,
      "auxiliary_loss_mlp": 0.01031682,
      "balance_loss_clip": 1.05147231,
      "balance_loss_mlp": 1.02386153,
      "epoch": 0.5655023146756448,
      "flos": 26979722910720.0,
      "grad_norm": 1.7288361627247184,
      "language_loss": 0.7252205,
      "learning_rate": 1.6742891125603605e-06,
      "loss": 0.74729097,
      "num_input_tokens_seen": 101464175,
      "step": 4703,
      "time_per_iteration": 2.464181661605835
    },
    {
      "auxiliary_loss_clip": 0.01161838,
      "auxiliary_loss_mlp": 0.01027017,
      "balance_loss_clip": 1.05123281,
      "balance_loss_mlp": 1.01860094,
      "epoch": 0.5656225575662839,
      "flos": 27669351104640.0,
      "grad_norm": 1.8079627743755122,
      "language_loss": 0.72152948,
      "learning_rate": 1.6735205649556185e-06,
      "loss": 0.74341804,
      "num_input_tokens_seen": 101484045,
      "step": 4704,
      "time_per_iteration": 2.515594959259033
    },
    {
      "auxiliary_loss_clip": 0.01138262,
      "auxiliary_loss_mlp": 0.01029459,
      "balance_loss_clip": 1.0479877,
      "balance_loss_mlp": 1.02154326,
      "epoch": 0.5657428004569229,
      "flos": 24349732997760.0,
      "grad_norm": 1.6270290300969787,
      "language_loss": 0.84739369,
      "learning_rate": 1.6727520668775476e-06,
      "loss": 0.86907089,
      "num_input_tokens_seen": 101504330,
      "step": 4705,
      "time_per_iteration": 2.556506633758545
    },
    {
      "auxiliary_loss_clip": 0.01179483,
      "auxiliary_loss_mlp": 0.01029435,
      "balance_loss_clip": 1.05179179,
      "balance_loss_mlp": 1.02099538,
      "epoch": 0.5658630433475621,
      "flos": 21944041562880.0,
      "grad_norm": 1.5995932013267553,
      "language_loss": 0.7536822,
      "learning_rate": 1.6719836184427275e-06,
      "loss": 0.77577138,
      "num_input_tokens_seen": 101524635,
      "step": 4706,
      "time_per_iteration": 2.4360644817352295
    },
    {
      "auxiliary_loss_clip": 0.0114712,
      "auxiliary_loss_mlp": 0.01026595,
      "balance_loss_clip": 1.04825616,
      "balance_loss_mlp": 1.01923943,
      "epoch": 0.5659832862382012,
      "flos": 30409012218240.0,
      "grad_norm": 2.0394749216234076,
      "language_loss": 0.64126742,
      "learning_rate": 1.671215219767733e-06,
      "loss": 0.66300452,
      "num_input_tokens_seen": 101544095,
      "step": 4707,
      "time_per_iteration": 2.5578556060791016
    },
    {
      "auxiliary_loss_clip": 0.01123661,
      "auxiliary_loss_mlp": 0.01031328,
      "balance_loss_clip": 1.04695809,
      "balance_loss_mlp": 1.02317405,
      "epoch": 0.5661035291288402,
      "flos": 13188194570880.0,
      "grad_norm": 2.291189978795589,
      "language_loss": 0.7618767,
      "learning_rate": 1.670446870969127e-06,
      "loss": 0.78342664,
      "num_input_tokens_seen": 101561760,
      "step": 4708,
      "time_per_iteration": 2.5221760272979736
    },
    {
      "auxiliary_loss_clip": 0.0115383,
      "auxiliary_loss_mlp": 0.01028261,
      "balance_loss_clip": 1.05117726,
      "balance_loss_mlp": 1.02042902,
      "epoch": 0.5662237720194794,
      "flos": 16143032108160.0,
      "grad_norm": 2.1385840302027335,
      "language_loss": 0.8010447,
      "learning_rate": 1.6696785721634685e-06,
      "loss": 0.82286561,
      "num_input_tokens_seen": 101576245,
      "step": 4709,
      "time_per_iteration": 3.2841742038726807
    },
    {
      "auxiliary_loss_clip": 0.01164776,
      "auxiliary_loss_mlp": 0.01033832,
      "balance_loss_clip": 1.0502218,
      "balance_loss_mlp": 1.02562487,
      "epoch": 0.5663440149101184,
      "flos": 17676848718720.0,
      "grad_norm": 2.0248618267950516,
      "language_loss": 0.73495239,
      "learning_rate": 1.6689103234673086e-06,
      "loss": 0.75693852,
      "num_input_tokens_seen": 101594565,
      "step": 4710,
      "time_per_iteration": 2.4507322311401367
    },
    {
      "auxiliary_loss_clip": 0.01148322,
      "auxiliary_loss_mlp": 0.01029706,
      "balance_loss_clip": 1.05052042,
      "balance_loss_mlp": 1.0216713,
      "epoch": 0.5664642578007575,
      "flos": 23368330627200.0,
      "grad_norm": 4.111201650649018,
      "language_loss": 0.76954913,
      "learning_rate": 1.668142124997189e-06,
      "loss": 0.79132938,
      "num_input_tokens_seen": 101614225,
      "step": 4711,
      "time_per_iteration": 2.5090441703796387
    },
    {
      "auxiliary_loss_clip": 0.01048453,
      "auxiliary_loss_mlp": 0.01004458,
      "balance_loss_clip": 1.01747072,
      "balance_loss_mlp": 1.0035342,
      "epoch": 0.5665845006913967,
      "flos": 65516470945920.0,
      "grad_norm": 0.7247073504934648,
      "language_loss": 0.59846222,
      "learning_rate": 1.6673739768696453e-06,
      "loss": 0.61899137,
      "num_input_tokens_seen": 101680795,
      "step": 4712,
      "time_per_iteration": 3.0884652137756348
    },
    {
      "auxiliary_loss_clip": 0.01156187,
      "auxiliary_loss_mlp": 0.01028721,
      "balance_loss_clip": 1.04947615,
      "balance_loss_mlp": 1.020105,
      "epoch": 0.5667047435820357,
      "flos": 26140885620480.0,
      "grad_norm": 2.56243291798559,
      "language_loss": 0.77386022,
      "learning_rate": 1.6666058792012052e-06,
      "loss": 0.79570925,
      "num_input_tokens_seen": 101701680,
      "step": 4713,
      "time_per_iteration": 3.3893685340881348
    },
    {
      "auxiliary_loss_clip": 0.01065614,
      "auxiliary_loss_mlp": 0.01001889,
      "balance_loss_clip": 1.01392603,
      "balance_loss_mlp": 1.00088787,
      "epoch": 0.5668249864726748,
      "flos": 71866949725440.0,
      "grad_norm": 0.881443577343635,
      "language_loss": 0.68800932,
      "learning_rate": 1.6658378321083878e-06,
      "loss": 0.70868433,
      "num_input_tokens_seen": 101766010,
      "step": 4714,
      "time_per_iteration": 3.082977056503296
    },
    {
      "auxiliary_loss_clip": 0.01111679,
      "auxiliary_loss_mlp": 0.01025905,
      "balance_loss_clip": 1.04485428,
      "balance_loss_mlp": 1.01834989,
      "epoch": 0.5669452293633139,
      "flos": 22195667312640.0,
      "grad_norm": 2.310161133539983,
      "language_loss": 0.82514113,
      "learning_rate": 1.6650698357077055e-06,
      "loss": 0.84651697,
      "num_input_tokens_seen": 101783055,
      "step": 4715,
      "time_per_iteration": 2.621511459350586
    },
    {
      "auxiliary_loss_clip": 0.01154776,
      "auxiliary_loss_mlp": 0.01032341,
      "balance_loss_clip": 1.04928899,
      "balance_loss_mlp": 1.02381158,
      "epoch": 0.567065472253953,
      "flos": 18223193560320.0,
      "grad_norm": 2.923296883424349,
      "language_loss": 0.80919373,
      "learning_rate": 1.6643018901156632e-06,
      "loss": 0.83106494,
      "num_input_tokens_seen": 101802150,
      "step": 4716,
      "time_per_iteration": 3.3085148334503174
    },
    {
      "auxiliary_loss_clip": 0.01155781,
      "auxiliary_loss_mlp": 0.01026654,
      "balance_loss_clip": 1.05090141,
      "balance_loss_mlp": 1.0188309,
      "epoch": 0.567185715144592,
      "flos": 20371548983040.0,
      "grad_norm": 3.0750420763824127,
      "language_loss": 0.79597688,
      "learning_rate": 1.6635339954487566e-06,
      "loss": 0.81780124,
      "num_input_tokens_seen": 101818025,
      "step": 4717,
      "time_per_iteration": 2.4891278743743896
    },
    {
      "auxiliary_loss_clip": 0.01154342,
      "auxiliary_loss_mlp": 0.01026684,
      "balance_loss_clip": 1.04996347,
      "balance_loss_mlp": 1.01880467,
      "epoch": 0.5673059580352312,
      "flos": 23221348174080.0,
      "grad_norm": 1.815736403216912,
      "language_loss": 0.82125729,
      "learning_rate": 1.6627661518234765e-06,
      "loss": 0.84306753,
      "num_input_tokens_seen": 101837280,
      "step": 4718,
      "time_per_iteration": 3.285160779953003
    },
    {
      "auxiliary_loss_clip": 0.01126167,
      "auxiliary_loss_mlp": 0.0102668,
      "balance_loss_clip": 1.04962659,
      "balance_loss_mlp": 1.01835918,
      "epoch": 0.5674262009258703,
      "flos": 21719599430400.0,
      "grad_norm": 1.5566206295639413,
      "language_loss": 0.85178322,
      "learning_rate": 1.661998359356302e-06,
      "loss": 0.87331164,
      "num_input_tokens_seen": 101856310,
      "step": 4719,
      "time_per_iteration": 2.6344470977783203
    },
    {
      "auxiliary_loss_clip": 0.01074504,
      "auxiliary_loss_mlp": 0.01003021,
      "balance_loss_clip": 1.01428938,
      "balance_loss_mlp": 1.00207376,
      "epoch": 0.5675464438165093,
      "flos": 67470369114240.0,
      "grad_norm": 0.7465739752672518,
      "language_loss": 0.55847228,
      "learning_rate": 1.6612306181637077e-06,
      "loss": 0.57924747,
      "num_input_tokens_seen": 101915635,
      "step": 4720,
      "time_per_iteration": 2.9846572875976562
    },
    {
      "auxiliary_loss_clip": 0.01132208,
      "auxiliary_loss_mlp": 0.01031774,
      "balance_loss_clip": 1.0465306,
      "balance_loss_mlp": 1.02393615,
      "epoch": 0.5676666867071485,
      "flos": 18879173688960.0,
      "grad_norm": 2.850211380728475,
      "language_loss": 0.65545309,
      "learning_rate": 1.6604629283621598e-06,
      "loss": 0.67709291,
      "num_input_tokens_seen": 101933565,
      "step": 4721,
      "time_per_iteration": 2.5548489093780518
    },
    {
      "auxiliary_loss_clip": 0.0118074,
      "auxiliary_loss_mlp": 0.01032694,
      "balance_loss_clip": 1.05319858,
      "balance_loss_mlp": 1.02409911,
      "epoch": 0.5677869295977875,
      "flos": 33546778744320.0,
      "grad_norm": 1.8710218947673942,
      "language_loss": 0.74162489,
      "learning_rate": 1.6596952900681152e-06,
      "loss": 0.76375926,
      "num_input_tokens_seen": 101954325,
      "step": 4722,
      "time_per_iteration": 2.5515873432159424
    },
    {
      "auxiliary_loss_clip": 0.01120108,
      "auxiliary_loss_mlp": 0.0103173,
      "balance_loss_clip": 1.053478,
      "balance_loss_mlp": 1.02299154,
      "epoch": 0.5679071724884266,
      "flos": 28037256157440.0,
      "grad_norm": 2.162052294578075,
      "language_loss": 0.81846702,
      "learning_rate": 1.658927703398025e-06,
      "loss": 0.83998537,
      "num_input_tokens_seen": 101974390,
      "step": 4723,
      "time_per_iteration": 2.6018714904785156
    },
    {
      "auxiliary_loss_clip": 0.01119953,
      "auxiliary_loss_mlp": 0.01026077,
      "balance_loss_clip": 1.04188263,
      "balance_loss_mlp": 1.01801205,
      "epoch": 0.5680274153790658,
      "flos": 23550110380800.0,
      "grad_norm": 2.5546751487288164,
      "language_loss": 0.7823928,
      "learning_rate": 1.6581601684683309e-06,
      "loss": 0.80385315,
      "num_input_tokens_seen": 101994815,
      "step": 4724,
      "time_per_iteration": 2.5922391414642334
    },
    {
      "auxiliary_loss_clip": 0.0116453,
      "auxiliary_loss_mlp": 0.0102915,
      "balance_loss_clip": 1.05205464,
      "balance_loss_mlp": 1.02184284,
      "epoch": 0.5681476582697048,
      "flos": 22455158140800.0,
      "grad_norm": 2.6734174527055248,
      "language_loss": 0.68262964,
      "learning_rate": 1.6573926853954674e-06,
      "loss": 0.70456648,
      "num_input_tokens_seen": 102012400,
      "step": 4725,
      "time_per_iteration": 2.4544126987457275
    },
    {
      "auxiliary_loss_clip": 0.01141419,
      "auxiliary_loss_mlp": 0.01025158,
      "balance_loss_clip": 1.0447166,
      "balance_loss_mlp": 1.01714468,
      "epoch": 0.5682679011603439,
      "flos": 19536913584000.0,
      "grad_norm": 1.8993748613005417,
      "language_loss": 0.8323307,
      "learning_rate": 1.6566252542958608e-06,
      "loss": 0.85399646,
      "num_input_tokens_seen": 102031900,
      "step": 4726,
      "time_per_iteration": 2.5696568489074707
    },
    {
      "auxiliary_loss_clip": 0.01131414,
      "auxiliary_loss_mlp": 0.01030684,
      "balance_loss_clip": 1.04947686,
      "balance_loss_mlp": 1.02264643,
      "epoch": 0.568388144050983,
      "flos": 28765488493440.0,
      "grad_norm": 1.9272812791611291,
      "language_loss": 0.78774786,
      "learning_rate": 1.6558578752859305e-06,
      "loss": 0.80936885,
      "num_input_tokens_seen": 102050860,
      "step": 4727,
      "time_per_iteration": 2.585646152496338
    },
    {
      "auxiliary_loss_clip": 0.01134925,
      "auxiliary_loss_mlp": 0.01025424,
      "balance_loss_clip": 1.04784811,
      "balance_loss_mlp": 1.01809299,
      "epoch": 0.5685083869416221,
      "flos": 21209452519680.0,
      "grad_norm": 3.4885767045964595,
      "language_loss": 0.78853911,
      "learning_rate": 1.6550905484820865e-06,
      "loss": 0.81014264,
      "num_input_tokens_seen": 102069320,
      "step": 4728,
      "time_per_iteration": 2.530128002166748
    },
    {
      "auxiliary_loss_clip": 0.01180409,
      "auxiliary_loss_mlp": 0.01028559,
      "balance_loss_clip": 1.05264819,
      "balance_loss_mlp": 1.02021992,
      "epoch": 0.5686286298322611,
      "flos": 24827021942400.0,
      "grad_norm": 2.1654557125574616,
      "language_loss": 0.79086977,
      "learning_rate": 1.6543232740007328e-06,
      "loss": 0.81295943,
      "num_input_tokens_seen": 102086435,
      "step": 4729,
      "time_per_iteration": 2.4443626403808594
    },
    {
      "auxiliary_loss_clip": 0.01167989,
      "auxiliary_loss_mlp": 0.01027817,
      "balance_loss_clip": 1.05272341,
      "balance_loss_mlp": 1.0197109,
      "epoch": 0.5687488727229003,
      "flos": 26615121909120.0,
      "grad_norm": 2.5494685832169597,
      "language_loss": 0.67172599,
      "learning_rate": 1.653556051958263e-06,
      "loss": 0.69368404,
      "num_input_tokens_seen": 102106115,
      "step": 4730,
      "time_per_iteration": 2.507558584213257
    },
    {
      "auxiliary_loss_clip": 0.0108906,
      "auxiliary_loss_mlp": 0.01026949,
      "balance_loss_clip": 1.04206955,
      "balance_loss_mlp": 1.01878262,
      "epoch": 0.5688691156135394,
      "flos": 20808725414400.0,
      "grad_norm": 2.099238030279477,
      "language_loss": 0.74115926,
      "learning_rate": 1.6527888824710642e-06,
      "loss": 0.76231939,
      "num_input_tokens_seen": 102125715,
      "step": 4731,
      "time_per_iteration": 2.610666036605835
    },
    {
      "auxiliary_loss_clip": 0.01128763,
      "auxiliary_loss_mlp": 0.0102978,
      "balance_loss_clip": 1.04457283,
      "balance_loss_mlp": 1.02127421,
      "epoch": 0.5689893585041784,
      "flos": 25880963829120.0,
      "grad_norm": 2.0808562393608097,
      "language_loss": 0.76409227,
      "learning_rate": 1.6520217656555166e-06,
      "loss": 0.78567767,
      "num_input_tokens_seen": 102145005,
      "step": 4732,
      "time_per_iteration": 2.553199529647827
    },
    {
      "auxiliary_loss_clip": 0.01138573,
      "auxiliary_loss_mlp": 0.01031371,
      "balance_loss_clip": 1.0479393,
      "balance_loss_mlp": 1.02349174,
      "epoch": 0.5691096013948175,
      "flos": 23477463123840.0,
      "grad_norm": 1.4975347240184178,
      "language_loss": 0.71006942,
      "learning_rate": 1.65125470162799e-06,
      "loss": 0.73176897,
      "num_input_tokens_seen": 102165360,
      "step": 4733,
      "time_per_iteration": 2.511319398880005
    },
    {
      "auxiliary_loss_clip": 0.01135727,
      "auxiliary_loss_mlp": 0.01030275,
      "balance_loss_clip": 1.04484749,
      "balance_loss_mlp": 1.02233577,
      "epoch": 0.5692298442854566,
      "flos": 18075600576000.0,
      "grad_norm": 1.9645453373813944,
      "language_loss": 0.69870341,
      "learning_rate": 1.6504876905048485e-06,
      "loss": 0.72036344,
      "num_input_tokens_seen": 102182320,
      "step": 4734,
      "time_per_iteration": 2.519029140472412
    },
    {
      "auxiliary_loss_clip": 0.01178031,
      "auxiliary_loss_mlp": 0.01025647,
      "balance_loss_clip": 1.05421352,
      "balance_loss_mlp": 1.01810122,
      "epoch": 0.5693500871760957,
      "flos": 23039317025280.0,
      "grad_norm": 1.855190993853517,
      "language_loss": 0.72103029,
      "learning_rate": 1.6497207324024464e-06,
      "loss": 0.74306709,
      "num_input_tokens_seen": 102201220,
      "step": 4735,
      "time_per_iteration": 2.4509618282318115
    },
    {
      "auxiliary_loss_clip": 0.01157478,
      "auxiliary_loss_mlp": 0.01029767,
      "balance_loss_clip": 1.04972386,
      "balance_loss_mlp": 1.02180362,
      "epoch": 0.5694703300667348,
      "flos": 18989670902400.0,
      "grad_norm": 13.772233856028635,
      "language_loss": 0.82644027,
      "learning_rate": 1.6489538274371305e-06,
      "loss": 0.84831274,
      "num_input_tokens_seen": 102219825,
      "step": 4736,
      "time_per_iteration": 3.3168396949768066
    },
    {
      "auxiliary_loss_clip": 0.01158419,
      "auxiliary_loss_mlp": 0.01028877,
      "balance_loss_clip": 1.05204082,
      "balance_loss_mlp": 1.02127445,
      "epoch": 0.5695905729573739,
      "flos": 21908705558400.0,
      "grad_norm": 1.871675628880176,
      "language_loss": 0.83240873,
      "learning_rate": 1.6481869757252396e-06,
      "loss": 0.85428166,
      "num_input_tokens_seen": 102238160,
      "step": 4737,
      "time_per_iteration": 2.4798386096954346
    },
    {
      "auxiliary_loss_clip": 0.01165729,
      "auxiliary_loss_mlp": 0.01030887,
      "balance_loss_clip": 1.05389452,
      "balance_loss_mlp": 1.02334714,
      "epoch": 0.569710815848013,
      "flos": 28476659232000.0,
      "grad_norm": 1.470773345299097,
      "language_loss": 0.71815145,
      "learning_rate": 1.647420177383105e-06,
      "loss": 0.74011767,
      "num_input_tokens_seen": 102261030,
      "step": 4738,
      "time_per_iteration": 2.5471749305725098
    },
    {
      "auxiliary_loss_clip": 0.01160067,
      "auxiliary_loss_mlp": 0.01024843,
      "balance_loss_clip": 1.05376577,
      "balance_loss_mlp": 1.01752019,
      "epoch": 0.569831058738652,
      "flos": 28366162018560.0,
      "grad_norm": 1.804050814597485,
      "language_loss": 0.72515523,
      "learning_rate": 1.646653432527049e-06,
      "loss": 0.74700433,
      "num_input_tokens_seen": 102281670,
      "step": 4739,
      "time_per_iteration": 3.3799397945404053
    },
    {
      "auxiliary_loss_clip": 0.01139071,
      "auxiliary_loss_mlp": 0.01025129,
      "balance_loss_clip": 1.05113959,
      "balance_loss_mlp": 1.01740456,
      "epoch": 0.5699513016292912,
      "flos": 25849973370240.0,
      "grad_norm": 1.5147571984855663,
      "language_loss": 0.74583161,
      "learning_rate": 1.645886741273387e-06,
      "loss": 0.76747358,
      "num_input_tokens_seen": 102303485,
      "step": 4740,
      "time_per_iteration": 2.582517623901367
    },
    {
      "auxiliary_loss_clip": 0.01134086,
      "auxiliary_loss_mlp": 0.01033202,
      "balance_loss_clip": 1.05308735,
      "balance_loss_mlp": 1.0250783,
      "epoch": 0.5700715445199303,
      "flos": 18037858360320.0,
      "grad_norm": 1.8819933248917702,
      "language_loss": 0.7388643,
      "learning_rate": 1.645120103738424e-06,
      "loss": 0.76053715,
      "num_input_tokens_seen": 102320995,
      "step": 4741,
      "time_per_iteration": 2.5203750133514404
    },
    {
      "auxiliary_loss_clip": 0.01151465,
      "auxiliary_loss_mlp": 0.00762225,
      "balance_loss_clip": 1.04902637,
      "balance_loss_mlp": 1.00041485,
      "epoch": 0.5701917874105693,
      "flos": 11473352392320.0,
      "grad_norm": 2.1662987553458968,
      "language_loss": 0.83098698,
      "learning_rate": 1.6443535200384591e-06,
      "loss": 0.85012382,
      "num_input_tokens_seen": 102339170,
      "step": 4742,
      "time_per_iteration": 3.3173928260803223
    },
    {
      "auxiliary_loss_clip": 0.01180091,
      "auxiliary_loss_mlp": 0.01032067,
      "balance_loss_clip": 1.05387616,
      "balance_loss_mlp": 1.02391887,
      "epoch": 0.5703120303012085,
      "flos": 21761759018880.0,
      "grad_norm": 1.7339324478765894,
      "language_loss": 0.70340449,
      "learning_rate": 1.6435869902897827e-06,
      "loss": 0.72552609,
      "num_input_tokens_seen": 102357750,
      "step": 4743,
      "time_per_iteration": 2.4915976524353027
    },
    {
      "auxiliary_loss_clip": 0.01042878,
      "auxiliary_loss_mlp": 0.01002232,
      "balance_loss_clip": 1.01622462,
      "balance_loss_mlp": 1.00115955,
      "epoch": 0.5704322731918475,
      "flos": 56746258513920.0,
      "grad_norm": 0.7916318192070402,
      "language_loss": 0.62012851,
      "learning_rate": 1.6428205146086764e-06,
      "loss": 0.6405797,
      "num_input_tokens_seen": 102419730,
      "step": 4744,
      "time_per_iteration": 3.156371593475342
    },
    {
      "auxiliary_loss_clip": 0.01155329,
      "auxiliary_loss_mlp": 0.01027641,
      "balance_loss_clip": 1.04986286,
      "balance_loss_mlp": 1.01962423,
      "epoch": 0.5705525160824866,
      "flos": 20741141975040.0,
      "grad_norm": 1.5881622900639671,
      "language_loss": 0.70522523,
      "learning_rate": 1.6420540931114142e-06,
      "loss": 0.72705495,
      "num_input_tokens_seen": 102440320,
      "step": 4745,
      "time_per_iteration": 3.334642171859741
    },
    {
      "auxiliary_loss_clip": 0.01150573,
      "auxiliary_loss_mlp": 0.01039313,
      "balance_loss_clip": 1.0490464,
      "balance_loss_mlp": 1.0315944,
      "epoch": 0.5706727589731257,
      "flos": 18771262254720.0,
      "grad_norm": 2.9952781073061967,
      "language_loss": 0.78852087,
      "learning_rate": 1.6412877259142616e-06,
      "loss": 0.81041968,
      "num_input_tokens_seen": 102460240,
      "step": 4746,
      "time_per_iteration": 2.5048701763153076
    },
    {
      "auxiliary_loss_clip": 0.01147025,
      "auxiliary_loss_mlp": 0.01026542,
      "balance_loss_clip": 1.05052066,
      "balance_loss_mlp": 1.0187099,
      "epoch": 0.5707930018637648,
      "flos": 27634733372160.0,
      "grad_norm": 3.0466808057231773,
      "language_loss": 0.73664892,
      "learning_rate": 1.6405214131334757e-06,
      "loss": 0.75838459,
      "num_input_tokens_seen": 102478765,
      "step": 4747,
      "time_per_iteration": 2.5273215770721436
    },
    {
      "auxiliary_loss_clip": 0.01119149,
      "auxiliary_loss_mlp": 0.0102502,
      "balance_loss_clip": 1.04923499,
      "balance_loss_mlp": 1.01717043,
      "epoch": 0.5709132447544039,
      "flos": 27597673514880.0,
      "grad_norm": 2.1510479692751954,
      "language_loss": 0.79685748,
      "learning_rate": 1.6397551548853052e-06,
      "loss": 0.81829917,
      "num_input_tokens_seen": 102496930,
      "step": 4748,
      "time_per_iteration": 2.60506534576416
    },
    {
      "auxiliary_loss_clip": 0.01150623,
      "auxiliary_loss_mlp": 0.01027717,
      "balance_loss_clip": 1.05115271,
      "balance_loss_mlp": 1.01989698,
      "epoch": 0.571033487645043,
      "flos": 21686095019520.0,
      "grad_norm": 1.707413932690761,
      "language_loss": 0.70669395,
      "learning_rate": 1.6389889512859917e-06,
      "loss": 0.72847736,
      "num_input_tokens_seen": 102516590,
      "step": 4749,
      "time_per_iteration": 2.495048999786377
    },
    {
      "auxiliary_loss_clip": 0.01051427,
      "auxiliary_loss_mlp": 0.01001945,
      "balance_loss_clip": 1.01452708,
      "balance_loss_mlp": 1.00094342,
      "epoch": 0.5711537305356821,
      "flos": 70181445980160.0,
      "grad_norm": 1.5239875771762392,
      "language_loss": 0.60341245,
      "learning_rate": 1.638222802451767e-06,
      "loss": 0.62394613,
      "num_input_tokens_seen": 102578070,
      "step": 4750,
      "time_per_iteration": 3.078207015991211
    },
    {
      "auxiliary_loss_clip": 0.01159788,
      "auxiliary_loss_mlp": 0.01025052,
      "balance_loss_clip": 1.05223393,
      "balance_loss_mlp": 1.01755369,
      "epoch": 0.5712739734263211,
      "flos": 24717494396160.0,
      "grad_norm": 3.5052093593161797,
      "language_loss": 0.75152946,
      "learning_rate": 1.6374567084988561e-06,
      "loss": 0.77337784,
      "num_input_tokens_seen": 102599255,
      "step": 4751,
      "time_per_iteration": 2.490955114364624
    },
    {
      "auxiliary_loss_clip": 0.01156749,
      "auxiliary_loss_mlp": 0.01027565,
      "balance_loss_clip": 1.05352747,
      "balance_loss_mlp": 1.01873159,
      "epoch": 0.5713942163169603,
      "flos": 26578169792640.0,
      "grad_norm": 1.9093210784622567,
      "language_loss": 0.76478082,
      "learning_rate": 1.6366906695434738e-06,
      "loss": 0.78662395,
      "num_input_tokens_seen": 102621775,
      "step": 4752,
      "time_per_iteration": 2.5401628017425537
    },
    {
      "auxiliary_loss_clip": 0.01167696,
      "auxiliary_loss_mlp": 0.01028785,
      "balance_loss_clip": 1.0564239,
      "balance_loss_mlp": 1.02113795,
      "epoch": 0.5715144592075994,
      "flos": 21142443697920.0,
      "grad_norm": 2.0891406145388656,
      "language_loss": 0.86240387,
      "learning_rate": 1.6359246857018275e-06,
      "loss": 0.88436872,
      "num_input_tokens_seen": 102639305,
      "step": 4753,
      "time_per_iteration": 2.456295967102051
    },
    {
      "auxiliary_loss_clip": 0.01121055,
      "auxiliary_loss_mlp": 0.01024696,
      "balance_loss_clip": 1.04586601,
      "balance_loss_mlp": 1.01672101,
      "epoch": 0.5716347020982384,
      "flos": 23330265189120.0,
      "grad_norm": 5.332455642797424,
      "language_loss": 0.7806741,
      "learning_rate": 1.6351587570901178e-06,
      "loss": 0.80213165,
      "num_input_tokens_seen": 102659430,
      "step": 4754,
      "time_per_iteration": 2.570239305496216
    },
    {
      "auxiliary_loss_clip": 0.0113945,
      "auxiliary_loss_mlp": 0.01030149,
      "balance_loss_clip": 1.05317783,
      "balance_loss_mlp": 1.02257323,
      "epoch": 0.5717549449888776,
      "flos": 17009555806080.0,
      "grad_norm": 3.52161720218778,
      "language_loss": 0.75976098,
      "learning_rate": 1.634392883824534e-06,
      "loss": 0.78145695,
      "num_input_tokens_seen": 102671430,
      "step": 4755,
      "time_per_iteration": 2.4697248935699463
    },
    {
      "auxiliary_loss_clip": 0.01124218,
      "auxiliary_loss_mlp": 0.0102753,
      "balance_loss_clip": 1.04681683,
      "balance_loss_mlp": 1.0193162,
      "epoch": 0.5718751878795166,
      "flos": 35518130922240.0,
      "grad_norm": 1.7099825837522165,
      "language_loss": 0.67718029,
      "learning_rate": 1.6336270660212595e-06,
      "loss": 0.69869781,
      "num_input_tokens_seen": 102693025,
      "step": 4756,
      "time_per_iteration": 2.679170608520508
    },
    {
      "auxiliary_loss_clip": 0.01151473,
      "auxiliary_loss_mlp": 0.01028719,
      "balance_loss_clip": 1.05642188,
      "balance_loss_mlp": 1.01994491,
      "epoch": 0.5719954307701557,
      "flos": 38613989255040.0,
      "grad_norm": 2.098109128742455,
      "language_loss": 0.65970272,
      "learning_rate": 1.6328613037964676e-06,
      "loss": 0.68150461,
      "num_input_tokens_seen": 102716090,
      "step": 4757,
      "time_per_iteration": 2.632429599761963
    },
    {
      "auxiliary_loss_clip": 0.0116407,
      "auxiliary_loss_mlp": 0.01024895,
      "balance_loss_clip": 1.05146694,
      "balance_loss_mlp": 1.01708078,
      "epoch": 0.5721156736607949,
      "flos": 20631111638400.0,
      "grad_norm": 2.2049568480010024,
      "language_loss": 0.67957032,
      "learning_rate": 1.6320955972663241e-06,
      "loss": 0.70145994,
      "num_input_tokens_seen": 102735685,
      "step": 4758,
      "time_per_iteration": 2.4628214836120605
    },
    {
      "auxiliary_loss_clip": 0.01165425,
      "auxiliary_loss_mlp": 0.01028752,
      "balance_loss_clip": 1.05173123,
      "balance_loss_mlp": 1.02092004,
      "epoch": 0.5722359165514339,
      "flos": 37415076076800.0,
      "grad_norm": 3.2790342293734995,
      "language_loss": 0.65478462,
      "learning_rate": 1.6313299465469857e-06,
      "loss": 0.67672646,
      "num_input_tokens_seen": 102758415,
      "step": 4759,
      "time_per_iteration": 2.6060216426849365
    },
    {
      "auxiliary_loss_clip": 0.0116235,
      "auxiliary_loss_mlp": 0.01029642,
      "balance_loss_clip": 1.05295944,
      "balance_loss_mlp": 1.02113032,
      "epoch": 0.572356159442073,
      "flos": 21972877205760.0,
      "grad_norm": 3.0941458673995004,
      "language_loss": 0.79842687,
      "learning_rate": 1.6305643517546014e-06,
      "loss": 0.82034677,
      "num_input_tokens_seen": 102773795,
      "step": 4760,
      "time_per_iteration": 2.4629385471343994
    },
    {
      "auxiliary_loss_clip": 0.01178226,
      "auxiliary_loss_mlp": 0.01036987,
      "balance_loss_clip": 1.05441713,
      "balance_loss_mlp": 1.02931631,
      "epoch": 0.5724764023327121,
      "flos": 19135540033920.0,
      "grad_norm": 2.205102241371106,
      "language_loss": 0.84799659,
      "learning_rate": 1.629798813005311e-06,
      "loss": 0.87014878,
      "num_input_tokens_seen": 102793515,
      "step": 4761,
      "time_per_iteration": 2.442936897277832
    },
    {
      "auxiliary_loss_clip": 0.01122685,
      "auxiliary_loss_mlp": 0.01028305,
      "balance_loss_clip": 1.04923034,
      "balance_loss_mlp": 1.02066636,
      "epoch": 0.5725966452233512,
      "flos": 22819759142400.0,
      "grad_norm": 2.060900104094086,
      "language_loss": 0.71421975,
      "learning_rate": 1.6290333304152473e-06,
      "loss": 0.73572958,
      "num_input_tokens_seen": 102813390,
      "step": 4762,
      "time_per_iteration": 3.454465866088867
    },
    {
      "auxiliary_loss_clip": 0.01150804,
      "auxiliary_loss_mlp": 0.01032457,
      "balance_loss_clip": 1.05635846,
      "balance_loss_mlp": 1.02433848,
      "epoch": 0.5727168881139902,
      "flos": 41496610498560.0,
      "grad_norm": 1.739227773846657,
      "language_loss": 0.56967151,
      "learning_rate": 1.6282679041005314e-06,
      "loss": 0.59150416,
      "num_input_tokens_seen": 102838980,
      "step": 4763,
      "time_per_iteration": 2.65539813041687
    },
    {
      "auxiliary_loss_clip": 0.01141807,
      "auxiliary_loss_mlp": 0.01023589,
      "balance_loss_clip": 1.04769182,
      "balance_loss_mlp": 1.01583421,
      "epoch": 0.5728371310046293,
      "flos": 14647675985280.0,
      "grad_norm": 4.273514517086232,
      "language_loss": 0.87447834,
      "learning_rate": 1.6275025341772789e-06,
      "loss": 0.89613229,
      "num_input_tokens_seen": 102855285,
      "step": 4764,
      "time_per_iteration": 2.451650857925415
    },
    {
      "auxiliary_loss_clip": 0.01150931,
      "auxiliary_loss_mlp": 0.01029313,
      "balance_loss_clip": 1.04970515,
      "balance_loss_mlp": 1.02058053,
      "epoch": 0.5729573738952685,
      "flos": 21506613736320.0,
      "grad_norm": 2.603330793803001,
      "language_loss": 0.81770635,
      "learning_rate": 1.626737220761596e-06,
      "loss": 0.83950877,
      "num_input_tokens_seen": 102872750,
      "step": 4765,
      "time_per_iteration": 2.4955108165740967
    },
    {
      "auxiliary_loss_clip": 0.01160917,
      "auxiliary_loss_mlp": 0.01028216,
      "balance_loss_clip": 1.0530231,
      "balance_loss_mlp": 1.02030087,
      "epoch": 0.5730776167859075,
      "flos": 23621680229760.0,
      "grad_norm": 2.3190990351956633,
      "language_loss": 0.7847755,
      "learning_rate": 1.62597196396958e-06,
      "loss": 0.80666685,
      "num_input_tokens_seen": 102890920,
      "step": 4766,
      "time_per_iteration": 3.3220200538635254
    },
    {
      "auxiliary_loss_clip": 0.01167287,
      "auxiliary_loss_mlp": 0.01026303,
      "balance_loss_clip": 1.05431175,
      "balance_loss_mlp": 1.01817298,
      "epoch": 0.5731978596765466,
      "flos": 25739224761600.0,
      "grad_norm": 1.854538852857469,
      "language_loss": 0.85461199,
      "learning_rate": 1.6252067639173197e-06,
      "loss": 0.87654787,
      "num_input_tokens_seen": 102912830,
      "step": 4767,
      "time_per_iteration": 2.495978832244873
    },
    {
      "auxiliary_loss_clip": 0.01165056,
      "auxiliary_loss_mlp": 0.01027951,
      "balance_loss_clip": 1.0521971,
      "balance_loss_mlp": 1.01991653,
      "epoch": 0.5733181025671857,
      "flos": 26359509749760.0,
      "grad_norm": 1.8074348970593332,
      "language_loss": 0.69824123,
      "learning_rate": 1.6244416207208956e-06,
      "loss": 0.72017127,
      "num_input_tokens_seen": 102933765,
      "step": 4768,
      "time_per_iteration": 3.297753095626831
    },
    {
      "auxiliary_loss_clip": 0.01138937,
      "auxiliary_loss_mlp": 0.01033193,
      "balance_loss_clip": 1.05142665,
      "balance_loss_mlp": 1.02530742,
      "epoch": 0.5734383454578248,
      "flos": 29423874833280.0,
      "grad_norm": 1.6160593431772499,
      "language_loss": 0.73711258,
      "learning_rate": 1.6236765344963787e-06,
      "loss": 0.75883389,
      "num_input_tokens_seen": 102955025,
      "step": 4769,
      "time_per_iteration": 2.586174964904785
    },
    {
      "auxiliary_loss_clip": 0.01151216,
      "auxiliary_loss_mlp": 0.01025143,
      "balance_loss_clip": 1.05109143,
      "balance_loss_mlp": 1.01734674,
      "epoch": 0.5735585883484638,
      "flos": 34969954487040.0,
      "grad_norm": 2.519054706805177,
      "language_loss": 0.68985176,
      "learning_rate": 1.6229115053598322e-06,
      "loss": 0.71161532,
      "num_input_tokens_seen": 102976780,
      "step": 4770,
      "time_per_iteration": 2.602721929550171
    },
    {
      "auxiliary_loss_clip": 0.01167753,
      "auxiliary_loss_mlp": 0.01030026,
      "balance_loss_clip": 1.05519235,
      "balance_loss_mlp": 1.02214599,
      "epoch": 0.573678831239103,
      "flos": 18770759464320.0,
      "grad_norm": 1.6986385436856126,
      "language_loss": 0.72035605,
      "learning_rate": 1.6221465334273108e-06,
      "loss": 0.74233383,
      "num_input_tokens_seen": 102995990,
      "step": 4771,
      "time_per_iteration": 3.1759603023529053
    },
    {
      "auxiliary_loss_clip": 0.01141856,
      "auxiliary_loss_mlp": 0.01024853,
      "balance_loss_clip": 1.04956734,
      "balance_loss_mlp": 1.01685977,
      "epoch": 0.5737990741297421,
      "flos": 25702883176320.0,
      "grad_norm": 2.1656902682139645,
      "language_loss": 0.61840439,
      "learning_rate": 1.6213816188148593e-06,
      "loss": 0.64007151,
      "num_input_tokens_seen": 103014695,
      "step": 4772,
      "time_per_iteration": 2.5560390949249268
    },
    {
      "auxiliary_loss_clip": 0.01145857,
      "auxiliary_loss_mlp": 0.01028749,
      "balance_loss_clip": 1.0553143,
      "balance_loss_mlp": 1.02075052,
      "epoch": 0.5739193170203811,
      "flos": 27269234530560.0,
      "grad_norm": 1.758216911477341,
      "language_loss": 0.77144563,
      "learning_rate": 1.6206167616385162e-06,
      "loss": 0.79319167,
      "num_input_tokens_seen": 103035760,
      "step": 4773,
      "time_per_iteration": 2.569058895111084
    },
    {
      "auxiliary_loss_clip": 0.0115817,
      "auxiliary_loss_mlp": 0.01028858,
      "balance_loss_clip": 1.05315411,
      "balance_loss_mlp": 1.02077007,
      "epoch": 0.5740395599110203,
      "flos": 12239721993600.0,
      "grad_norm": 3.3084818464407175,
      "language_loss": 0.73481417,
      "learning_rate": 1.6198519620143078e-06,
      "loss": 0.75668442,
      "num_input_tokens_seen": 103052915,
      "step": 4774,
      "time_per_iteration": 2.472527503967285
    },
    {
      "auxiliary_loss_clip": 0.01142225,
      "auxiliary_loss_mlp": 0.01032314,
      "balance_loss_clip": 1.0526861,
      "balance_loss_mlp": 1.02453017,
      "epoch": 0.5741598028016593,
      "flos": 25921399564800.0,
      "grad_norm": 1.5924289218745713,
      "language_loss": 0.78225183,
      "learning_rate": 1.6190872200582546e-06,
      "loss": 0.80399728,
      "num_input_tokens_seen": 103074655,
      "step": 4775,
      "time_per_iteration": 2.5884151458740234
    },
    {
      "auxiliary_loss_clip": 0.01146656,
      "auxiliary_loss_mlp": 0.00762394,
      "balance_loss_clip": 1.05057156,
      "balance_loss_mlp": 1.00039613,
      "epoch": 0.5742800456922984,
      "flos": 19244133826560.0,
      "grad_norm": 2.085643995961095,
      "language_loss": 0.7782892,
      "learning_rate": 1.6183225358863676e-06,
      "loss": 0.79737973,
      "num_input_tokens_seen": 103091550,
      "step": 4776,
      "time_per_iteration": 2.4914186000823975
    },
    {
      "auxiliary_loss_clip": 0.01141872,
      "auxiliary_loss_mlp": 0.01028435,
      "balance_loss_clip": 1.04854012,
      "balance_loss_mlp": 1.01994717,
      "epoch": 0.5744002885829376,
      "flos": 30920487932160.0,
      "grad_norm": 2.43199619117625,
      "language_loss": 0.71885306,
      "learning_rate": 1.617557909614648e-06,
      "loss": 0.74055612,
      "num_input_tokens_seen": 103110985,
      "step": 4777,
      "time_per_iteration": 2.5706887245178223
    },
    {
      "auxiliary_loss_clip": 0.01134175,
      "auxiliary_loss_mlp": 0.01027744,
      "balance_loss_clip": 1.04754949,
      "balance_loss_mlp": 1.02012062,
      "epoch": 0.5745205314735766,
      "flos": 23840017050240.0,
      "grad_norm": 2.1036283836514804,
      "language_loss": 0.86193657,
      "learning_rate": 1.6167933413590899e-06,
      "loss": 0.88355571,
      "num_input_tokens_seen": 103129890,
      "step": 4778,
      "time_per_iteration": 2.548201084136963
    },
    {
      "auxiliary_loss_clip": 0.01163781,
      "auxiliary_loss_mlp": 0.01031255,
      "balance_loss_clip": 1.05230212,
      "balance_loss_mlp": 1.02337563,
      "epoch": 0.5746407743642157,
      "flos": 12311902373760.0,
      "grad_norm": 7.07477411183321,
      "language_loss": 0.90704578,
      "learning_rate": 1.6160288312356773e-06,
      "loss": 0.92899609,
      "num_input_tokens_seen": 103147020,
      "step": 4779,
      "time_per_iteration": 2.458759307861328
    },
    {
      "auxiliary_loss_clip": 0.01168738,
      "auxiliary_loss_mlp": 0.01027353,
      "balance_loss_clip": 1.05215931,
      "balance_loss_mlp": 1.01888895,
      "epoch": 0.5747610172548548,
      "flos": 24133658734080.0,
      "grad_norm": 1.5443493866202302,
      "language_loss": 0.81743348,
      "learning_rate": 1.6152643793603857e-06,
      "loss": 0.83939433,
      "num_input_tokens_seen": 103167370,
      "step": 4780,
      "time_per_iteration": 2.4920387268066406
    },
    {
      "auxiliary_loss_clip": 0.01179624,
      "auxiliary_loss_mlp": 0.01027761,
      "balance_loss_clip": 1.05525708,
      "balance_loss_mlp": 1.02006555,
      "epoch": 0.5748812601454939,
      "flos": 25408451393280.0,
      "grad_norm": 1.7335684289408109,
      "language_loss": 0.87785637,
      "learning_rate": 1.6144999858491815e-06,
      "loss": 0.89993024,
      "num_input_tokens_seen": 103186000,
      "step": 4781,
      "time_per_iteration": 2.4937713146209717
    },
    {
      "auxiliary_loss_clip": 0.01156142,
      "auxiliary_loss_mlp": 0.01024267,
      "balance_loss_clip": 1.05169177,
      "balance_loss_mlp": 1.01584518,
      "epoch": 0.575001503036133,
      "flos": 30624942827520.0,
      "grad_norm": 1.595080321016195,
      "language_loss": 0.85633057,
      "learning_rate": 1.6137356508180232e-06,
      "loss": 0.87813461,
      "num_input_tokens_seen": 103207710,
      "step": 4782,
      "time_per_iteration": 2.57137131690979
    },
    {
      "auxiliary_loss_clip": 0.01180839,
      "auxiliary_loss_mlp": 0.00762296,
      "balance_loss_clip": 1.05387294,
      "balance_loss_mlp": 1.00040984,
      "epoch": 0.5751217459267721,
      "flos": 21726566668800.0,
      "grad_norm": 2.235736534573803,
      "language_loss": 0.81298852,
      "learning_rate": 1.6129713743828593e-06,
      "loss": 0.83241987,
      "num_input_tokens_seen": 103226720,
      "step": 4783,
      "time_per_iteration": 2.448772430419922
    },
    {
      "auxiliary_loss_clip": 0.01151621,
      "auxiliary_loss_mlp": 0.01027815,
      "balance_loss_clip": 1.04941893,
      "balance_loss_mlp": 1.02030456,
      "epoch": 0.5752419888174112,
      "flos": 21651620941440.0,
      "grad_norm": 1.523641375629479,
      "language_loss": 0.75432396,
      "learning_rate": 1.6122071566596306e-06,
      "loss": 0.77611828,
      "num_input_tokens_seen": 103246995,
      "step": 4784,
      "time_per_iteration": 2.5059280395507812
    },
    {
      "auxiliary_loss_clip": 0.01168195,
      "auxiliary_loss_mlp": 0.01029577,
      "balance_loss_clip": 1.05327964,
      "balance_loss_mlp": 1.02119637,
      "epoch": 0.5753622317080502,
      "flos": 17775997234560.0,
      "grad_norm": 2.744317140698451,
      "language_loss": 0.83043426,
      "learning_rate": 1.6114429977642674e-06,
      "loss": 0.85241199,
      "num_input_tokens_seen": 103261500,
      "step": 4785,
      "time_per_iteration": 2.424600601196289
    },
    {
      "auxiliary_loss_clip": 0.01166476,
      "auxiliary_loss_mlp": 0.0102614,
      "balance_loss_clip": 1.05516362,
      "balance_loss_mlp": 1.01873064,
      "epoch": 0.5754824745986894,
      "flos": 19789616741760.0,
      "grad_norm": 1.7930361664239713,
      "language_loss": 0.73213071,
      "learning_rate": 1.6106788978126926e-06,
      "loss": 0.75405687,
      "num_input_tokens_seen": 103280475,
      "step": 4786,
      "time_per_iteration": 2.4599227905273438
    },
    {
      "auxiliary_loss_clip": 0.01118287,
      "auxiliary_loss_mlp": 0.01032201,
      "balance_loss_clip": 1.04471731,
      "balance_loss_mlp": 1.02342129,
      "epoch": 0.5756027174893285,
      "flos": 30985665160320.0,
      "grad_norm": 5.146094576227733,
      "language_loss": 0.79100758,
      "learning_rate": 1.6099148569208196e-06,
      "loss": 0.81251246,
      "num_input_tokens_seen": 103297695,
      "step": 4787,
      "time_per_iteration": 2.624391794204712
    },
    {
      "auxiliary_loss_clip": 0.01154895,
      "auxiliary_loss_mlp": 0.01034565,
      "balance_loss_clip": 1.05490386,
      "balance_loss_mlp": 1.02555287,
      "epoch": 0.5757229603799675,
      "flos": 28546864364160.0,
      "grad_norm": 2.6128869188007537,
      "language_loss": 0.62864387,
      "learning_rate": 1.6091508752045523e-06,
      "loss": 0.6505385,
      "num_input_tokens_seen": 103318575,
      "step": 4788,
      "time_per_iteration": 2.5386650562286377
    },
    {
      "auxiliary_loss_clip": 0.01127626,
      "auxiliary_loss_mlp": 0.01025739,
      "balance_loss_clip": 1.04600096,
      "balance_loss_mlp": 1.01770473,
      "epoch": 0.5758432032706067,
      "flos": 22999024944000.0,
      "grad_norm": 2.168295974326995,
      "language_loss": 0.86428696,
      "learning_rate": 1.608386952779787e-06,
      "loss": 0.88582063,
      "num_input_tokens_seen": 103337945,
      "step": 4789,
      "time_per_iteration": 3.3623316287994385
    },
    {
      "auxiliary_loss_clip": 0.011567,
      "auxiliary_loss_mlp": 0.01026606,
      "balance_loss_clip": 1.05294979,
      "balance_loss_mlp": 1.01910138,
      "epoch": 0.5759634461612457,
      "flos": 25739727552000.0,
      "grad_norm": 1.6391115248064005,
      "language_loss": 0.7456978,
      "learning_rate": 1.6076230897624098e-06,
      "loss": 0.76753086,
      "num_input_tokens_seen": 103360150,
      "step": 4790,
      "time_per_iteration": 2.5459370613098145
    },
    {
      "auxiliary_loss_clip": 0.01165915,
      "auxiliary_loss_mlp": 0.01030199,
      "balance_loss_clip": 1.05057538,
      "balance_loss_mlp": 1.02169979,
      "epoch": 0.5760836890518848,
      "flos": 30591761639040.0,
      "grad_norm": 2.297407168241599,
      "language_loss": 0.77500349,
      "learning_rate": 1.6068592862682974e-06,
      "loss": 0.79696465,
      "num_input_tokens_seen": 103378305,
      "step": 4791,
      "time_per_iteration": 2.5299479961395264
    },
    {
      "auxiliary_loss_clip": 0.01153211,
      "auxiliary_loss_mlp": 0.01031202,
      "balance_loss_clip": 1.05213237,
      "balance_loss_mlp": 1.02332222,
      "epoch": 0.576203931942524,
      "flos": 36538963447680.0,
      "grad_norm": 1.773175782769849,
      "language_loss": 0.73385715,
      "learning_rate": 1.6060955424133187e-06,
      "loss": 0.7557013,
      "num_input_tokens_seen": 103399230,
      "step": 4792,
      "time_per_iteration": 3.5237619876861572
    },
    {
      "auxiliary_loss_clip": 0.01165923,
      "auxiliary_loss_mlp": 0.01026662,
      "balance_loss_clip": 1.05542064,
      "balance_loss_mlp": 1.01798964,
      "epoch": 0.576324174833163,
      "flos": 25516937445120.0,
      "grad_norm": 1.743717787566893,
      "language_loss": 0.89414316,
      "learning_rate": 1.6053318583133332e-06,
      "loss": 0.91606903,
      "num_input_tokens_seen": 103420100,
      "step": 4793,
      "time_per_iteration": 2.5145506858825684
    },
    {
      "auxiliary_loss_clip": 0.01164846,
      "auxiliary_loss_mlp": 0.01032138,
      "balance_loss_clip": 1.05392146,
      "balance_loss_mlp": 1.02375805,
      "epoch": 0.5764444177238021,
      "flos": 25119262995840.0,
      "grad_norm": 2.0376276274853775,
      "language_loss": 0.75491732,
      "learning_rate": 1.6045682340841907e-06,
      "loss": 0.77688712,
      "num_input_tokens_seen": 103439025,
      "step": 4794,
      "time_per_iteration": 2.5021474361419678
    },
    {
      "auxiliary_loss_clip": 0.0105585,
      "auxiliary_loss_mlp": 0.00752672,
      "balance_loss_clip": 1.02720571,
      "balance_loss_mlp": 0.99973947,
      "epoch": 0.5765646606144411,
      "flos": 62212687758720.0,
      "grad_norm": 0.752540202647363,
      "language_loss": 0.58022308,
      "learning_rate": 1.6038046698417336e-06,
      "loss": 0.59830832,
      "num_input_tokens_seen": 103499920,
      "step": 4795,
      "time_per_iteration": 3.94370436668396
    },
    {
      "auxiliary_loss_clip": 0.01164884,
      "auxiliary_loss_mlp": 0.01024995,
      "balance_loss_clip": 1.05216634,
      "balance_loss_mlp": 1.01725233,
      "epoch": 0.5766849035050803,
      "flos": 25118760205440.0,
      "grad_norm": 1.9003486803110985,
      "language_loss": 0.68532211,
      "learning_rate": 1.6030411657017919e-06,
      "loss": 0.70722091,
      "num_input_tokens_seen": 103519575,
      "step": 4796,
      "time_per_iteration": 2.5120344161987305
    },
    {
      "auxiliary_loss_clip": 0.01156231,
      "auxiliary_loss_mlp": 0.01025154,
      "balance_loss_clip": 1.05059648,
      "balance_loss_mlp": 1.0174408,
      "epoch": 0.5768051463957193,
      "flos": 15991093578240.0,
      "grad_norm": 1.761590501920671,
      "language_loss": 0.84264827,
      "learning_rate": 1.6022777217801903e-06,
      "loss": 0.86446214,
      "num_input_tokens_seen": 103536530,
      "step": 4797,
      "time_per_iteration": 3.161177158355713
    },
    {
      "auxiliary_loss_clip": 0.01136033,
      "auxiliary_loss_mlp": 0.01023375,
      "balance_loss_clip": 1.05145061,
      "balance_loss_mlp": 1.0154984,
      "epoch": 0.5769253892863584,
      "flos": 22163635359360.0,
      "grad_norm": 1.8800946414537691,
      "language_loss": 0.73611188,
      "learning_rate": 1.601514338192742e-06,
      "loss": 0.75770599,
      "num_input_tokens_seen": 103556460,
      "step": 4798,
      "time_per_iteration": 2.5353286266326904
    },
    {
      "auxiliary_loss_clip": 0.01175039,
      "auxiliary_loss_mlp": 0.01023674,
      "balance_loss_clip": 1.05287385,
      "balance_loss_mlp": 1.01635456,
      "epoch": 0.5770456321769976,
      "flos": 22856388036480.0,
      "grad_norm": 2.114391634578596,
      "language_loss": 0.71742213,
      "learning_rate": 1.6007510150552514e-06,
      "loss": 0.73940921,
      "num_input_tokens_seen": 103574520,
      "step": 4799,
      "time_per_iteration": 2.4333620071411133
    },
    {
      "auxiliary_loss_clip": 0.01168884,
      "auxiliary_loss_mlp": 0.01027255,
      "balance_loss_clip": 1.05176425,
      "balance_loss_mlp": 1.01856422,
      "epoch": 0.5771658750676366,
      "flos": 46353672489600.0,
      "grad_norm": 1.5141077532487481,
      "language_loss": 0.62086558,
      "learning_rate": 1.599987752483515e-06,
      "loss": 0.64282703,
      "num_input_tokens_seen": 103598965,
      "step": 4800,
      "time_per_iteration": 2.6748929023742676
    },
    {
      "auxiliary_loss_clip": 0.01130554,
      "auxiliary_loss_mlp": 0.01028828,
      "balance_loss_clip": 1.04643154,
      "balance_loss_mlp": 1.02079892,
      "epoch": 0.5772861179582757,
      "flos": 22159972172160.0,
      "grad_norm": 1.6100844400104255,
      "language_loss": 0.67872632,
      "learning_rate": 1.5992245505933184e-06,
      "loss": 0.70032012,
      "num_input_tokens_seen": 103618665,
      "step": 4801,
      "time_per_iteration": 2.5230495929718018
    },
    {
      "auxiliary_loss_clip": 0.01180662,
      "auxiliary_loss_mlp": 0.01028828,
      "balance_loss_clip": 1.05408692,
      "balance_loss_mlp": 1.02134156,
      "epoch": 0.5774063608489148,
      "flos": 31248926916480.0,
      "grad_norm": 2.002559568933912,
      "language_loss": 0.7088384,
      "learning_rate": 1.5984614095004388e-06,
      "loss": 0.73093331,
      "num_input_tokens_seen": 103639800,
      "step": 4802,
      "time_per_iteration": 2.4983036518096924
    },
    {
      "auxiliary_loss_clip": 0.01158881,
      "auxiliary_loss_mlp": 0.01030225,
      "balance_loss_clip": 1.05158973,
      "balance_loss_mlp": 1.02246118,
      "epoch": 0.5775266037395539,
      "flos": 22527123039360.0,
      "grad_norm": 2.80413296871043,
      "language_loss": 0.81024653,
      "learning_rate": 1.5976983293206438e-06,
      "loss": 0.83213758,
      "num_input_tokens_seen": 103655605,
      "step": 4803,
      "time_per_iteration": 2.449223518371582
    },
    {
      "auxiliary_loss_clip": 0.01145952,
      "auxiliary_loss_mlp": 0.01029542,
      "balance_loss_clip": 1.04780221,
      "balance_loss_mlp": 1.02193069,
      "epoch": 0.577646846630193,
      "flos": 21068790860160.0,
      "grad_norm": 1.8254404834279567,
      "language_loss": 0.71511841,
      "learning_rate": 1.5969353101696928e-06,
      "loss": 0.73687339,
      "num_input_tokens_seen": 103674045,
      "step": 4804,
      "time_per_iteration": 2.4780876636505127
    },
    {
      "auxiliary_loss_clip": 0.01164843,
      "auxiliary_loss_mlp": 0.01030267,
      "balance_loss_clip": 1.05211675,
      "balance_loss_mlp": 1.02312672,
      "epoch": 0.5777670895208321,
      "flos": 29714284293120.0,
      "grad_norm": 1.5720162114038148,
      "language_loss": 0.79401541,
      "learning_rate": 1.5961723521633341e-06,
      "loss": 0.81596649,
      "num_input_tokens_seen": 103695285,
      "step": 4805,
      "time_per_iteration": 2.527066230773926
    },
    {
      "auxiliary_loss_clip": 0.01145788,
      "auxiliary_loss_mlp": 0.01031655,
      "balance_loss_clip": 1.04825234,
      "balance_loss_mlp": 1.02364397,
      "epoch": 0.5778873324114712,
      "flos": 19500428344320.0,
      "grad_norm": 2.2742097060379605,
      "language_loss": 0.90755296,
      "learning_rate": 1.5954094554173097e-06,
      "loss": 0.92932737,
      "num_input_tokens_seen": 103713275,
      "step": 4806,
      "time_per_iteration": 2.4670984745025635
    },
    {
      "auxiliary_loss_clip": 0.01156293,
      "auxiliary_loss_mlp": 0.01027005,
      "balance_loss_clip": 1.05265927,
      "balance_loss_mlp": 1.01957774,
      "epoch": 0.5780075753021102,
      "flos": 14136846716160.0,
      "grad_norm": 2.099187690766042,
      "language_loss": 0.78771853,
      "learning_rate": 1.5946466200473482e-06,
      "loss": 0.80955154,
      "num_input_tokens_seen": 103731185,
      "step": 4807,
      "time_per_iteration": 2.4643871784210205
    },
    {
      "auxiliary_loss_clip": 0.011533,
      "auxiliary_loss_mlp": 0.01030682,
      "balance_loss_clip": 1.04941022,
      "balance_loss_mlp": 1.02274859,
      "epoch": 0.5781278181927494,
      "flos": 15262178883840.0,
      "grad_norm": 1.9045103062158144,
      "language_loss": 0.83419228,
      "learning_rate": 1.5938838461691723e-06,
      "loss": 0.85603213,
      "num_input_tokens_seen": 103748095,
      "step": 4808,
      "time_per_iteration": 2.4726104736328125
    },
    {
      "auxiliary_loss_clip": 0.01181256,
      "auxiliary_loss_mlp": 0.01027813,
      "balance_loss_clip": 1.05584574,
      "balance_loss_mlp": 1.02013552,
      "epoch": 0.5782480610833884,
      "flos": 16726831856640.0,
      "grad_norm": 2.6893097111012914,
      "language_loss": 0.82849038,
      "learning_rate": 1.593121133898494e-06,
      "loss": 0.85058105,
      "num_input_tokens_seen": 103765300,
      "step": 4809,
      "time_per_iteration": 2.404103994369507
    },
    {
      "auxiliary_loss_clip": 0.01170514,
      "auxiliary_loss_mlp": 0.0102681,
      "balance_loss_clip": 1.05332327,
      "balance_loss_mlp": 1.01931202,
      "epoch": 0.5783683039740275,
      "flos": 25482140144640.0,
      "grad_norm": 2.3154858738439033,
      "language_loss": 0.79078865,
      "learning_rate": 1.592358483351016e-06,
      "loss": 0.8127619,
      "num_input_tokens_seen": 103785475,
      "step": 4810,
      "time_per_iteration": 2.4880356788635254
    },
    {
      "auxiliary_loss_clip": 0.01159595,
      "auxiliary_loss_mlp": 0.01024556,
      "balance_loss_clip": 1.05054951,
      "balance_loss_mlp": 1.01746881,
      "epoch": 0.5784885468646667,
      "flos": 18405835240320.0,
      "grad_norm": 1.9558081568064312,
      "language_loss": 0.72437102,
      "learning_rate": 1.5915958946424326e-06,
      "loss": 0.74621248,
      "num_input_tokens_seen": 103804160,
      "step": 4811,
      "time_per_iteration": 2.433809518814087
    },
    {
      "auxiliary_loss_clip": 0.01138734,
      "auxiliary_loss_mlp": 0.00763159,
      "balance_loss_clip": 1.05050564,
      "balance_loss_mlp": 1.00041556,
      "epoch": 0.5786087897553057,
      "flos": 46100717936640.0,
      "grad_norm": 1.6131716401641405,
      "language_loss": 0.74637413,
      "learning_rate": 1.5908333678884271e-06,
      "loss": 0.76539308,
      "num_input_tokens_seen": 103830580,
      "step": 4812,
      "time_per_iteration": 2.761343002319336
    },
    {
      "auxiliary_loss_clip": 0.01162204,
      "auxiliary_loss_mlp": 0.01027746,
      "balance_loss_clip": 1.05179417,
      "balance_loss_mlp": 1.02013421,
      "epoch": 0.5787290326459448,
      "flos": 12385950261120.0,
      "grad_norm": 2.60731347072189,
      "language_loss": 0.73593378,
      "learning_rate": 1.5900709032046743e-06,
      "loss": 0.75783324,
      "num_input_tokens_seen": 103848655,
      "step": 4813,
      "time_per_iteration": 2.429922580718994
    },
    {
      "auxiliary_loss_clip": 0.01147975,
      "auxiliary_loss_mlp": 0.01027722,
      "balance_loss_clip": 1.0530237,
      "balance_loss_mlp": 1.01994109,
      "epoch": 0.5788492755365839,
      "flos": 23290332243840.0,
      "grad_norm": 2.2008917073525547,
      "language_loss": 0.78439766,
      "learning_rate": 1.5893085007068391e-06,
      "loss": 0.80615461,
      "num_input_tokens_seen": 103866215,
      "step": 4814,
      "time_per_iteration": 2.4926669597625732
    },
    {
      "auxiliary_loss_clip": 0.01138574,
      "auxiliary_loss_mlp": 0.01025892,
      "balance_loss_clip": 1.04767525,
      "balance_loss_mlp": 1.01736546,
      "epoch": 0.578969518427223,
      "flos": 24061047390720.0,
      "grad_norm": 1.7970503846266523,
      "language_loss": 0.70889968,
      "learning_rate": 1.5885461605105786e-06,
      "loss": 0.73054433,
      "num_input_tokens_seen": 103887815,
      "step": 4815,
      "time_per_iteration": 2.5430924892425537
    },
    {
      "auxiliary_loss_clip": 0.01154508,
      "auxiliary_loss_mlp": 0.01028425,
      "balance_loss_clip": 1.0535754,
      "balance_loss_mlp": 1.02012205,
      "epoch": 0.579089761317862,
      "flos": 21871825269120.0,
      "grad_norm": 1.792600951521911,
      "language_loss": 0.76843095,
      "learning_rate": 1.5877838827315375e-06,
      "loss": 0.79026026,
      "num_input_tokens_seen": 103906360,
      "step": 4816,
      "time_per_iteration": 3.320629358291626
    },
    {
      "auxiliary_loss_clip": 0.01178697,
      "auxiliary_loss_mlp": 0.01030352,
      "balance_loss_clip": 1.05510771,
      "balance_loss_mlp": 1.02273405,
      "epoch": 0.5792100042085012,
      "flos": 22929681738240.0,
      "grad_norm": 1.6887431972782252,
      "language_loss": 0.70308185,
      "learning_rate": 1.587021667485355e-06,
      "loss": 0.72517234,
      "num_input_tokens_seen": 103925730,
      "step": 4817,
      "time_per_iteration": 2.450834035873413
    },
    {
      "auxiliary_loss_clip": 0.01150962,
      "auxiliary_loss_mlp": 0.01022438,
      "balance_loss_clip": 1.04798841,
      "balance_loss_mlp": 1.0148263,
      "epoch": 0.5793302470991403,
      "flos": 21470056669440.0,
      "grad_norm": 1.7952666705475375,
      "language_loss": 0.78592384,
      "learning_rate": 1.5862595148876559e-06,
      "loss": 0.80765784,
      "num_input_tokens_seen": 103945835,
      "step": 4818,
      "time_per_iteration": 2.5076398849487305
    },
    {
      "auxiliary_loss_clip": 0.01125559,
      "auxiliary_loss_mlp": 0.01028747,
      "balance_loss_clip": 1.0495286,
      "balance_loss_mlp": 1.02084947,
      "epoch": 0.5794504899897793,
      "flos": 12711013367040.0,
      "grad_norm": 2.2706713707006374,
      "language_loss": 0.76332724,
      "learning_rate": 1.58549742505406e-06,
      "loss": 0.78487033,
      "num_input_tokens_seen": 103960580,
      "step": 4819,
      "time_per_iteration": 3.3719420433044434
    },
    {
      "auxiliary_loss_clip": 0.01178578,
      "auxiliary_loss_mlp": 0.01026739,
      "balance_loss_clip": 1.05333555,
      "balance_loss_mlp": 1.01927006,
      "epoch": 0.5795707328804185,
      "flos": 14867054300160.0,
      "grad_norm": 2.5759263894171474,
      "language_loss": 0.75158703,
      "learning_rate": 1.5847353981001747e-06,
      "loss": 0.77364022,
      "num_input_tokens_seen": 103977760,
      "step": 4820,
      "time_per_iteration": 2.443037271499634
    },
    {
      "auxiliary_loss_clip": 0.0114201,
      "auxiliary_loss_mlp": 0.01029647,
      "balance_loss_clip": 1.04654431,
      "balance_loss_mlp": 1.02160025,
      "epoch": 0.5796909757710575,
      "flos": 36430046432640.0,
      "grad_norm": 1.7062196958755085,
      "language_loss": 0.69907612,
      "learning_rate": 1.5839734341415993e-06,
      "loss": 0.72079271,
      "num_input_tokens_seen": 103999960,
      "step": 4821,
      "time_per_iteration": 3.46303653717041
    },
    {
      "auxiliary_loss_clip": 0.01158311,
      "auxiliary_loss_mlp": 0.01026743,
      "balance_loss_clip": 1.05479527,
      "balance_loss_mlp": 1.019418,
      "epoch": 0.5798112186616966,
      "flos": 23039891642880.0,
      "grad_norm": 7.62861136487302,
      "language_loss": 0.76747185,
      "learning_rate": 1.5832115332939238e-06,
      "loss": 0.78932238,
      "num_input_tokens_seen": 104018400,
      "step": 4822,
      "time_per_iteration": 2.4939842224121094
    },
    {
      "auxiliary_loss_clip": 0.01165624,
      "auxiliary_loss_mlp": 0.01031723,
      "balance_loss_clip": 1.05331111,
      "balance_loss_mlp": 1.0240159,
      "epoch": 0.5799314615523358,
      "flos": 16652604401280.0,
      "grad_norm": 2.0397647987742,
      "language_loss": 0.74556226,
      "learning_rate": 1.5824496956727272e-06,
      "loss": 0.76753575,
      "num_input_tokens_seen": 104035605,
      "step": 4823,
      "time_per_iteration": 2.455172538757324
    },
    {
      "auxiliary_loss_clip": 0.01150368,
      "auxiliary_loss_mlp": 0.01025473,
      "balance_loss_clip": 1.05048776,
      "balance_loss_mlp": 1.01774251,
      "epoch": 0.5800517044429748,
      "flos": 20485673470080.0,
      "grad_norm": 1.6543905942696686,
      "language_loss": 0.73162234,
      "learning_rate": 1.5816879213935797e-06,
      "loss": 0.75338078,
      "num_input_tokens_seen": 104054415,
      "step": 4824,
      "time_per_iteration": 3.2578377723693848
    },
    {
      "auxiliary_loss_clip": 0.01159574,
      "auxiliary_loss_mlp": 0.01029769,
      "balance_loss_clip": 1.05142653,
      "balance_loss_mlp": 1.0224911,
      "epoch": 0.5801719473336139,
      "flos": 31538258968320.0,
      "grad_norm": 1.6112439783685095,
      "language_loss": 0.79972327,
      "learning_rate": 1.5809262105720416e-06,
      "loss": 0.82161665,
      "num_input_tokens_seen": 104075455,
      "step": 4825,
      "time_per_iteration": 2.543463945388794
    },
    {
      "auxiliary_loss_clip": 0.01175541,
      "auxiliary_loss_mlp": 0.01031363,
      "balance_loss_clip": 1.05400872,
      "balance_loss_mlp": 1.02417207,
      "epoch": 0.580292190224253,
      "flos": 20375966355840.0,
      "grad_norm": 1.508661381725362,
      "language_loss": 0.79546809,
      "learning_rate": 1.5801645633236644e-06,
      "loss": 0.81753707,
      "num_input_tokens_seen": 104096440,
      "step": 4826,
      "time_per_iteration": 2.461289167404175
    },
    {
      "auxiliary_loss_clip": 0.01142754,
      "auxiliary_loss_mlp": 0.01030793,
      "balance_loss_clip": 1.04756069,
      "balance_loss_mlp": 1.02289271,
      "epoch": 0.5804124331148921,
      "flos": 26615373304320.0,
      "grad_norm": 1.9444833460267235,
      "language_loss": 0.77387327,
      "learning_rate": 1.579402979763989e-06,
      "loss": 0.79560876,
      "num_input_tokens_seen": 104116775,
      "step": 4827,
      "time_per_iteration": 2.5271782875061035
    },
    {
      "auxiliary_loss_clip": 0.01121406,
      "auxiliary_loss_mlp": 0.01024271,
      "balance_loss_clip": 1.05055141,
      "balance_loss_mlp": 1.0168587,
      "epoch": 0.5805326760055312,
      "flos": 13478496289920.0,
      "grad_norm": 2.1982238170008617,
      "language_loss": 0.81359017,
      "learning_rate": 1.578641460008548e-06,
      "loss": 0.83504689,
      "num_input_tokens_seen": 104134510,
      "step": 4828,
      "time_per_iteration": 2.5493481159210205
    },
    {
      "auxiliary_loss_clip": 0.01162447,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.05247688,
      "balance_loss_mlp": 1.01814032,
      "epoch": 0.5806529188961702,
      "flos": 12091374823680.0,
      "grad_norm": 2.1794946307205882,
      "language_loss": 0.67927456,
      "learning_rate": 1.5778800041728613e-06,
      "loss": 0.70116264,
      "num_input_tokens_seen": 104150800,
      "step": 4829,
      "time_per_iteration": 2.423433303833008
    },
    {
      "auxiliary_loss_clip": 0.01159302,
      "auxiliary_loss_mlp": 0.01021144,
      "balance_loss_clip": 1.05304146,
      "balance_loss_mlp": 1.01371145,
      "epoch": 0.5807731617868094,
      "flos": 26214107495040.0,
      "grad_norm": 1.4959524687369703,
      "language_loss": 0.66090465,
      "learning_rate": 1.577118612372443e-06,
      "loss": 0.6827091,
      "num_input_tokens_seen": 104172640,
      "step": 4830,
      "time_per_iteration": 2.5437004566192627
    },
    {
      "auxiliary_loss_clip": 0.01142239,
      "auxiliary_loss_mlp": 0.00762631,
      "balance_loss_clip": 1.04601967,
      "balance_loss_mlp": 1.0005002,
      "epoch": 0.5808934046774484,
      "flos": 37962139190400.0,
      "grad_norm": 1.6426443878281551,
      "language_loss": 0.70309186,
      "learning_rate": 1.5763572847227943e-06,
      "loss": 0.72214061,
      "num_input_tokens_seen": 104193525,
      "step": 4831,
      "time_per_iteration": 2.665447950363159
    },
    {
      "auxiliary_loss_clip": 0.01160254,
      "auxiliary_loss_mlp": 0.01025853,
      "balance_loss_clip": 1.0496273,
      "balance_loss_mlp": 1.01862288,
      "epoch": 0.5810136475680875,
      "flos": 20485853038080.0,
      "grad_norm": 1.9382805648751418,
      "language_loss": 0.81237346,
      "learning_rate": 1.5755960213394091e-06,
      "loss": 0.8342346,
      "num_input_tokens_seen": 104210625,
      "step": 4832,
      "time_per_iteration": 2.4658761024475098
    },
    {
      "auxiliary_loss_clip": 0.01136401,
      "auxiliary_loss_mlp": 0.01027521,
      "balance_loss_clip": 1.04745483,
      "balance_loss_mlp": 1.02035367,
      "epoch": 0.5811338904587267,
      "flos": 17530153574400.0,
      "grad_norm": 1.7978367172209813,
      "language_loss": 0.78827655,
      "learning_rate": 1.5748348223377703e-06,
      "loss": 0.80991578,
      "num_input_tokens_seen": 104228180,
      "step": 4833,
      "time_per_iteration": 2.5076098442077637
    },
    {
      "auxiliary_loss_clip": 0.01146458,
      "auxiliary_loss_mlp": 0.01027632,
      "balance_loss_clip": 1.05074251,
      "balance_loss_mlp": 1.02035737,
      "epoch": 0.5812541333493657,
      "flos": 19458017360640.0,
      "grad_norm": 1.6187735851716132,
      "language_loss": 0.77801555,
      "learning_rate": 1.5740736878333507e-06,
      "loss": 0.79975647,
      "num_input_tokens_seen": 104246020,
      "step": 4834,
      "time_per_iteration": 2.4869067668914795
    },
    {
      "auxiliary_loss_clip": 0.01152377,
      "auxiliary_loss_mlp": 0.01024147,
      "balance_loss_clip": 1.05032754,
      "balance_loss_mlp": 1.01639867,
      "epoch": 0.5813743762400048,
      "flos": 20594949621120.0,
      "grad_norm": 2.767879829741655,
      "language_loss": 0.77816725,
      "learning_rate": 1.5733126179416143e-06,
      "loss": 0.79993248,
      "num_input_tokens_seen": 104260505,
      "step": 4835,
      "time_per_iteration": 2.5152225494384766
    },
    {
      "auxiliary_loss_clip": 0.0116264,
      "auxiliary_loss_mlp": 0.01025491,
      "balance_loss_clip": 1.05145264,
      "balance_loss_mlp": 1.01820779,
      "epoch": 0.5814946191306439,
      "flos": 33178227246720.0,
      "grad_norm": 2.749408754252559,
      "language_loss": 0.72580385,
      "learning_rate": 1.5725516127780137e-06,
      "loss": 0.74768519,
      "num_input_tokens_seen": 104282640,
      "step": 4836,
      "time_per_iteration": 2.57684326171875
    },
    {
      "auxiliary_loss_clip": 0.01168099,
      "auxiliary_loss_mlp": 0.0103019,
      "balance_loss_clip": 1.04977918,
      "balance_loss_mlp": 1.0219171,
      "epoch": 0.581614862021283,
      "flos": 16143283503360.0,
      "grad_norm": 2.068136660915475,
      "language_loss": 0.88011634,
      "learning_rate": 1.5717906724579943e-06,
      "loss": 0.90209925,
      "num_input_tokens_seen": 104299700,
      "step": 4837,
      "time_per_iteration": 2.4465553760528564
    },
    {
      "auxiliary_loss_clip": 0.01144227,
      "auxiliary_loss_mlp": 0.01024046,
      "balance_loss_clip": 1.05093455,
      "balance_loss_mlp": 1.01672626,
      "epoch": 0.581735104911922,
      "flos": 33802642298880.0,
      "grad_norm": 2.022580939214022,
      "language_loss": 0.68180704,
      "learning_rate": 1.571029797096989e-06,
      "loss": 0.70348978,
      "num_input_tokens_seen": 104320805,
      "step": 4838,
      "time_per_iteration": 2.6637215614318848
    },
    {
      "auxiliary_loss_clip": 0.0117482,
      "auxiliary_loss_mlp": 0.01029864,
      "balance_loss_clip": 1.05176306,
      "balance_loss_mlp": 1.02255654,
      "epoch": 0.5818553478025612,
      "flos": 23331163029120.0,
      "grad_norm": 1.7913687619389325,
      "language_loss": 0.78718591,
      "learning_rate": 1.570268986810423e-06,
      "loss": 0.80923277,
      "num_input_tokens_seen": 104340700,
      "step": 4839,
      "time_per_iteration": 2.4550280570983887
    },
    {
      "auxiliary_loss_clip": 0.01145935,
      "auxiliary_loss_mlp": 0.01026746,
      "balance_loss_clip": 1.04870248,
      "balance_loss_mlp": 1.01996565,
      "epoch": 0.5819755906932003,
      "flos": 20996143603200.0,
      "grad_norm": 1.985929185702165,
      "language_loss": 0.74562323,
      "learning_rate": 1.5695082417137096e-06,
      "loss": 0.76735002,
      "num_input_tokens_seen": 104358575,
      "step": 4840,
      "time_per_iteration": 2.4990272521972656
    },
    {
      "auxiliary_loss_clip": 0.01144232,
      "auxiliary_loss_mlp": 0.01025733,
      "balance_loss_clip": 1.04634285,
      "balance_loss_mlp": 1.01869965,
      "epoch": 0.5820958335838393,
      "flos": 21431668008960.0,
      "grad_norm": 1.720241073364023,
      "language_loss": 0.75308704,
      "learning_rate": 1.5687475619222539e-06,
      "loss": 0.77478671,
      "num_input_tokens_seen": 104378530,
      "step": 4841,
      "time_per_iteration": 2.547492504119873
    },
    {
      "auxiliary_loss_clip": 0.01141072,
      "auxiliary_loss_mlp": 0.01025842,
      "balance_loss_clip": 1.04540944,
      "balance_loss_mlp": 1.01801324,
      "epoch": 0.5822160764744785,
      "flos": 17967473660160.0,
      "grad_norm": 2.3256871300788577,
      "language_loss": 0.73508853,
      "learning_rate": 1.5679869475514496e-06,
      "loss": 0.75675774,
      "num_input_tokens_seen": 104395465,
      "step": 4842,
      "time_per_iteration": 3.2962448596954346
    },
    {
      "auxiliary_loss_clip": 0.0116497,
      "auxiliary_loss_mlp": 0.01030106,
      "balance_loss_clip": 1.05276275,
      "balance_loss_mlp": 1.02196956,
      "epoch": 0.5823363193651175,
      "flos": 23033858158080.0,
      "grad_norm": 2.1838748952135205,
      "language_loss": 0.81499696,
      "learning_rate": 1.567226398716682e-06,
      "loss": 0.83694774,
      "num_input_tokens_seen": 104415380,
      "step": 4843,
      "time_per_iteration": 2.5237386226654053
    },
    {
      "auxiliary_loss_clip": 0.01154551,
      "auxiliary_loss_mlp": 0.01023428,
      "balance_loss_clip": 1.04970193,
      "balance_loss_mlp": 1.01489604,
      "epoch": 0.5824565622557566,
      "flos": 32891840110080.0,
      "grad_norm": 1.8471673957133083,
      "language_loss": 0.6191833,
      "learning_rate": 1.566465915533326e-06,
      "loss": 0.64096314,
      "num_input_tokens_seen": 104437410,
      "step": 4844,
      "time_per_iteration": 2.6162548065185547
    },
    {
      "auxiliary_loss_clip": 0.01160571,
      "auxiliary_loss_mlp": 0.01025233,
      "balance_loss_clip": 1.0516479,
      "balance_loss_mlp": 1.01764512,
      "epoch": 0.5825768051463958,
      "flos": 22229674513920.0,
      "grad_norm": 1.9538527415128801,
      "language_loss": 0.88267934,
      "learning_rate": 1.5657054981167458e-06,
      "loss": 0.90453744,
      "num_input_tokens_seen": 104456305,
      "step": 4845,
      "time_per_iteration": 3.3563575744628906
    },
    {
      "auxiliary_loss_clip": 0.01159165,
      "auxiliary_loss_mlp": 0.0102485,
      "balance_loss_clip": 1.0512867,
      "balance_loss_mlp": 1.01805472,
      "epoch": 0.5826970480370348,
      "flos": 28001561016960.0,
      "grad_norm": 2.000017001513208,
      "language_loss": 0.67855966,
      "learning_rate": 1.5649451465822965e-06,
      "loss": 0.70039982,
      "num_input_tokens_seen": 104477695,
      "step": 4846,
      "time_per_iteration": 2.5291800498962402
    },
    {
      "auxiliary_loss_clip": 0.01119701,
      "auxiliary_loss_mlp": 0.01028635,
      "balance_loss_clip": 1.04843366,
      "balance_loss_mlp": 1.02113056,
      "epoch": 0.5828172909276739,
      "flos": 17858053854720.0,
      "grad_norm": 1.7766887218450944,
      "language_loss": 0.83740079,
      "learning_rate": 1.5641848610453218e-06,
      "loss": 0.85888422,
      "num_input_tokens_seen": 104496355,
      "step": 4847,
      "time_per_iteration": 2.5439999103546143
    },
    {
      "auxiliary_loss_clip": 0.01159178,
      "auxiliary_loss_mlp": 0.01025862,
      "balance_loss_clip": 1.05213666,
      "balance_loss_mlp": 1.01814342,
      "epoch": 0.582937533818313,
      "flos": 19865244827520.0,
      "grad_norm": 2.107255172201198,
      "language_loss": 0.86087084,
      "learning_rate": 1.563424641621158e-06,
      "loss": 0.8827213,
      "num_input_tokens_seen": 104515535,
      "step": 4848,
      "time_per_iteration": 3.325834274291992
    },
    {
      "auxiliary_loss_clip": 0.01152505,
      "auxiliary_loss_mlp": 0.01025521,
      "balance_loss_clip": 1.04987502,
      "balance_loss_mlp": 1.01762342,
      "epoch": 0.5830577767089521,
      "flos": 26870734068480.0,
      "grad_norm": 1.94068994315908,
      "language_loss": 0.69884402,
      "learning_rate": 1.5626644884251282e-06,
      "loss": 0.72062427,
      "num_input_tokens_seen": 104535055,
      "step": 4849,
      "time_per_iteration": 2.5377447605133057
    },
    {
      "auxiliary_loss_clip": 0.01176486,
      "auxiliary_loss_mlp": 0.01023693,
      "balance_loss_clip": 1.05314493,
      "balance_loss_mlp": 1.01675177,
      "epoch": 0.5831780195995911,
      "flos": 25298205575040.0,
      "grad_norm": 1.5909733309391125,
      "language_loss": 0.88058019,
      "learning_rate": 1.5619044015725488e-06,
      "loss": 0.90258199,
      "num_input_tokens_seen": 104554745,
      "step": 4850,
      "time_per_iteration": 2.465125322341919
    },
    {
      "auxiliary_loss_clip": 0.01184149,
      "auxiliary_loss_mlp": 0.01030931,
      "balance_loss_clip": 1.05670619,
      "balance_loss_mlp": 1.02236032,
      "epoch": 0.5832982624902303,
      "flos": 14756988049920.0,
      "grad_norm": 3.1330635107007696,
      "language_loss": 0.86991805,
      "learning_rate": 1.5611443811787224e-06,
      "loss": 0.89206892,
      "num_input_tokens_seen": 104568870,
      "step": 4851,
      "time_per_iteration": 3.156158924102783
    },
    {
      "auxiliary_loss_clip": 0.01160862,
      "auxiliary_loss_mlp": 0.01023479,
      "balance_loss_clip": 1.05103397,
      "balance_loss_mlp": 1.01597786,
      "epoch": 0.5834185053808694,
      "flos": 20444555376000.0,
      "grad_norm": 2.2519021943582413,
      "language_loss": 0.69193029,
      "learning_rate": 1.560384427358945e-06,
      "loss": 0.71377373,
      "num_input_tokens_seen": 104588415,
      "step": 4852,
      "time_per_iteration": 2.4471395015716553
    },
    {
      "auxiliary_loss_clip": 0.01140894,
      "auxiliary_loss_mlp": 0.01028994,
      "balance_loss_clip": 1.04445791,
      "balance_loss_mlp": 1.02146542,
      "epoch": 0.5835387482715084,
      "flos": 27200394115200.0,
      "grad_norm": 2.463433953067411,
      "language_loss": 0.73123449,
      "learning_rate": 1.5596245402284998e-06,
      "loss": 0.75293338,
      "num_input_tokens_seen": 104611940,
      "step": 4853,
      "time_per_iteration": 2.5585215091705322
    },
    {
      "auxiliary_loss_clip": 0.01166485,
      "auxiliary_loss_mlp": 0.01027264,
      "balance_loss_clip": 1.05385637,
      "balance_loss_mlp": 1.01940215,
      "epoch": 0.5836589911621476,
      "flos": 16654615562880.0,
      "grad_norm": 1.6673030764860544,
      "language_loss": 0.8178246,
      "learning_rate": 1.5588647199026619e-06,
      "loss": 0.83976197,
      "num_input_tokens_seen": 104629675,
      "step": 4854,
      "time_per_iteration": 2.441371440887451
    },
    {
      "auxiliary_loss_clip": 0.01183148,
      "auxiliary_loss_mlp": 0.01024559,
      "balance_loss_clip": 1.05733562,
      "balance_loss_mlp": 1.01665592,
      "epoch": 0.5837792340527866,
      "flos": 20446817932800.0,
      "grad_norm": 2.0831662708254886,
      "language_loss": 0.87713569,
      "learning_rate": 1.5581049664966956e-06,
      "loss": 0.89921272,
      "num_input_tokens_seen": 104647435,
      "step": 4855,
      "time_per_iteration": 2.4172589778900146
    },
    {
      "auxiliary_loss_clip": 0.01032293,
      "auxiliary_loss_mlp": 0.01003257,
      "balance_loss_clip": 1.02152658,
      "balance_loss_mlp": 1.00216007,
      "epoch": 0.5838994769434257,
      "flos": 65995480765440.0,
      "grad_norm": 0.9961906891607405,
      "language_loss": 0.65095389,
      "learning_rate": 1.5573452801258545e-06,
      "loss": 0.67130935,
      "num_input_tokens_seen": 104694605,
      "step": 4856,
      "time_per_iteration": 2.949613332748413
    },
    {
      "auxiliary_loss_clip": 0.01168817,
      "auxiliary_loss_mlp": 0.01034306,
      "balance_loss_clip": 1.05254018,
      "balance_loss_mlp": 1.02656889,
      "epoch": 0.5840197198340649,
      "flos": 21470523546240.0,
      "grad_norm": 2.360438211936364,
      "language_loss": 0.63235027,
      "learning_rate": 1.5565856609053824e-06,
      "loss": 0.65438151,
      "num_input_tokens_seen": 104713400,
      "step": 4857,
      "time_per_iteration": 2.4557480812072754
    },
    {
      "auxiliary_loss_clip": 0.01177818,
      "auxiliary_loss_mlp": 0.01025229,
      "balance_loss_clip": 1.05423403,
      "balance_loss_mlp": 1.01728415,
      "epoch": 0.5841399627247039,
      "flos": 19135144984320.0,
      "grad_norm": 2.0485681925859223,
      "language_loss": 0.80327988,
      "learning_rate": 1.5558261089505127e-06,
      "loss": 0.82531041,
      "num_input_tokens_seen": 104732130,
      "step": 4858,
      "time_per_iteration": 2.3986682891845703
    },
    {
      "auxiliary_loss_clip": 0.01164291,
      "auxiliary_loss_mlp": 0.01026393,
      "balance_loss_clip": 1.05311632,
      "balance_loss_mlp": 1.01891589,
      "epoch": 0.584260205615343,
      "flos": 26425692558720.0,
      "grad_norm": 2.6940012537396596,
      "language_loss": 0.80158663,
      "learning_rate": 1.5550666243764697e-06,
      "loss": 0.82349348,
      "num_input_tokens_seen": 104750290,
      "step": 4859,
      "time_per_iteration": 2.508812427520752
    },
    {
      "auxiliary_loss_clip": 0.01162938,
      "auxiliary_loss_mlp": 0.01029103,
      "balance_loss_clip": 1.05108821,
      "balance_loss_mlp": 1.02162242,
      "epoch": 0.584380448505982,
      "flos": 13881809174400.0,
      "grad_norm": 2.038169763437529,
      "language_loss": 0.77519751,
      "learning_rate": 1.554307207298465e-06,
      "loss": 0.79711789,
      "num_input_tokens_seen": 104768550,
      "step": 4860,
      "time_per_iteration": 2.437427043914795
    },
    {
      "auxiliary_loss_clip": 0.01182664,
      "auxiliary_loss_mlp": 0.01031477,
      "balance_loss_clip": 1.05650914,
      "balance_loss_mlp": 1.02346015,
      "epoch": 0.5845006913966212,
      "flos": 21543709507200.0,
      "grad_norm": 1.860142245788919,
      "language_loss": 0.78734183,
      "learning_rate": 1.553547857831704e-06,
      "loss": 0.80948329,
      "num_input_tokens_seen": 104785060,
      "step": 4861,
      "time_per_iteration": 2.440162181854248
    },
    {
      "auxiliary_loss_clip": 0.0108054,
      "auxiliary_loss_mlp": 0.01000403,
      "balance_loss_clip": 1.02137733,
      "balance_loss_mlp": 0.99944383,
      "epoch": 0.5846209342872603,
      "flos": 58375452712320.0,
      "grad_norm": 0.8851175172939759,
      "language_loss": 0.64188695,
      "learning_rate": 1.5527885760913771e-06,
      "loss": 0.66269636,
      "num_input_tokens_seen": 104834950,
      "step": 4862,
      "time_per_iteration": 2.851132869720459
    },
    {
      "auxiliary_loss_clip": 0.01147899,
      "auxiliary_loss_mlp": 0.01026378,
      "balance_loss_clip": 1.05136418,
      "balance_loss_mlp": 1.01900458,
      "epoch": 0.5847411771778993,
      "flos": 18588045957120.0,
      "grad_norm": 1.5871219602991582,
      "language_loss": 0.76283348,
      "learning_rate": 1.552029362192668e-06,
      "loss": 0.78457618,
      "num_input_tokens_seen": 104854210,
      "step": 4863,
      "time_per_iteration": 2.48380184173584
    },
    {
      "auxiliary_loss_clip": 0.01129701,
      "auxiliary_loss_mlp": 0.01031475,
      "balance_loss_clip": 1.04677558,
      "balance_loss_mlp": 1.02391434,
      "epoch": 0.5848614200685385,
      "flos": 24240780069120.0,
      "grad_norm": 1.8966944328291402,
      "language_loss": 0.72216862,
      "learning_rate": 1.5512702162507478e-06,
      "loss": 0.74378037,
      "num_input_tokens_seen": 104874525,
      "step": 4864,
      "time_per_iteration": 2.5531554222106934
    },
    {
      "auxiliary_loss_clip": 0.01059597,
      "auxiliary_loss_mlp": 0.01001672,
      "balance_loss_clip": 1.02088821,
      "balance_loss_mlp": 1.00066471,
      "epoch": 0.5849816629591775,
      "flos": 71660245933440.0,
      "grad_norm": 1.1317534729688938,
      "language_loss": 0.55839682,
      "learning_rate": 1.5505111383807792e-06,
      "loss": 0.57900953,
      "num_input_tokens_seen": 104937195,
      "step": 4865,
      "time_per_iteration": 3.136577606201172
    },
    {
      "auxiliary_loss_clip": 0.01123058,
      "auxiliary_loss_mlp": 0.01024534,
      "balance_loss_clip": 1.0455687,
      "balance_loss_mlp": 1.01750982,
      "epoch": 0.5851019058498166,
      "flos": 23802095266560.0,
      "grad_norm": 1.7223347079823759,
      "language_loss": 0.80646664,
      "learning_rate": 1.5497521286979138e-06,
      "loss": 0.82794255,
      "num_input_tokens_seen": 104957435,
      "step": 4866,
      "time_per_iteration": 2.5836257934570312
    },
    {
      "auxiliary_loss_clip": 0.01138844,
      "auxiliary_loss_mlp": 0.01029247,
      "balance_loss_clip": 1.04933167,
      "balance_loss_mlp": 1.02107573,
      "epoch": 0.5852221487404557,
      "flos": 24388516707840.0,
      "grad_norm": 1.9035069135276117,
      "language_loss": 0.74253416,
      "learning_rate": 1.5489931873172927e-06,
      "loss": 0.76421505,
      "num_input_tokens_seen": 104978755,
      "step": 4867,
      "time_per_iteration": 2.568427085876465
    },
    {
      "auxiliary_loss_clip": 0.01087025,
      "auxiliary_loss_mlp": 0.01028749,
      "balance_loss_clip": 1.03933191,
      "balance_loss_mlp": 1.02133465,
      "epoch": 0.5853423916310948,
      "flos": 27271425260160.0,
      "grad_norm": 1.6779308275106,
      "language_loss": 0.79110068,
      "learning_rate": 1.5482343143540467e-06,
      "loss": 0.81225848,
      "num_input_tokens_seen": 105000020,
      "step": 4868,
      "time_per_iteration": 2.625805139541626
    },
    {
      "auxiliary_loss_clip": 0.01135247,
      "auxiliary_loss_mlp": 0.00761814,
      "balance_loss_clip": 1.04805756,
      "balance_loss_mlp": 1.00038552,
      "epoch": 0.5854626345217339,
      "flos": 11983786611840.0,
      "grad_norm": 1.926357160625032,
      "language_loss": 0.8255769,
      "learning_rate": 1.547475509923295e-06,
      "loss": 0.84454751,
      "num_input_tokens_seen": 105017060,
      "step": 4869,
      "time_per_iteration": 3.3666861057281494
    },
    {
      "auxiliary_loss_clip": 0.01040175,
      "auxiliary_loss_mlp": 0.01002552,
      "balance_loss_clip": 1.01994157,
      "balance_loss_mlp": 1.0015744,
      "epoch": 0.585582877412373,
      "flos": 64342335173760.0,
      "grad_norm": 0.7328504637338263,
      "language_loss": 0.56075072,
      "learning_rate": 1.5467167741401495e-06,
      "loss": 0.58117795,
      "num_input_tokens_seen": 105078540,
      "step": 4870,
      "time_per_iteration": 3.132795572280884
    },
    {
      "auxiliary_loss_clip": 0.01143404,
      "auxiliary_loss_mlp": 0.01028324,
      "balance_loss_clip": 1.04508471,
      "balance_loss_mlp": 1.02055144,
      "epoch": 0.5857031203030121,
      "flos": 17011926103680.0,
      "grad_norm": 2.673943306379921,
      "language_loss": 0.71054357,
      "learning_rate": 1.5459581071197083e-06,
      "loss": 0.73226088,
      "num_input_tokens_seen": 105094200,
      "step": 4871,
      "time_per_iteration": 2.476003646850586
    },
    {
      "auxiliary_loss_clip": 0.01168802,
      "auxiliary_loss_mlp": 0.01022287,
      "balance_loss_clip": 1.05585456,
      "balance_loss_mlp": 1.01464605,
      "epoch": 0.5858233631936511,
      "flos": 20885682303360.0,
      "grad_norm": 2.0639407047433993,
      "language_loss": 0.83156538,
      "learning_rate": 1.5451995089770624e-06,
      "loss": 0.85347629,
      "num_input_tokens_seen": 105113985,
      "step": 4872,
      "time_per_iteration": 3.325416088104248
    },
    {
      "auxiliary_loss_clip": 0.01175918,
      "auxiliary_loss_mlp": 0.01024246,
      "balance_loss_clip": 1.05331874,
      "balance_loss_mlp": 1.01719499,
      "epoch": 0.5859436060842903,
      "flos": 23191902000000.0,
      "grad_norm": 9.177022466088836,
      "language_loss": 0.7179935,
      "learning_rate": 1.5444409798272885e-06,
      "loss": 0.73999512,
      "num_input_tokens_seen": 105138075,
      "step": 4873,
      "time_per_iteration": 2.550305128097534
    },
    {
      "auxiliary_loss_clip": 0.01136295,
      "auxiliary_loss_mlp": 0.01031032,
      "balance_loss_clip": 1.04809999,
      "balance_loss_mlp": 1.02355802,
      "epoch": 0.5860638489749294,
      "flos": 22492648961280.0,
      "grad_norm": 2.7690078253871917,
      "language_loss": 0.80556107,
      "learning_rate": 1.543682519785456e-06,
      "loss": 0.82723439,
      "num_input_tokens_seen": 105156555,
      "step": 4874,
      "time_per_iteration": 2.590604066848755
    },
    {
      "auxiliary_loss_clip": 0.01148519,
      "auxiliary_loss_mlp": 0.01029442,
      "balance_loss_clip": 1.05019605,
      "balance_loss_mlp": 1.02231646,
      "epoch": 0.5861840918655684,
      "flos": 17566243764480.0,
      "grad_norm": 2.4416744322458652,
      "language_loss": 0.80483055,
      "learning_rate": 1.5429241289666219e-06,
      "loss": 0.82661021,
      "num_input_tokens_seen": 105174055,
      "step": 4875,
      "time_per_iteration": 3.315791130065918
    },
    {
      "auxiliary_loss_clip": 0.01140726,
      "auxiliary_loss_mlp": 0.01026856,
      "balance_loss_clip": 1.04901659,
      "balance_loss_mlp": 1.01955175,
      "epoch": 0.5863043347562076,
      "flos": 25556152118400.0,
      "grad_norm": 1.9229775416406538,
      "language_loss": 0.69987988,
      "learning_rate": 1.5421658074858342e-06,
      "loss": 0.72155571,
      "num_input_tokens_seen": 105192160,
      "step": 4876,
      "time_per_iteration": 2.535771131515503
    },
    {
      "auxiliary_loss_clip": 0.01145201,
      "auxiliary_loss_mlp": 0.01029946,
      "balance_loss_clip": 1.05000031,
      "balance_loss_mlp": 1.02225113,
      "epoch": 0.5864245776468466,
      "flos": 20667525050880.0,
      "grad_norm": 2.7927631739534875,
      "language_loss": 0.66187465,
      "learning_rate": 1.5414075554581298e-06,
      "loss": 0.68362617,
      "num_input_tokens_seen": 105210205,
      "step": 4877,
      "time_per_iteration": 3.263946771621704
    },
    {
      "auxiliary_loss_clip": 0.01178424,
      "auxiliary_loss_mlp": 0.01027163,
      "balance_loss_clip": 1.05255139,
      "balance_loss_mlp": 1.01991558,
      "epoch": 0.5865448205374857,
      "flos": 28913907490560.0,
      "grad_norm": 2.465698080662706,
      "language_loss": 0.78675997,
      "learning_rate": 1.5406493729985348e-06,
      "loss": 0.80881584,
      "num_input_tokens_seen": 105229400,
      "step": 4878,
      "time_per_iteration": 2.476013422012329
    },
    {
      "auxiliary_loss_clip": 0.01124357,
      "auxiliary_loss_mlp": 0.00762172,
      "balance_loss_clip": 1.04953027,
      "balance_loss_mlp": 1.00028741,
      "epoch": 0.5866650634281249,
      "flos": 25842575168640.0,
      "grad_norm": 2.697156231673269,
      "language_loss": 0.7248143,
      "learning_rate": 1.5398912602220644e-06,
      "loss": 0.74367952,
      "num_input_tokens_seen": 105248675,
      "step": 4879,
      "time_per_iteration": 2.583902359008789
    },
    {
      "auxiliary_loss_clip": 0.01133181,
      "auxiliary_loss_mlp": 0.01027374,
      "balance_loss_clip": 1.0483495,
      "balance_loss_mlp": 1.01992297,
      "epoch": 0.5867853063187639,
      "flos": 17052325925760.0,
      "grad_norm": 2.2094827278864617,
      "language_loss": 0.78712022,
      "learning_rate": 1.539133217243724e-06,
      "loss": 0.80872571,
      "num_input_tokens_seen": 105265695,
      "step": 4880,
      "time_per_iteration": 2.518782615661621
    },
    {
      "auxiliary_loss_clip": 0.01139508,
      "auxiliary_loss_mlp": 0.01027867,
      "balance_loss_clip": 1.04864419,
      "balance_loss_mlp": 1.01960599,
      "epoch": 0.586905549209403,
      "flos": 24645026707200.0,
      "grad_norm": 2.0988796510168375,
      "language_loss": 0.76015568,
      "learning_rate": 1.5383752441785081e-06,
      "loss": 0.78182948,
      "num_input_tokens_seen": 105284920,
      "step": 4881,
      "time_per_iteration": 2.5902581214904785
    },
    {
      "auxiliary_loss_clip": 0.01167349,
      "auxiliary_loss_mlp": 0.01035072,
      "balance_loss_clip": 1.05216777,
      "balance_loss_mlp": 1.02748466,
      "epoch": 0.5870257921000421,
      "flos": 14720538723840.0,
      "grad_norm": 2.12705509228889,
      "language_loss": 0.85114139,
      "learning_rate": 1.5376173411414003e-06,
      "loss": 0.87316561,
      "num_input_tokens_seen": 105302960,
      "step": 4882,
      "time_per_iteration": 2.4419870376586914
    },
    {
      "auxiliary_loss_clip": 0.01148606,
      "auxiliary_loss_mlp": 0.01030796,
      "balance_loss_clip": 1.04713047,
      "balance_loss_mlp": 1.02263618,
      "epoch": 0.5871460349906812,
      "flos": 23914998691200.0,
      "grad_norm": 2.286116162266255,
      "language_loss": 0.78641808,
      "learning_rate": 1.5368595082473753e-06,
      "loss": 0.80821216,
      "num_input_tokens_seen": 105321260,
      "step": 4883,
      "time_per_iteration": 2.5248355865478516
    },
    {
      "auxiliary_loss_clip": 0.01163735,
      "auxiliary_loss_mlp": 0.01020461,
      "balance_loss_clip": 1.0496695,
      "balance_loss_mlp": 1.01328456,
      "epoch": 0.5872662778813202,
      "flos": 22164174063360.0,
      "grad_norm": 1.6252387607282661,
      "language_loss": 0.77787477,
      "learning_rate": 1.5361017456113935e-06,
      "loss": 0.79971671,
      "num_input_tokens_seen": 105341610,
      "step": 4884,
      "time_per_iteration": 2.4605295658111572
    },
    {
      "auxiliary_loss_clip": 0.01165495,
      "auxiliary_loss_mlp": 0.01027055,
      "balance_loss_clip": 1.05140829,
      "balance_loss_mlp": 1.01914513,
      "epoch": 0.5873865207719594,
      "flos": 18441925430400.0,
      "grad_norm": 1.9039322470754632,
      "language_loss": 0.86321974,
      "learning_rate": 1.5353440533484085e-06,
      "loss": 0.88514519,
      "num_input_tokens_seen": 105360465,
      "step": 4885,
      "time_per_iteration": 2.44521427154541
    },
    {
      "auxiliary_loss_clip": 0.01151262,
      "auxiliary_loss_mlp": 0.01029163,
      "balance_loss_clip": 1.05050683,
      "balance_loss_mlp": 1.02125978,
      "epoch": 0.5875067636625985,
      "flos": 54015321427200.0,
      "grad_norm": 1.8363400797070706,
      "language_loss": 0.65927649,
      "learning_rate": 1.534586431573361e-06,
      "loss": 0.6810807,
      "num_input_tokens_seen": 105385405,
      "step": 4886,
      "time_per_iteration": 2.792919397354126
    },
    {
      "auxiliary_loss_clip": 0.01106488,
      "auxiliary_loss_mlp": 0.01025683,
      "balance_loss_clip": 1.04269648,
      "balance_loss_mlp": 1.01675391,
      "epoch": 0.5876270065532375,
      "flos": 27995707100160.0,
      "grad_norm": 1.8618837842434977,
      "language_loss": 0.78928101,
      "learning_rate": 1.5338288804011817e-06,
      "loss": 0.81060272,
      "num_input_tokens_seen": 105404905,
      "step": 4887,
      "time_per_iteration": 2.654822826385498
    },
    {
      "auxiliary_loss_clip": 0.0114405,
      "auxiliary_loss_mlp": 0.01030356,
      "balance_loss_clip": 1.04710913,
      "balance_loss_mlp": 1.02238119,
      "epoch": 0.5877472494438767,
      "flos": 21361462876800.0,
      "grad_norm": 1.862960841958464,
      "language_loss": 0.71203065,
      "learning_rate": 1.533071399946791e-06,
      "loss": 0.73377466,
      "num_input_tokens_seen": 105423650,
      "step": 4888,
      "time_per_iteration": 2.4955098628997803
    },
    {
      "auxiliary_loss_clip": 0.01150916,
      "auxiliary_loss_mlp": 0.01026224,
      "balance_loss_clip": 1.04857135,
      "balance_loss_mlp": 1.01899385,
      "epoch": 0.5878674923345157,
      "flos": 22383013674240.0,
      "grad_norm": 2.0136479059696826,
      "language_loss": 0.57435399,
      "learning_rate": 1.5323139903250977e-06,
      "loss": 0.59612542,
      "num_input_tokens_seen": 105444255,
      "step": 4889,
      "time_per_iteration": 2.533618688583374
    },
    {
      "auxiliary_loss_clip": 0.01151524,
      "auxiliary_loss_mlp": 0.01024734,
      "balance_loss_clip": 1.05260658,
      "balance_loss_mlp": 1.01745296,
      "epoch": 0.5879877352251548,
      "flos": 21868664872320.0,
      "grad_norm": 1.7949150722762295,
      "language_loss": 0.77045381,
      "learning_rate": 1.5315566516510002e-06,
      "loss": 0.79221642,
      "num_input_tokens_seen": 105462425,
      "step": 4890,
      "time_per_iteration": 2.495255708694458
    },
    {
      "auxiliary_loss_clip": 0.01177807,
      "auxiliary_loss_mlp": 0.01027324,
      "balance_loss_clip": 1.05443501,
      "balance_loss_mlp": 1.01940525,
      "epoch": 0.5881079781157939,
      "flos": 17493811989120.0,
      "grad_norm": 1.7239018469422087,
      "language_loss": 0.67731953,
      "learning_rate": 1.5307993840393857e-06,
      "loss": 0.69937086,
      "num_input_tokens_seen": 105480505,
      "step": 4891,
      "time_per_iteration": 2.4426653385162354
    },
    {
      "auxiliary_loss_clip": 0.0117616,
      "auxiliary_loss_mlp": 0.0102398,
      "balance_loss_clip": 1.05199265,
      "balance_loss_mlp": 1.01665497,
      "epoch": 0.588228221006433,
      "flos": 22601853285120.0,
      "grad_norm": 1.846636804355916,
      "language_loss": 0.80427814,
      "learning_rate": 1.530042187605132e-06,
      "loss": 0.82627952,
      "num_input_tokens_seen": 105499760,
      "step": 4892,
      "time_per_iteration": 2.4386227130889893
    },
    {
      "auxiliary_loss_clip": 0.01162774,
      "auxiliary_loss_mlp": 0.00761443,
      "balance_loss_clip": 1.05201054,
      "balance_loss_mlp": 1.00024509,
      "epoch": 0.5883484638970721,
      "flos": 26176939896960.0,
      "grad_norm": 1.3633597318893786,
      "language_loss": 0.84165776,
      "learning_rate": 1.5292850624631044e-06,
      "loss": 0.86089993,
      "num_input_tokens_seen": 105521955,
      "step": 4893,
      "time_per_iteration": 2.5085768699645996
    },
    {
      "auxiliary_loss_clip": 0.01160156,
      "auxiliary_loss_mlp": 0.01028469,
      "balance_loss_clip": 1.05295575,
      "balance_loss_mlp": 1.02048826,
      "epoch": 0.5884687067877111,
      "flos": 30443737691520.0,
      "grad_norm": 1.909389229039017,
      "language_loss": 0.80411786,
      "learning_rate": 1.5285280087281593e-06,
      "loss": 0.82600415,
      "num_input_tokens_seen": 105542685,
      "step": 4894,
      "time_per_iteration": 2.5591704845428467
    },
    {
      "auxiliary_loss_clip": 0.01061128,
      "auxiliary_loss_mlp": 0.01001673,
      "balance_loss_clip": 1.02113819,
      "balance_loss_mlp": 1.0007019,
      "epoch": 0.5885889496783503,
      "flos": 70507550580480.0,
      "grad_norm": 0.6507384048749758,
      "language_loss": 0.56575453,
      "learning_rate": 1.5277710265151398e-06,
      "loss": 0.58638263,
      "num_input_tokens_seen": 105612165,
      "step": 4895,
      "time_per_iteration": 3.230508804321289
    },
    {
      "auxiliary_loss_clip": 0.0116331,
      "auxiliary_loss_mlp": 0.01023341,
      "balance_loss_clip": 1.05184925,
      "balance_loss_mlp": 1.01520443,
      "epoch": 0.5887091925689893,
      "flos": 19098767485440.0,
      "grad_norm": 2.3995892345281353,
      "language_loss": 0.77127254,
      "learning_rate": 1.5270141159388803e-06,
      "loss": 0.7931391,
      "num_input_tokens_seen": 105629185,
      "step": 4896,
      "time_per_iteration": 3.265119791030884
    },
    {
      "auxiliary_loss_clip": 0.01175512,
      "auxiliary_loss_mlp": 0.010242,
      "balance_loss_clip": 1.0510962,
      "balance_loss_mlp": 1.01646328,
      "epoch": 0.5888294354596284,
      "flos": 23294282739840.0,
      "grad_norm": 1.6262157937744788,
      "language_loss": 0.80495906,
      "learning_rate": 1.526257277114203e-06,
      "loss": 0.82695621,
      "num_input_tokens_seen": 105650260,
      "step": 4897,
      "time_per_iteration": 2.4233932495117188
    },
    {
      "auxiliary_loss_clip": 0.01143025,
      "auxiliary_loss_mlp": 0.01024012,
      "balance_loss_clip": 1.04937696,
      "balance_loss_mlp": 1.01666224,
      "epoch": 0.5889496783502676,
      "flos": 21981532383360.0,
      "grad_norm": 1.8312968996154904,
      "language_loss": 0.79475182,
      "learning_rate": 1.5255005101559201e-06,
      "loss": 0.81642222,
      "num_input_tokens_seen": 105667870,
      "step": 4898,
      "time_per_iteration": 2.489795684814453
    },
    {
      "auxiliary_loss_clip": 0.01166231,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.05195093,
      "balance_loss_mlp": 1.01932025,
      "epoch": 0.5890699212409066,
      "flos": 21685233093120.0,
      "grad_norm": 2.1078549330762324,
      "language_loss": 0.76870197,
      "learning_rate": 1.524743815178833e-06,
      "loss": 0.7906279,
      "num_input_tokens_seen": 105685830,
      "step": 4899,
      "time_per_iteration": 3.3152313232421875
    },
    {
      "auxiliary_loss_clip": 0.0114926,
      "auxiliary_loss_mlp": 0.01024229,
      "balance_loss_clip": 1.04817283,
      "balance_loss_mlp": 1.0168587,
      "epoch": 0.5891901641315457,
      "flos": 19464553635840.0,
      "grad_norm": 3.166271870474221,
      "language_loss": 0.8101716,
      "learning_rate": 1.5239871922977315e-06,
      "loss": 0.83190656,
      "num_input_tokens_seen": 105705745,
      "step": 4900,
      "time_per_iteration": 2.5020203590393066
    },
    {
      "auxiliary_loss_clip": 0.01145567,
      "auxiliary_loss_mlp": 0.01032255,
      "balance_loss_clip": 1.04646659,
      "balance_loss_mlp": 1.02446473,
      "epoch": 0.5893104070221848,
      "flos": 19609884063360.0,
      "grad_norm": 1.846445680777024,
      "language_loss": 0.8989383,
      "learning_rate": 1.523230641627394e-06,
      "loss": 0.92071652,
      "num_input_tokens_seen": 105724730,
      "step": 4901,
      "time_per_iteration": 3.3320086002349854
    },
    {
      "auxiliary_loss_clip": 0.01122301,
      "auxiliary_loss_mlp": 0.01026981,
      "balance_loss_clip": 1.04358888,
      "balance_loss_mlp": 1.01954246,
      "epoch": 0.5894306499128239,
      "flos": 29060063930880.0,
      "grad_norm": 1.9612376460470438,
      "language_loss": 0.72871953,
      "learning_rate": 1.5224741632825888e-06,
      "loss": 0.75021231,
      "num_input_tokens_seen": 105744920,
      "step": 4902,
      "time_per_iteration": 2.615467071533203
    },
    {
      "auxiliary_loss_clip": 0.01181289,
      "auxiliary_loss_mlp": 0.01024252,
      "balance_loss_clip": 1.05576873,
      "balance_loss_mlp": 1.01619053,
      "epoch": 0.589550892803463,
      "flos": 42298890721920.0,
      "grad_norm": 1.6752828781043458,
      "language_loss": 0.69171691,
      "learning_rate": 1.521717757378074e-06,
      "loss": 0.7137723,
      "num_input_tokens_seen": 105765465,
      "step": 4903,
      "time_per_iteration": 2.6308889389038086
    },
    {
      "auxiliary_loss_clip": 0.01168553,
      "auxiliary_loss_mlp": 0.01030049,
      "balance_loss_clip": 1.05305803,
      "balance_loss_mlp": 1.02199602,
      "epoch": 0.5896711356941021,
      "flos": 14137062197760.0,
      "grad_norm": 1.8413498162275888,
      "language_loss": 0.69518149,
      "learning_rate": 1.5209614240285943e-06,
      "loss": 0.7171675,
      "num_input_tokens_seen": 105783120,
      "step": 4904,
      "time_per_iteration": 3.2270171642303467
    },
    {
      "auxiliary_loss_clip": 0.01174072,
      "auxiliary_loss_mlp": 0.00761931,
      "balance_loss_clip": 1.04998147,
      "balance_loss_mlp": 1.00037193,
      "epoch": 0.5897913785847412,
      "flos": 17201355454080.0,
      "grad_norm": 2.593005582368182,
      "language_loss": 0.85191512,
      "learning_rate": 1.520205163348887e-06,
      "loss": 0.87127519,
      "num_input_tokens_seen": 105801055,
      "step": 4905,
      "time_per_iteration": 2.4040586948394775
    },
    {
      "auxiliary_loss_clip": 0.01051347,
      "auxiliary_loss_mlp": 0.010015,
      "balance_loss_clip": 1.01999092,
      "balance_loss_mlp": 1.0003916,
      "epoch": 0.5899116214753802,
      "flos": 48794164202880.0,
      "grad_norm": 0.7262631845310534,
      "language_loss": 0.57018745,
      "learning_rate": 1.519448975453674e-06,
      "loss": 0.590716,
      "num_input_tokens_seen": 105856155,
      "step": 4906,
      "time_per_iteration": 2.957052707672119
    },
    {
      "auxiliary_loss_clip": 0.01164717,
      "auxiliary_loss_mlp": 0.00762303,
      "balance_loss_clip": 1.05371881,
      "balance_loss_mlp": 1.00042319,
      "epoch": 0.5900318643660194,
      "flos": 21103659987840.0,
      "grad_norm": 2.197850812735842,
      "language_loss": 0.75720716,
      "learning_rate": 1.5186928604576696e-06,
      "loss": 0.77647734,
      "num_input_tokens_seen": 105873350,
      "step": 4907,
      "time_per_iteration": 2.457343101501465
    },
    {
      "auxiliary_loss_clip": 0.01147913,
      "auxiliary_loss_mlp": 0.01028378,
      "balance_loss_clip": 1.04785371,
      "balance_loss_mlp": 1.02092731,
      "epoch": 0.5901521072566585,
      "flos": 21178390233600.0,
      "grad_norm": 2.07140024042001,
      "language_loss": 0.76665926,
      "learning_rate": 1.5179368184755752e-06,
      "loss": 0.78842217,
      "num_input_tokens_seen": 105891435,
      "step": 4908,
      "time_per_iteration": 2.48539400100708
    },
    {
      "auxiliary_loss_clip": 0.01146949,
      "auxiliary_loss_mlp": 0.0102247,
      "balance_loss_clip": 1.05015516,
      "balance_loss_mlp": 1.01515639,
      "epoch": 0.5902723501472975,
      "flos": 20225967160320.0,
      "grad_norm": 1.5581074641132817,
      "language_loss": 0.82744849,
      "learning_rate": 1.5171808496220821e-06,
      "loss": 0.84914261,
      "num_input_tokens_seen": 105910190,
      "step": 4909,
      "time_per_iteration": 2.4843177795410156
    },
    {
      "auxiliary_loss_clip": 0.01153602,
      "auxiliary_loss_mlp": 0.01025007,
      "balance_loss_clip": 1.0498457,
      "balance_loss_mlp": 1.01782131,
      "epoch": 0.5903925930379367,
      "flos": 22964407211520.0,
      "grad_norm": 1.711347472887412,
      "language_loss": 0.81179714,
      "learning_rate": 1.5164249540118708e-06,
      "loss": 0.83358324,
      "num_input_tokens_seen": 105929315,
      "step": 4910,
      "time_per_iteration": 2.5136327743530273
    },
    {
      "auxiliary_loss_clip": 0.01108093,
      "auxiliary_loss_mlp": 0.01028028,
      "balance_loss_clip": 1.04377532,
      "balance_loss_mlp": 1.0202291,
      "epoch": 0.5905128359285757,
      "flos": 23367720096000.0,
      "grad_norm": 1.7194469707999254,
      "language_loss": 0.83124453,
      "learning_rate": 1.5156691317596093e-06,
      "loss": 0.8526057,
      "num_input_tokens_seen": 105950740,
      "step": 4911,
      "time_per_iteration": 2.5977957248687744
    },
    {
      "auxiliary_loss_clip": 0.01166619,
      "auxiliary_loss_mlp": 0.00762103,
      "balance_loss_clip": 1.05225146,
      "balance_loss_mlp": 1.00041032,
      "epoch": 0.5906330788192148,
      "flos": 28032335994240.0,
      "grad_norm": 2.351751233723427,
      "language_loss": 0.66862053,
      "learning_rate": 1.5149133829799556e-06,
      "loss": 0.6879077,
      "num_input_tokens_seen": 105968735,
      "step": 4912,
      "time_per_iteration": 2.5219383239746094
    },
    {
      "auxiliary_loss_clip": 0.01154809,
      "auxiliary_loss_mlp": 0.01034103,
      "balance_loss_clip": 1.04871333,
      "balance_loss_mlp": 1.02611589,
      "epoch": 0.590753321709854,
      "flos": 18477943793280.0,
      "grad_norm": 2.1764494683531366,
      "language_loss": 0.8045311,
      "learning_rate": 1.5141577077875556e-06,
      "loss": 0.82642019,
      "num_input_tokens_seen": 105986060,
      "step": 4913,
      "time_per_iteration": 2.460291624069214
    },
    {
      "auxiliary_loss_clip": 0.01166466,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 1.05300474,
      "balance_loss_mlp": 1.01912963,
      "epoch": 0.590873564600493,
      "flos": 16873706568960.0,
      "grad_norm": 1.7998377767036837,
      "language_loss": 0.72579432,
      "learning_rate": 1.5134021062970451e-06,
      "loss": 0.74772632,
      "num_input_tokens_seen": 106004440,
      "step": 4914,
      "time_per_iteration": 2.4363203048706055
    },
    {
      "auxiliary_loss_clip": 0.01130013,
      "auxiliary_loss_mlp": 0.01028509,
      "balance_loss_clip": 1.05197072,
      "balance_loss_mlp": 1.02078462,
      "epoch": 0.5909938074911321,
      "flos": 13516166678400.0,
      "grad_norm": 2.0512883741565373,
      "language_loss": 0.80964404,
      "learning_rate": 1.5126465786230483e-06,
      "loss": 0.83122927,
      "num_input_tokens_seen": 106021215,
      "step": 4915,
      "time_per_iteration": 2.478245258331299
    },
    {
      "auxiliary_loss_clip": 0.01175081,
      "auxiliary_loss_mlp": 0.01026926,
      "balance_loss_clip": 1.05142224,
      "balance_loss_mlp": 1.01912951,
      "epoch": 0.5911140503817712,
      "flos": 26024067613440.0,
      "grad_norm": 1.7527029326285188,
      "language_loss": 0.82141525,
      "learning_rate": 1.5118911248801787e-06,
      "loss": 0.84343535,
      "num_input_tokens_seen": 106039225,
      "step": 4916,
      "time_per_iteration": 2.444098949432373
    },
    {
      "auxiliary_loss_clip": 0.01159578,
      "auxiliary_loss_mlp": 0.01027247,
      "balance_loss_clip": 1.05018401,
      "balance_loss_mlp": 1.02036572,
      "epoch": 0.5912342932724103,
      "flos": 23258731253760.0,
      "grad_norm": 1.979301126389231,
      "language_loss": 0.79502845,
      "learning_rate": 1.5111357451830364e-06,
      "loss": 0.81689668,
      "num_input_tokens_seen": 106057920,
      "step": 4917,
      "time_per_iteration": 2.460336923599243
    },
    {
      "auxiliary_loss_clip": 0.01162486,
      "auxiliary_loss_mlp": 0.01027043,
      "balance_loss_clip": 1.05022955,
      "balance_loss_mlp": 1.01943111,
      "epoch": 0.5913545361630493,
      "flos": 19573039687680.0,
      "grad_norm": 1.9666009456757982,
      "language_loss": 0.71161103,
      "learning_rate": 1.5103804396462131e-06,
      "loss": 0.73350626,
      "num_input_tokens_seen": 106077855,
      "step": 4918,
      "time_per_iteration": 2.4605956077575684
    },
    {
      "auxiliary_loss_clip": 0.01165644,
      "auxiliary_loss_mlp": 0.01030637,
      "balance_loss_clip": 1.04921901,
      "balance_loss_mlp": 1.02252448,
      "epoch": 0.5914747790536885,
      "flos": 26213532877440.0,
      "grad_norm": 3.088505184687068,
      "language_loss": 0.79835898,
      "learning_rate": 1.5096252083842877e-06,
      "loss": 0.8203218,
      "num_input_tokens_seen": 106097065,
      "step": 4919,
      "time_per_iteration": 2.4927666187286377
    },
    {
      "auxiliary_loss_clip": 0.01158832,
      "auxiliary_loss_mlp": 0.01024323,
      "balance_loss_clip": 1.04779673,
      "balance_loss_mlp": 1.01669347,
      "epoch": 0.5915950219443276,
      "flos": 27417545786880.0,
      "grad_norm": 1.6976556706417651,
      "language_loss": 0.84898341,
      "learning_rate": 1.5088700515118285e-06,
      "loss": 0.87081504,
      "num_input_tokens_seen": 106116385,
      "step": 4920,
      "time_per_iteration": 2.503058910369873
    },
    {
      "auxiliary_loss_clip": 0.01130942,
      "auxiliary_loss_mlp": 0.01028448,
      "balance_loss_clip": 1.04954422,
      "balance_loss_mlp": 1.02028191,
      "epoch": 0.5917152648349666,
      "flos": 21907879545600.0,
      "grad_norm": 1.6698580532702783,
      "language_loss": 0.66167057,
      "learning_rate": 1.508114969143392e-06,
      "loss": 0.68326449,
      "num_input_tokens_seen": 106136370,
      "step": 4921,
      "time_per_iteration": 2.513958692550659
    },
    {
      "auxiliary_loss_clip": 0.01149561,
      "auxiliary_loss_mlp": 0.01029695,
      "balance_loss_clip": 1.04787827,
      "balance_loss_mlp": 1.0225482,
      "epoch": 0.5918355077256057,
      "flos": 28109185142400.0,
      "grad_norm": 1.783833279866674,
      "language_loss": 0.77339721,
      "learning_rate": 1.5073599613935238e-06,
      "loss": 0.7951898,
      "num_input_tokens_seen": 106158490,
      "step": 4922,
      "time_per_iteration": 3.415992259979248
    },
    {
      "auxiliary_loss_clip": 0.01149563,
      "auxiliary_loss_mlp": 0.01027678,
      "balance_loss_clip": 1.04928064,
      "balance_loss_mlp": 1.01992917,
      "epoch": 0.5919557506162448,
      "flos": 28183807647360.0,
      "grad_norm": 2.123727696147373,
      "language_loss": 0.57717514,
      "learning_rate": 1.5066050283767574e-06,
      "loss": 0.59894753,
      "num_input_tokens_seen": 106179170,
      "step": 4923,
      "time_per_iteration": 2.5488946437835693
    },
    {
      "auxiliary_loss_clip": 0.0114406,
      "auxiliary_loss_mlp": 0.01028344,
      "balance_loss_clip": 1.04959166,
      "balance_loss_mlp": 1.02095032,
      "epoch": 0.5920759935068839,
      "flos": 12094355652480.0,
      "grad_norm": 1.9618520512369764,
      "language_loss": 0.82455289,
      "learning_rate": 1.505850170207616e-06,
      "loss": 0.84627688,
      "num_input_tokens_seen": 106196035,
      "step": 4924,
      "time_per_iteration": 2.5083770751953125
    },
    {
      "auxiliary_loss_clip": 0.0114584,
      "auxiliary_loss_mlp": 0.01026718,
      "balance_loss_clip": 1.04682088,
      "balance_loss_mlp": 1.01937485,
      "epoch": 0.592196236397523,
      "flos": 29424772673280.0,
      "grad_norm": 2.189166603395096,
      "language_loss": 0.7800976,
      "learning_rate": 1.505095387000611e-06,
      "loss": 0.80182314,
      "num_input_tokens_seen": 106218335,
      "step": 4925,
      "time_per_iteration": 2.5520801544189453
    },
    {
      "auxiliary_loss_clip": 0.01138334,
      "auxiliary_loss_mlp": 0.01026449,
      "balance_loss_clip": 1.0482161,
      "balance_loss_mlp": 1.01918066,
      "epoch": 0.5923164792881621,
      "flos": 24384709866240.0,
      "grad_norm": 2.065637656285934,
      "language_loss": 0.74177647,
      "learning_rate": 1.504340678870242e-06,
      "loss": 0.76342428,
      "num_input_tokens_seen": 106236550,
      "step": 4926,
      "time_per_iteration": 3.3669991493225098
    },
    {
      "auxiliary_loss_clip": 0.01162038,
      "auxiliary_loss_mlp": 0.0103202,
      "balance_loss_clip": 1.05101717,
      "balance_loss_mlp": 1.02484989,
      "epoch": 0.5924367221788012,
      "flos": 24024238928640.0,
      "grad_norm": 1.9737458741265004,
      "language_loss": 0.89623493,
      "learning_rate": 1.5035860459309989e-06,
      "loss": 0.91817546,
      "num_input_tokens_seen": 106254265,
      "step": 4927,
      "time_per_iteration": 2.4698445796966553
    },
    {
      "auxiliary_loss_clip": 0.01143996,
      "auxiliary_loss_mlp": 0.01029548,
      "balance_loss_clip": 1.04748595,
      "balance_loss_mlp": 1.02116728,
      "epoch": 0.5925569650694402,
      "flos": 26870590414080.0,
      "grad_norm": 1.8080600223539682,
      "language_loss": 0.63587487,
      "learning_rate": 1.5028314882973568e-06,
      "loss": 0.6576103,
      "num_input_tokens_seen": 106274670,
      "step": 4928,
      "time_per_iteration": 3.3737740516662598
    },
    {
      "auxiliary_loss_clip": 0.01151784,
      "auxiliary_loss_mlp": 0.01030708,
      "balance_loss_clip": 1.05230355,
      "balance_loss_mlp": 1.02272153,
      "epoch": 0.5926772079600794,
      "flos": 22302788647680.0,
      "grad_norm": 2.645334717007061,
      "language_loss": 0.84577751,
      "learning_rate": 1.502077006083783e-06,
      "loss": 0.86760241,
      "num_input_tokens_seen": 106293330,
      "step": 4929,
      "time_per_iteration": 2.4817793369293213
    },
    {
      "auxiliary_loss_clip": 0.01167471,
      "auxiliary_loss_mlp": 0.00761328,
      "balance_loss_clip": 1.05173385,
      "balance_loss_mlp": 1.00029147,
      "epoch": 0.5927974508507184,
      "flos": 19865244827520.0,
      "grad_norm": 1.7331667546283522,
      "language_loss": 0.76869094,
      "learning_rate": 1.5013225994047315e-06,
      "loss": 0.78797901,
      "num_input_tokens_seen": 106310960,
      "step": 4930,
      "time_per_iteration": 2.4470012187957764
    },
    {
      "auxiliary_loss_clip": 0.01165454,
      "auxiliary_loss_mlp": 0.00761324,
      "balance_loss_clip": 1.05366147,
      "balance_loss_mlp": 1.00030732,
      "epoch": 0.5929176937413575,
      "flos": 15776743167360.0,
      "grad_norm": 1.5171864093309735,
      "language_loss": 0.80508769,
      "learning_rate": 1.5005682683746452e-06,
      "loss": 0.82435542,
      "num_input_tokens_seen": 106329475,
      "step": 4931,
      "time_per_iteration": 3.183997869491577
    },
    {
      "auxiliary_loss_clip": 0.01163788,
      "auxiliary_loss_mlp": 0.01027969,
      "balance_loss_clip": 1.05344498,
      "balance_loss_mlp": 1.02032518,
      "epoch": 0.5930379366319967,
      "flos": 17601472028160.0,
      "grad_norm": 2.2819889038742365,
      "language_loss": 0.72695696,
      "learning_rate": 1.4998140131079553e-06,
      "loss": 0.74887455,
      "num_input_tokens_seen": 106345565,
      "step": 4932,
      "time_per_iteration": 2.418487310409546
    },
    {
      "auxiliary_loss_clip": 0.01102791,
      "auxiliary_loss_mlp": 0.00761458,
      "balance_loss_clip": 1.04363,
      "balance_loss_mlp": 1.00026798,
      "epoch": 0.5931581795226357,
      "flos": 17704283731200.0,
      "grad_norm": 1.7946972982747897,
      "language_loss": 0.73921323,
      "learning_rate": 1.4990598337190821e-06,
      "loss": 0.75785571,
      "num_input_tokens_seen": 106361920,
      "step": 4933,
      "time_per_iteration": 2.531195640563965
    },
    {
      "auxiliary_loss_clip": 0.01175481,
      "auxiliary_loss_mlp": 0.00762408,
      "balance_loss_clip": 1.05142677,
      "balance_loss_mlp": 1.00027764,
      "epoch": 0.5932784224132748,
      "flos": 24280102483200.0,
      "grad_norm": 1.8443745540758933,
      "language_loss": 0.67920095,
      "learning_rate": 1.4983057303224338e-06,
      "loss": 0.69857979,
      "num_input_tokens_seen": 106381735,
      "step": 4934,
      "time_per_iteration": 2.4451987743377686
    },
    {
      "auxiliary_loss_clip": 0.01118944,
      "auxiliary_loss_mlp": 0.01029881,
      "balance_loss_clip": 1.04617429,
      "balance_loss_mlp": 1.0221796,
      "epoch": 0.5933986653039139,
      "flos": 22926700909440.0,
      "grad_norm": 1.6630106734573074,
      "language_loss": 0.87887239,
      "learning_rate": 1.4975517030324072e-06,
      "loss": 0.90036064,
      "num_input_tokens_seen": 106399745,
      "step": 4935,
      "time_per_iteration": 2.543041944503784
    },
    {
      "auxiliary_loss_clip": 0.01073839,
      "auxiliary_loss_mlp": 0.00752771,
      "balance_loss_clip": 1.01507378,
      "balance_loss_mlp": 0.9997347,
      "epoch": 0.593518908194553,
      "flos": 71121730256640.0,
      "grad_norm": 0.7858956376765502,
      "language_loss": 0.61843598,
      "learning_rate": 1.4967977519633882e-06,
      "loss": 0.63670206,
      "num_input_tokens_seen": 106457205,
      "step": 4936,
      "time_per_iteration": 3.118154525756836
    },
    {
      "auxiliary_loss_clip": 0.0113402,
      "auxiliary_loss_mlp": 0.01024581,
      "balance_loss_clip": 1.04871881,
      "balance_loss_mlp": 1.0169394,
      "epoch": 0.593639151085192,
      "flos": 20448649526400.0,
      "grad_norm": 1.9104499777932717,
      "language_loss": 0.78172088,
      "learning_rate": 1.4960438772297494e-06,
      "loss": 0.80330694,
      "num_input_tokens_seen": 106474250,
      "step": 4937,
      "time_per_iteration": 2.512204647064209
    },
    {
      "auxiliary_loss_clip": 0.01149633,
      "auxiliary_loss_mlp": 0.01027627,
      "balance_loss_clip": 1.04809988,
      "balance_loss_mlp": 1.02007508,
      "epoch": 0.5937593939758312,
      "flos": 30883428074880.0,
      "grad_norm": 2.2284603403364525,
      "language_loss": 0.73346782,
      "learning_rate": 1.495290078945855e-06,
      "loss": 0.75524038,
      "num_input_tokens_seen": 106494015,
      "step": 4938,
      "time_per_iteration": 2.5780420303344727
    },
    {
      "auxiliary_loss_clip": 0.01175281,
      "auxiliary_loss_mlp": 0.01029293,
      "balance_loss_clip": 1.05246115,
      "balance_loss_mlp": 1.02158952,
      "epoch": 0.5938796368664703,
      "flos": 36898069668480.0,
      "grad_norm": 2.3131837347137636,
      "language_loss": 0.73908448,
      "learning_rate": 1.4945363572260529e-06,
      "loss": 0.76113027,
      "num_input_tokens_seen": 106515010,
      "step": 4939,
      "time_per_iteration": 2.553023099899292
    },
    {
      "auxiliary_loss_clip": 0.01162091,
      "auxiliary_loss_mlp": 0.0102468,
      "balance_loss_clip": 1.04995251,
      "balance_loss_mlp": 1.01750398,
      "epoch": 0.5939998797571093,
      "flos": 23842926051840.0,
      "grad_norm": 2.280127823288347,
      "language_loss": 0.67816079,
      "learning_rate": 1.4937827121846845e-06,
      "loss": 0.70002848,
      "num_input_tokens_seen": 106535265,
      "step": 4940,
      "time_per_iteration": 2.52632474899292
    },
    {
      "auxiliary_loss_clip": 0.01133648,
      "auxiliary_loss_mlp": 0.01034691,
      "balance_loss_clip": 1.0529623,
      "balance_loss_mlp": 1.02721667,
      "epoch": 0.5941201226477485,
      "flos": 25191407462400.0,
      "grad_norm": 1.513079214174461,
      "language_loss": 0.73617417,
      "learning_rate": 1.4930291439360755e-06,
      "loss": 0.75785756,
      "num_input_tokens_seen": 106557830,
      "step": 4941,
      "time_per_iteration": 2.567063570022583
    },
    {
      "auxiliary_loss_clip": 0.0116539,
      "auxiliary_loss_mlp": 0.01029056,
      "balance_loss_clip": 1.05236495,
      "balance_loss_mlp": 1.02051449,
      "epoch": 0.5942403655383875,
      "flos": 22418996123520.0,
      "grad_norm": 1.7173236475614415,
      "language_loss": 0.79456139,
      "learning_rate": 1.4922756525945427e-06,
      "loss": 0.81650585,
      "num_input_tokens_seen": 106577140,
      "step": 4942,
      "time_per_iteration": 2.447849988937378
    },
    {
      "auxiliary_loss_clip": 0.01063626,
      "auxiliary_loss_mlp": 0.01001372,
      "balance_loss_clip": 1.01418185,
      "balance_loss_mlp": 1.0002749,
      "epoch": 0.5943606084290266,
      "flos": 67629310796160.0,
      "grad_norm": 1.176669132312266,
      "language_loss": 0.59675837,
      "learning_rate": 1.4915222382743894e-06,
      "loss": 0.61740828,
      "num_input_tokens_seen": 106635975,
      "step": 4943,
      "time_per_iteration": 3.1054978370666504
    },
    {
      "auxiliary_loss_clip": 0.01165791,
      "auxiliary_loss_mlp": 0.01027476,
      "balance_loss_clip": 1.05288613,
      "balance_loss_mlp": 1.01963186,
      "epoch": 0.5944808513196658,
      "flos": 18223157646720.0,
      "grad_norm": 2.511971399840203,
      "language_loss": 0.71873939,
      "learning_rate": 1.4907689010899085e-06,
      "loss": 0.74067211,
      "num_input_tokens_seen": 106653555,
      "step": 4944,
      "time_per_iteration": 2.431511640548706
    },
    {
      "auxiliary_loss_clip": 0.01148534,
      "auxiliary_loss_mlp": 0.01024146,
      "balance_loss_clip": 1.04858875,
      "balance_loss_mlp": 1.01636744,
      "epoch": 0.5946010942103048,
      "flos": 24790824011520.0,
      "grad_norm": 2.005001366906365,
      "language_loss": 0.6286431,
      "learning_rate": 1.4900156411553804e-06,
      "loss": 0.65036988,
      "num_input_tokens_seen": 106673385,
      "step": 4945,
      "time_per_iteration": 2.5283596515655518
    },
    {
      "auxiliary_loss_clip": 0.01153577,
      "auxiliary_loss_mlp": 0.01026671,
      "balance_loss_clip": 1.05174875,
      "balance_loss_mlp": 1.01916051,
      "epoch": 0.5947213371009439,
      "flos": 15231619388160.0,
      "grad_norm": 1.8912554429473638,
      "language_loss": 0.85471523,
      "learning_rate": 1.4892624585850739e-06,
      "loss": 0.87651777,
      "num_input_tokens_seen": 106691740,
      "step": 4946,
      "time_per_iteration": 2.4547576904296875
    },
    {
      "auxiliary_loss_clip": 0.01180647,
      "auxiliary_loss_mlp": 0.01029913,
      "balance_loss_clip": 1.05407166,
      "balance_loss_mlp": 1.02195621,
      "epoch": 0.594841579991583,
      "flos": 25848069949440.0,
      "grad_norm": 1.948354772301302,
      "language_loss": 0.79370081,
      "learning_rate": 1.4885093534932465e-06,
      "loss": 0.81580639,
      "num_input_tokens_seen": 106709705,
      "step": 4947,
      "time_per_iteration": 2.451784610748291
    },
    {
      "auxiliary_loss_clip": 0.01149525,
      "auxiliary_loss_mlp": 0.01027932,
      "balance_loss_clip": 1.05171728,
      "balance_loss_mlp": 1.01985598,
      "epoch": 0.5949618228822221,
      "flos": 23981109672960.0,
      "grad_norm": 2.1113095811427676,
      "language_loss": 0.71349955,
      "learning_rate": 1.4877563259941433e-06,
      "loss": 0.7352742,
      "num_input_tokens_seen": 106727560,
      "step": 4948,
      "time_per_iteration": 2.5894393920898438
    },
    {
      "auxiliary_loss_clip": 0.01169289,
      "auxiliary_loss_mlp": 0.0102833,
      "balance_loss_clip": 1.05242014,
      "balance_loss_mlp": 1.02004457,
      "epoch": 0.5950820657728612,
      "flos": 40547491476480.0,
      "grad_norm": 1.9119158894215622,
      "language_loss": 0.67768896,
      "learning_rate": 1.4870033762019988e-06,
      "loss": 0.69966519,
      "num_input_tokens_seen": 106747725,
      "step": 4949,
      "time_per_iteration": 3.4546549320220947
    },
    {
      "auxiliary_loss_clip": 0.01148257,
      "auxiliary_loss_mlp": 0.01028029,
      "balance_loss_clip": 1.05020618,
      "balance_loss_mlp": 1.01979768,
      "epoch": 0.5952023086635003,
      "flos": 23184467884800.0,
      "grad_norm": 1.8718595646818428,
      "language_loss": 0.73295224,
      "learning_rate": 1.4862505042310334e-06,
      "loss": 0.75471509,
      "num_input_tokens_seen": 106767010,
      "step": 4950,
      "time_per_iteration": 2.4868078231811523
    },
    {
      "auxiliary_loss_clip": 0.01140951,
      "auxiliary_loss_mlp": 0.01032313,
      "balance_loss_clip": 1.04924345,
      "balance_loss_mlp": 1.02516675,
      "epoch": 0.5953225515541394,
      "flos": 33653289548160.0,
      "grad_norm": 2.7747454963618057,
      "language_loss": 0.69862986,
      "learning_rate": 1.4854977101954587e-06,
      "loss": 0.72036242,
      "num_input_tokens_seen": 106789230,
      "step": 4951,
      "time_per_iteration": 2.602067470550537
    },
    {
      "auxiliary_loss_clip": 0.01163366,
      "auxiliary_loss_mlp": 0.01028298,
      "balance_loss_clip": 1.04802966,
      "balance_loss_mlp": 1.02034664,
      "epoch": 0.5954427944447784,
      "flos": 24459619680000.0,
      "grad_norm": 1.8566278373746716,
      "language_loss": 0.86410409,
      "learning_rate": 1.4847449942094716e-06,
      "loss": 0.88602066,
      "num_input_tokens_seen": 106808110,
      "step": 4952,
      "time_per_iteration": 2.5694382190704346
    },
    {
      "auxiliary_loss_clip": 0.01145011,
      "auxiliary_loss_mlp": 0.01026784,
      "balance_loss_clip": 1.0497334,
      "balance_loss_mlp": 1.01909447,
      "epoch": 0.5955630373354175,
      "flos": 18551848026240.0,
      "grad_norm": 2.1879984913400503,
      "language_loss": 0.8626225,
      "learning_rate": 1.4839923563872598e-06,
      "loss": 0.88434047,
      "num_input_tokens_seen": 106826650,
      "step": 4953,
      "time_per_iteration": 3.299710750579834
    },
    {
      "auxiliary_loss_clip": 0.01138145,
      "auxiliary_loss_mlp": 0.01028121,
      "balance_loss_clip": 1.05129373,
      "balance_loss_mlp": 1.01990759,
      "epoch": 0.5956832802260567,
      "flos": 19791699730560.0,
      "grad_norm": 2.0111740021682514,
      "language_loss": 0.76011419,
      "learning_rate": 1.483239796842997e-06,
      "loss": 0.78177691,
      "num_input_tokens_seen": 106844680,
      "step": 4954,
      "time_per_iteration": 2.4914398193359375
    },
    {
      "auxiliary_loss_clip": 0.01135083,
      "auxiliary_loss_mlp": 0.01028849,
      "balance_loss_clip": 1.04863405,
      "balance_loss_mlp": 1.02138078,
      "epoch": 0.5958035231166957,
      "flos": 19750868945280.0,
      "grad_norm": 2.391598636700139,
      "language_loss": 0.83933699,
      "learning_rate": 1.4824873156908462e-06,
      "loss": 0.86097634,
      "num_input_tokens_seen": 106862605,
      "step": 4955,
      "time_per_iteration": 3.3377203941345215
    },
    {
      "auxiliary_loss_clip": 0.01163082,
      "auxiliary_loss_mlp": 0.00763065,
      "balance_loss_clip": 1.0517801,
      "balance_loss_mlp": 1.0003221,
      "epoch": 0.5959237660073348,
      "flos": 21652806090240.0,
      "grad_norm": 1.6410390504623167,
      "language_loss": 0.7558533,
      "learning_rate": 1.4817349130449584e-06,
      "loss": 0.77511477,
      "num_input_tokens_seen": 106882325,
      "step": 4956,
      "time_per_iteration": 2.455779552459717
    },
    {
      "auxiliary_loss_clip": 0.01160479,
      "auxiliary_loss_mlp": 0.01024626,
      "balance_loss_clip": 1.05075622,
      "balance_loss_mlp": 1.01708627,
      "epoch": 0.5960440088979739,
      "flos": 21171207513600.0,
      "grad_norm": 1.7988731833587812,
      "language_loss": 0.83128268,
      "learning_rate": 1.4809825890194717e-06,
      "loss": 0.8531338,
      "num_input_tokens_seen": 106900995,
      "step": 4957,
      "time_per_iteration": 3.168745756149292
    },
    {
      "auxiliary_loss_clip": 0.01143143,
      "auxiliary_loss_mlp": 0.01023334,
      "balance_loss_clip": 1.0467366,
      "balance_loss_mlp": 1.01525807,
      "epoch": 0.596164251788613,
      "flos": 14757526753920.0,
      "grad_norm": 1.8143343458085588,
      "language_loss": 0.77339411,
      "learning_rate": 1.4802303437285139e-06,
      "loss": 0.79505885,
      "num_input_tokens_seen": 106918265,
      "step": 4958,
      "time_per_iteration": 2.451044797897339
    },
    {
      "auxiliary_loss_clip": 0.01145635,
      "auxiliary_loss_mlp": 0.01030751,
      "balance_loss_clip": 1.0466224,
      "balance_loss_mlp": 1.02278733,
      "epoch": 0.596284494679252,
      "flos": 20485924865280.0,
      "grad_norm": 2.2989540994973168,
      "language_loss": 0.8101123,
      "learning_rate": 1.4794781772861994e-06,
      "loss": 0.83187616,
      "num_input_tokens_seen": 106934760,
      "step": 4959,
      "time_per_iteration": 2.4854226112365723
    },
    {
      "auxiliary_loss_clip": 0.01146987,
      "auxiliary_loss_mlp": 0.00762551,
      "balance_loss_clip": 1.04849386,
      "balance_loss_mlp": 1.00034475,
      "epoch": 0.5964047375698912,
      "flos": 31212262108800.0,
      "grad_norm": 2.0492302027120926,
      "language_loss": 0.66953945,
      "learning_rate": 1.4787260898066324e-06,
      "loss": 0.68863487,
      "num_input_tokens_seen": 106954760,
      "step": 4960,
      "time_per_iteration": 2.557249069213867
    },
    {
      "auxiliary_loss_clip": 0.01176359,
      "auxiliary_loss_mlp": 0.0102893,
      "balance_loss_clip": 1.05356097,
      "balance_loss_mlp": 1.02102649,
      "epoch": 0.5965249804605303,
      "flos": 27483620855040.0,
      "grad_norm": 1.9676453578294528,
      "language_loss": 0.85881627,
      "learning_rate": 1.4779740814039023e-06,
      "loss": 0.88086915,
      "num_input_tokens_seen": 106974845,
      "step": 4961,
      "time_per_iteration": 2.4634146690368652
    },
    {
      "auxiliary_loss_clip": 0.01175868,
      "auxiliary_loss_mlp": 0.0102955,
      "balance_loss_clip": 1.05139518,
      "balance_loss_mlp": 1.02110434,
      "epoch": 0.5966452233511693,
      "flos": 30773936442240.0,
      "grad_norm": 2.139889861102823,
      "language_loss": 0.68651414,
      "learning_rate": 1.4772221521920894e-06,
      "loss": 0.70856833,
      "num_input_tokens_seen": 106994870,
      "step": 4962,
      "time_per_iteration": 2.473191976547241
    },
    {
      "auxiliary_loss_clip": 0.01152617,
      "auxiliary_loss_mlp": 0.01025369,
      "balance_loss_clip": 1.05465364,
      "balance_loss_mlp": 1.01762605,
      "epoch": 0.5967654662418085,
      "flos": 25481170477440.0,
      "grad_norm": 1.9149100296852453,
      "language_loss": 0.74028611,
      "learning_rate": 1.4764703022852598e-06,
      "loss": 0.76206601,
      "num_input_tokens_seen": 107015390,
      "step": 4963,
      "time_per_iteration": 2.5281455516815186
    },
    {
      "auxiliary_loss_clip": 0.01096019,
      "auxiliary_loss_mlp": 0.01023436,
      "balance_loss_clip": 1.04384995,
      "balance_loss_mlp": 1.01601553,
      "epoch": 0.5968857091324475,
      "flos": 19099126621440.0,
      "grad_norm": 1.839949121094876,
      "language_loss": 0.77111888,
      "learning_rate": 1.4757185317974696e-06,
      "loss": 0.79231346,
      "num_input_tokens_seen": 107033775,
      "step": 4964,
      "time_per_iteration": 2.5622169971466064
    },
    {
      "auxiliary_loss_clip": 0.01164558,
      "auxiliary_loss_mlp": 0.01028244,
      "balance_loss_clip": 1.05113244,
      "balance_loss_mlp": 1.02014971,
      "epoch": 0.5970059520230866,
      "flos": 23692711374720.0,
      "grad_norm": 2.331835218485058,
      "language_loss": 0.70733988,
      "learning_rate": 1.474966840842761e-06,
      "loss": 0.72926784,
      "num_input_tokens_seen": 107053355,
      "step": 4965,
      "time_per_iteration": 2.462054491043091
    },
    {
      "auxiliary_loss_clip": 0.01167546,
      "auxiliary_loss_mlp": 0.01027751,
      "balance_loss_clip": 1.0524137,
      "balance_loss_mlp": 1.01996338,
      "epoch": 0.5971261949137258,
      "flos": 23185545292800.0,
      "grad_norm": 1.7218439076553258,
      "language_loss": 0.86805224,
      "learning_rate": 1.4742152295351655e-06,
      "loss": 0.89000523,
      "num_input_tokens_seen": 107072510,
      "step": 4966,
      "time_per_iteration": 2.4483320713043213
    },
    {
      "auxiliary_loss_clip": 0.01163634,
      "auxiliary_loss_mlp": 0.00762949,
      "balance_loss_clip": 1.05017376,
      "balance_loss_mlp": 1.00041175,
      "epoch": 0.5972464378043648,
      "flos": 20557710195840.0,
      "grad_norm": 2.4383486074991283,
      "language_loss": 0.64000165,
      "learning_rate": 1.4734636979887016e-06,
      "loss": 0.65926743,
      "num_input_tokens_seen": 107089970,
      "step": 4967,
      "time_per_iteration": 2.4462952613830566
    },
    {
      "auxiliary_loss_clip": 0.01138416,
      "auxiliary_loss_mlp": 0.01032135,
      "balance_loss_clip": 1.04806232,
      "balance_loss_mlp": 1.02404666,
      "epoch": 0.5973666806950039,
      "flos": 29387030457600.0,
      "grad_norm": 2.0295601572211126,
      "language_loss": 0.90170622,
      "learning_rate": 1.4727122463173755e-06,
      "loss": 0.92341173,
      "num_input_tokens_seen": 107108500,
      "step": 4968,
      "time_per_iteration": 2.5615103244781494
    },
    {
      "auxiliary_loss_clip": 0.0114958,
      "auxiliary_loss_mlp": 0.01025528,
      "balance_loss_clip": 1.05158031,
      "balance_loss_mlp": 1.01778495,
      "epoch": 0.597486923585643,
      "flos": 22273522041600.0,
      "grad_norm": 1.7394379454645812,
      "language_loss": 0.64354867,
      "learning_rate": 1.471960874635183e-06,
      "loss": 0.66529977,
      "num_input_tokens_seen": 107128060,
      "step": 4969,
      "time_per_iteration": 2.4700706005096436
    },
    {
      "auxiliary_loss_clip": 0.01145599,
      "auxiliary_loss_mlp": 0.01030358,
      "balance_loss_clip": 1.04819679,
      "balance_loss_mlp": 1.02221036,
      "epoch": 0.5976071664762821,
      "flos": 13772461196160.0,
      "grad_norm": 2.187790786830356,
      "language_loss": 0.70690691,
      "learning_rate": 1.4712095830561055e-06,
      "loss": 0.72866654,
      "num_input_tokens_seen": 107146550,
      "step": 4970,
      "time_per_iteration": 2.4548113346099854
    },
    {
      "auxiliary_loss_clip": 0.01147186,
      "auxiliary_loss_mlp": 0.01028012,
      "balance_loss_clip": 1.0471102,
      "balance_loss_mlp": 1.02043581,
      "epoch": 0.5977274093669211,
      "flos": 19098623831040.0,
      "grad_norm": 1.8106387547572447,
      "language_loss": 0.80656016,
      "learning_rate": 1.4704583716941147e-06,
      "loss": 0.8283121,
      "num_input_tokens_seen": 107165415,
      "step": 4971,
      "time_per_iteration": 2.4699997901916504
    },
    {
      "auxiliary_loss_clip": 0.01156526,
      "auxiliary_loss_mlp": 0.01033366,
      "balance_loss_clip": 1.05182981,
      "balance_loss_mlp": 1.02576685,
      "epoch": 0.5978476522575603,
      "flos": 20376002269440.0,
      "grad_norm": 2.1702936540877924,
      "language_loss": 0.72501111,
      "learning_rate": 1.4697072406631672e-06,
      "loss": 0.74691004,
      "num_input_tokens_seen": 107185320,
      "step": 4972,
      "time_per_iteration": 2.4371893405914307
    },
    {
      "auxiliary_loss_clip": 0.01125817,
      "auxiliary_loss_mlp": 0.01034691,
      "balance_loss_clip": 1.05018544,
      "balance_loss_mlp": 1.02584517,
      "epoch": 0.5979678951481994,
      "flos": 29023147728000.0,
      "grad_norm": 1.588791497033447,
      "language_loss": 0.72839105,
      "learning_rate": 1.4689561900772097e-06,
      "loss": 0.74999613,
      "num_input_tokens_seen": 107205380,
      "step": 4973,
      "time_per_iteration": 2.6014435291290283
    },
    {
      "auxiliary_loss_clip": 0.01146675,
      "auxiliary_loss_mlp": 0.01029572,
      "balance_loss_clip": 1.04700828,
      "balance_loss_mlp": 1.02213907,
      "epoch": 0.5980881380388384,
      "flos": 17967689141760.0,
      "grad_norm": 2.0716523326837435,
      "language_loss": 0.72641158,
      "learning_rate": 1.4682052200501758e-06,
      "loss": 0.74817407,
      "num_input_tokens_seen": 107222585,
      "step": 4974,
      "time_per_iteration": 2.4566664695739746
    },
    {
      "auxiliary_loss_clip": 0.01176278,
      "auxiliary_loss_mlp": 0.01027127,
      "balance_loss_clip": 1.05194044,
      "balance_loss_mlp": 1.01902044,
      "epoch": 0.5982083809294776,
      "flos": 22962827013120.0,
      "grad_norm": 1.8513017249803914,
      "language_loss": 0.79629666,
      "learning_rate": 1.4674543306959876e-06,
      "loss": 0.81833076,
      "num_input_tokens_seen": 107242055,
      "step": 4975,
      "time_per_iteration": 2.4262027740478516
    },
    {
      "auxiliary_loss_clip": 0.01156405,
      "auxiliary_loss_mlp": 0.01028812,
      "balance_loss_clip": 1.05264425,
      "balance_loss_mlp": 1.02031815,
      "epoch": 0.5983286238201166,
      "flos": 20991941712000.0,
      "grad_norm": 2.3764702166331664,
      "language_loss": 0.84969616,
      "learning_rate": 1.4667035221285535e-06,
      "loss": 0.87154835,
      "num_input_tokens_seen": 107259695,
      "step": 4976,
      "time_per_iteration": 3.314951181411743
    },
    {
      "auxiliary_loss_clip": 0.01161551,
      "auxiliary_loss_mlp": 0.01027003,
      "balance_loss_clip": 1.05235291,
      "balance_loss_mlp": 1.0192008,
      "epoch": 0.5984488667107557,
      "flos": 28183448511360.0,
      "grad_norm": 1.9596747127444862,
      "language_loss": 0.74235994,
      "learning_rate": 1.4659527944617715e-06,
      "loss": 0.76424545,
      "num_input_tokens_seen": 107279640,
      "step": 4977,
      "time_per_iteration": 2.49161958694458
    },
    {
      "auxiliary_loss_clip": 0.01101519,
      "auxiliary_loss_mlp": 0.01026815,
      "balance_loss_clip": 1.04230893,
      "balance_loss_mlp": 1.018929,
      "epoch": 0.5985691096013949,
      "flos": 16471794314880.0,
      "grad_norm": 1.8085265572665044,
      "language_loss": 0.75897062,
      "learning_rate": 1.465202147809526e-06,
      "loss": 0.78025389,
      "num_input_tokens_seen": 107298135,
      "step": 4978,
      "time_per_iteration": 2.5493552684783936
    },
    {
      "auxiliary_loss_clip": 0.01177732,
      "auxiliary_loss_mlp": 0.01026295,
      "balance_loss_clip": 1.05335546,
      "balance_loss_mlp": 1.01910388,
      "epoch": 0.5986893524920339,
      "flos": 26719046933760.0,
      "grad_norm": 2.0609474560671837,
      "language_loss": 0.76020217,
      "learning_rate": 1.4644515822856888e-06,
      "loss": 0.78224242,
      "num_input_tokens_seen": 107316570,
      "step": 4979,
      "time_per_iteration": 3.2906670570373535
    },
    {
      "auxiliary_loss_clip": 0.01042761,
      "auxiliary_loss_mlp": 0.01002341,
      "balance_loss_clip": 1.01489282,
      "balance_loss_mlp": 1.00126767,
      "epoch": 0.598809595382673,
      "flos": 61608061100160.0,
      "grad_norm": 0.9209046664704985,
      "language_loss": 0.5651983,
      "learning_rate": 1.4637010980041215e-06,
      "loss": 0.58564937,
      "num_input_tokens_seen": 107378680,
      "step": 4980,
      "time_per_iteration": 3.1087265014648438
    },
    {
      "auxiliary_loss_clip": 0.01178441,
      "auxiliary_loss_mlp": 0.01029064,
      "balance_loss_clip": 1.05274665,
      "balance_loss_mlp": 1.02085042,
      "epoch": 0.5989298382733121,
      "flos": 11801719549440.0,
      "grad_norm": 2.3279182921558594,
      "language_loss": 0.89423263,
      "learning_rate": 1.4629506950786707e-06,
      "loss": 0.91630769,
      "num_input_tokens_seen": 107394860,
      "step": 4981,
      "time_per_iteration": 3.2508251667022705
    },
    {
      "auxiliary_loss_clip": 0.01073633,
      "auxiliary_loss_mlp": 0.01000602,
      "balance_loss_clip": 1.01515174,
      "balance_loss_mlp": 0.99955332,
      "epoch": 0.5990500811639512,
      "flos": 60025800021120.0,
      "grad_norm": 0.8080336682060275,
      "language_loss": 0.56101525,
      "learning_rate": 1.4622003736231733e-06,
      "loss": 0.58175755,
      "num_input_tokens_seen": 107453850,
      "step": 4982,
      "time_per_iteration": 3.0760984420776367
    },
    {
      "auxiliary_loss_clip": 0.01163708,
      "auxiliary_loss_mlp": 0.01027783,
      "balance_loss_clip": 1.05302715,
      "balance_loss_mlp": 1.01946187,
      "epoch": 0.5991703240545903,
      "flos": 18222726683520.0,
      "grad_norm": 2.3769213074901163,
      "language_loss": 0.804515,
      "learning_rate": 1.461450133751451e-06,
      "loss": 0.8264299,
      "num_input_tokens_seen": 107471920,
      "step": 4983,
      "time_per_iteration": 2.456864833831787
    },
    {
      "auxiliary_loss_clip": 0.01167607,
      "auxiliary_loss_mlp": 0.01029591,
      "balance_loss_clip": 1.05290306,
      "balance_loss_mlp": 1.02141893,
      "epoch": 0.5992905669452293,
      "flos": 27709894581120.0,
      "grad_norm": 1.8922090337392568,
      "language_loss": 0.76022387,
      "learning_rate": 1.4606999755773153e-06,
      "loss": 0.78219581,
      "num_input_tokens_seen": 107493125,
      "step": 4984,
      "time_per_iteration": 3.260646343231201
    },
    {
      "auxiliary_loss_clip": 0.01176605,
      "auxiliary_loss_mlp": 0.01028571,
      "balance_loss_clip": 1.05327678,
      "balance_loss_mlp": 1.02060187,
      "epoch": 0.5994108098358685,
      "flos": 20449008662400.0,
      "grad_norm": 1.97051652426096,
      "language_loss": 0.821702,
      "learning_rate": 1.4599498992145643e-06,
      "loss": 0.8437537,
      "num_input_tokens_seen": 107513150,
      "step": 4985,
      "time_per_iteration": 2.4697866439819336
    },
    {
      "auxiliary_loss_clip": 0.01155634,
      "auxiliary_loss_mlp": 0.00762361,
      "balance_loss_clip": 1.05148196,
      "balance_loss_mlp": 1.00047624,
      "epoch": 0.5995310527265075,
      "flos": 22269966595200.0,
      "grad_norm": 1.8964918959056367,
      "language_loss": 0.70976949,
      "learning_rate": 1.4591999047769846e-06,
      "loss": 0.72894943,
      "num_input_tokens_seen": 107532005,
      "step": 4986,
      "time_per_iteration": 2.499343156814575
    },
    {
      "auxiliary_loss_clip": 0.01102257,
      "auxiliary_loss_mlp": 0.0103256,
      "balance_loss_clip": 1.04158103,
      "balance_loss_mlp": 1.0239768,
      "epoch": 0.5996512956171466,
      "flos": 18916951818240.0,
      "grad_norm": 1.6569863288346862,
      "language_loss": 0.75477016,
      "learning_rate": 1.4584499923783486e-06,
      "loss": 0.77611834,
      "num_input_tokens_seen": 107550585,
      "step": 4987,
      "time_per_iteration": 2.5597124099731445
    },
    {
      "auxiliary_loss_clip": 0.01146626,
      "auxiliary_loss_mlp": 0.01022737,
      "balance_loss_clip": 1.04883158,
      "balance_loss_mlp": 1.01527464,
      "epoch": 0.5997715385077858,
      "flos": 15370916330880.0,
      "grad_norm": 1.7460331297603642,
      "language_loss": 0.76018357,
      "learning_rate": 1.457700162132419e-06,
      "loss": 0.78187728,
      "num_input_tokens_seen": 107567575,
      "step": 4988,
      "time_per_iteration": 2.455214262008667
    },
    {
      "auxiliary_loss_clip": 0.01116324,
      "auxiliary_loss_mlp": 0.01024681,
      "balance_loss_clip": 1.04639053,
      "balance_loss_mlp": 1.01710546,
      "epoch": 0.5998917813984248,
      "flos": 25264844818560.0,
      "grad_norm": 2.5868413614548573,
      "language_loss": 0.72516936,
      "learning_rate": 1.4569504141529433e-06,
      "loss": 0.74657941,
      "num_input_tokens_seen": 107585410,
      "step": 4989,
      "time_per_iteration": 2.582700252532959
    },
    {
      "auxiliary_loss_clip": 0.01165248,
      "auxiliary_loss_mlp": 0.01032739,
      "balance_loss_clip": 1.05479109,
      "balance_loss_mlp": 1.02441216,
      "epoch": 0.6000120242890639,
      "flos": 22054502862720.0,
      "grad_norm": 1.9744354764354737,
      "language_loss": 0.72019613,
      "learning_rate": 1.456200748553658e-06,
      "loss": 0.74217606,
      "num_input_tokens_seen": 107603405,
      "step": 4990,
      "time_per_iteration": 2.449049472808838
    },
    {
      "auxiliary_loss_clip": 0.01179614,
      "auxiliary_loss_mlp": 0.01031814,
      "balance_loss_clip": 1.0533433,
      "balance_loss_mlp": 1.02327228,
      "epoch": 0.600132267179703,
      "flos": 29863421562240.0,
      "grad_norm": 1.774715116444949,
      "language_loss": 0.78607428,
      "learning_rate": 1.455451165448287e-06,
      "loss": 0.80818862,
      "num_input_tokens_seen": 107626060,
      "step": 4991,
      "time_per_iteration": 2.491271734237671
    },
    {
      "auxiliary_loss_clip": 0.01146083,
      "auxiliary_loss_mlp": 0.01027813,
      "balance_loss_clip": 1.05040836,
      "balance_loss_mlp": 1.01975465,
      "epoch": 0.6002525100703421,
      "flos": 25045358762880.0,
      "grad_norm": 2.1315986861374623,
      "language_loss": 0.73357773,
      "learning_rate": 1.4547016649505407e-06,
      "loss": 0.75531662,
      "num_input_tokens_seen": 107644070,
      "step": 4992,
      "time_per_iteration": 2.5270144939422607
    },
    {
      "auxiliary_loss_clip": 0.01133068,
      "auxiliary_loss_mlp": 0.01023193,
      "balance_loss_clip": 1.04659677,
      "balance_loss_mlp": 1.01486278,
      "epoch": 0.6003727529609811,
      "flos": 20849592113280.0,
      "grad_norm": 2.0925473028567425,
      "language_loss": 0.84769738,
      "learning_rate": 1.4539522471741193e-06,
      "loss": 0.86925995,
      "num_input_tokens_seen": 107661495,
      "step": 4993,
      "time_per_iteration": 2.575850248336792
    },
    {
      "auxiliary_loss_clip": 0.01165259,
      "auxiliary_loss_mlp": 0.01029749,
      "balance_loss_clip": 1.04939103,
      "balance_loss_mlp": 1.02135623,
      "epoch": 0.6004929958516203,
      "flos": 15594604277760.0,
      "grad_norm": 2.0274479076688436,
      "language_loss": 0.71188104,
      "learning_rate": 1.4532029122327067e-06,
      "loss": 0.73383111,
      "num_input_tokens_seen": 107678280,
      "step": 4994,
      "time_per_iteration": 2.423466682434082
    },
    {
      "auxiliary_loss_clip": 0.01132062,
      "auxiliary_loss_mlp": 0.01029756,
      "balance_loss_clip": 1.05140328,
      "balance_loss_mlp": 1.02209139,
      "epoch": 0.6006132387422594,
      "flos": 21763267390080.0,
      "grad_norm": 1.9483110586840606,
      "language_loss": 0.75176615,
      "learning_rate": 1.4524536602399783e-06,
      "loss": 0.77338433,
      "num_input_tokens_seen": 107697370,
      "step": 4995,
      "time_per_iteration": 2.5061392784118652
    },
    {
      "auxiliary_loss_clip": 0.01143593,
      "auxiliary_loss_mlp": 0.01033968,
      "balance_loss_clip": 1.04965007,
      "balance_loss_mlp": 1.02633905,
      "epoch": 0.6007334816328984,
      "flos": 22858542852480.0,
      "grad_norm": 1.575291398430298,
      "language_loss": 0.77695823,
      "learning_rate": 1.4517044913095938e-06,
      "loss": 0.79873389,
      "num_input_tokens_seen": 107717790,
      "step": 4996,
      "time_per_iteration": 2.5041141510009766
    },
    {
      "auxiliary_loss_clip": 0.01163615,
      "auxiliary_loss_mlp": 0.01026292,
      "balance_loss_clip": 1.05148876,
      "balance_loss_mlp": 1.0180068,
      "epoch": 0.6008537245235376,
      "flos": 28324577047680.0,
      "grad_norm": 2.175484317456789,
      "language_loss": 0.81437993,
      "learning_rate": 1.4509554055552022e-06,
      "loss": 0.83627903,
      "num_input_tokens_seen": 107738020,
      "step": 4997,
      "time_per_iteration": 2.5001513957977295
    },
    {
      "auxiliary_loss_clip": 0.0114588,
      "auxiliary_loss_mlp": 0.01029076,
      "balance_loss_clip": 1.04908776,
      "balance_loss_mlp": 1.02102661,
      "epoch": 0.6009739674141766,
      "flos": 20886113266560.0,
      "grad_norm": 3.6005132846392973,
      "language_loss": 0.83950353,
      "learning_rate": 1.450206403090439e-06,
      "loss": 0.86125308,
      "num_input_tokens_seen": 107756215,
      "step": 4998,
      "time_per_iteration": 2.4637374877929688
    },
    {
      "auxiliary_loss_clip": 0.01163042,
      "auxiliary_loss_mlp": 0.01024263,
      "balance_loss_clip": 1.05340576,
      "balance_loss_mlp": 1.01642537,
      "epoch": 0.6010942103048157,
      "flos": 20481004702080.0,
      "grad_norm": 2.753614681859002,
      "language_loss": 0.86094439,
      "learning_rate": 1.4494574840289274e-06,
      "loss": 0.88281751,
      "num_input_tokens_seen": 107773330,
      "step": 4999,
      "time_per_iteration": 2.4368371963500977
    },
    {
      "auxiliary_loss_clip": 0.01166374,
      "auxiliary_loss_mlp": 0.01033472,
      "balance_loss_clip": 1.04971433,
      "balance_loss_mlp": 1.02477598,
      "epoch": 0.6012144531954549,
      "flos": 23805973935360.0,
      "grad_norm": 5.324664305094547,
      "language_loss": 0.73814762,
      "learning_rate": 1.4487086484842782e-06,
      "loss": 0.76014602,
      "num_input_tokens_seen": 107791975,
      "step": 5000,
      "time_per_iteration": 2.4726297855377197
    },
    {
      "auxiliary_loss_clip": 0.01171895,
      "auxiliary_loss_mlp": 0.0102602,
      "balance_loss_clip": 1.04896939,
      "balance_loss_mlp": 1.01844716,
      "epoch": 0.6013346960860939,
      "flos": 18988378012800.0,
      "grad_norm": 2.252309271280477,
      "language_loss": 0.60172224,
      "learning_rate": 1.4479598965700878e-06,
      "loss": 0.62370133,
      "num_input_tokens_seen": 107809240,
      "step": 5001,
      "time_per_iteration": 2.3883323669433594
    },
    {
      "auxiliary_loss_clip": 0.01132745,
      "auxiliary_loss_mlp": 0.01027517,
      "balance_loss_clip": 1.04567719,
      "balance_loss_mlp": 1.01950622,
      "epoch": 0.601454938976733,
      "flos": 24025316336640.0,
      "grad_norm": 2.272489443577085,
      "language_loss": 0.68891734,
      "learning_rate": 1.4472112283999427e-06,
      "loss": 0.71051997,
      "num_input_tokens_seen": 107827895,
      "step": 5002,
      "time_per_iteration": 3.3464436531066895
    },
    {
      "auxiliary_loss_clip": 0.01160421,
      "auxiliary_loss_mlp": 0.01032569,
      "balance_loss_clip": 1.05323637,
      "balance_loss_mlp": 1.0249815,
      "epoch": 0.6015751818673721,
      "flos": 26427129102720.0,
      "grad_norm": 2.0363594406857852,
      "language_loss": 0.6939925,
      "learning_rate": 1.4464626440874143e-06,
      "loss": 0.71592242,
      "num_input_tokens_seen": 107847010,
      "step": 5003,
      "time_per_iteration": 2.494530439376831
    },
    {
      "auxiliary_loss_clip": 0.01126397,
      "auxiliary_loss_mlp": 0.01028353,
      "balance_loss_clip": 1.04351139,
      "balance_loss_mlp": 1.02006209,
      "epoch": 0.6016954247580112,
      "flos": 13115260005120.0,
      "grad_norm": 2.402330950953752,
      "language_loss": 0.74155611,
      "learning_rate": 1.4457141437460636e-06,
      "loss": 0.7631036,
      "num_input_tokens_seen": 107864235,
      "step": 5004,
      "time_per_iteration": 2.50895619392395
    },
    {
      "auxiliary_loss_clip": 0.01149278,
      "auxiliary_loss_mlp": 0.0102956,
      "balance_loss_clip": 1.04833245,
      "balance_loss_mlp": 1.02110839,
      "epoch": 0.6018156676486502,
      "flos": 23768447201280.0,
      "grad_norm": 1.7299193761026348,
      "language_loss": 0.73381329,
      "learning_rate": 1.444965727489436e-06,
      "loss": 0.7556017,
      "num_input_tokens_seen": 107883680,
      "step": 5005,
      "time_per_iteration": 3.347539186477661
    },
    {
      "auxiliary_loss_clip": 0.01130165,
      "auxiliary_loss_mlp": 0.01029062,
      "balance_loss_clip": 1.04347682,
      "balance_loss_mlp": 1.02115822,
      "epoch": 0.6019359105392894,
      "flos": 26469360518400.0,
      "grad_norm": 9.282858128044536,
      "language_loss": 0.63305753,
      "learning_rate": 1.444217395431066e-06,
      "loss": 0.65464979,
      "num_input_tokens_seen": 107906220,
      "step": 5006,
      "time_per_iteration": 2.5798964500427246
    },
    {
      "auxiliary_loss_clip": 0.01038194,
      "auxiliary_loss_mlp": 0.01005458,
      "balance_loss_clip": 1.01464319,
      "balance_loss_mlp": 1.00442088,
      "epoch": 0.6020561534299285,
      "flos": 69190849728000.0,
      "grad_norm": 0.8039692586969128,
      "language_loss": 0.55863523,
      "learning_rate": 1.4434691476844755e-06,
      "loss": 0.57907176,
      "num_input_tokens_seen": 107967195,
      "step": 5007,
      "time_per_iteration": 3.0605690479278564
    },
    {
      "auxiliary_loss_clip": 0.01143008,
      "auxiliary_loss_mlp": 0.01026873,
      "balance_loss_clip": 1.04916382,
      "balance_loss_mlp": 1.0197922,
      "epoch": 0.6021763963205675,
      "flos": 21835304115840.0,
      "grad_norm": 2.123492861379607,
      "language_loss": 0.6677469,
      "learning_rate": 1.4427209843631729e-06,
      "loss": 0.68944573,
      "num_input_tokens_seen": 107984245,
      "step": 5008,
      "time_per_iteration": 3.344230890274048
    },
    {
      "auxiliary_loss_clip": 0.01175014,
      "auxiliary_loss_mlp": 0.00762186,
      "balance_loss_clip": 1.05259132,
      "balance_loss_mlp": 1.0005331,
      "epoch": 0.6022966392112067,
      "flos": 26578636669440.0,
      "grad_norm": 1.83676117665767,
      "language_loss": 0.81273806,
      "learning_rate": 1.4419729055806534e-06,
      "loss": 0.83211005,
      "num_input_tokens_seen": 108003680,
      "step": 5009,
      "time_per_iteration": 2.489734172821045
    },
    {
      "auxiliary_loss_clip": 0.01144772,
      "auxiliary_loss_mlp": 0.00762106,
      "balance_loss_clip": 1.05130374,
      "balance_loss_mlp": 1.0005089,
      "epoch": 0.6024168821018457,
      "flos": 20703722981760.0,
      "grad_norm": 2.1334655888156724,
      "language_loss": 0.82075316,
      "learning_rate": 1.441224911450401e-06,
      "loss": 0.83982193,
      "num_input_tokens_seen": 108019635,
      "step": 5010,
      "time_per_iteration": 3.2493977546691895
    },
    {
      "auxiliary_loss_clip": 0.01166597,
      "auxiliary_loss_mlp": 0.01035405,
      "balance_loss_clip": 1.05022347,
      "balance_loss_mlp": 1.02731037,
      "epoch": 0.6025371249924848,
      "flos": 24680973242880.0,
      "grad_norm": 2.7722133494753143,
      "language_loss": 0.82099724,
      "learning_rate": 1.4404770020858851e-06,
      "loss": 0.84301722,
      "num_input_tokens_seen": 108039120,
      "step": 5011,
      "time_per_iteration": 2.47942852973938
    },
    {
      "auxiliary_loss_clip": 0.0115398,
      "auxiliary_loss_mlp": 0.01027947,
      "balance_loss_clip": 1.0478797,
      "balance_loss_mlp": 1.02033556,
      "epoch": 0.602657367883124,
      "flos": 25955801815680.0,
      "grad_norm": 1.59993587240379,
      "language_loss": 0.85915565,
      "learning_rate": 1.439729177600563e-06,
      "loss": 0.88097489,
      "num_input_tokens_seen": 108059615,
      "step": 5012,
      "time_per_iteration": 2.4997735023498535
    },
    {
      "auxiliary_loss_clip": 0.01137371,
      "auxiliary_loss_mlp": 0.01026793,
      "balance_loss_clip": 1.04735422,
      "balance_loss_mlp": 1.01887453,
      "epoch": 0.602777610773763,
      "flos": 16690633925760.0,
      "grad_norm": 2.0473062333553855,
      "language_loss": 0.73011851,
      "learning_rate": 1.4389814381078793e-06,
      "loss": 0.75176024,
      "num_input_tokens_seen": 108078855,
      "step": 5013,
      "time_per_iteration": 2.539184331893921
    },
    {
      "auxiliary_loss_clip": 0.01087009,
      "auxiliary_loss_mlp": 0.01034419,
      "balance_loss_clip": 1.0457232,
      "balance_loss_mlp": 1.02707553,
      "epoch": 0.6028978536644021,
      "flos": 13334243270400.0,
      "grad_norm": 3.7071607608245536,
      "language_loss": 0.8010776,
      "learning_rate": 1.438233783721265e-06,
      "loss": 0.82229185,
      "num_input_tokens_seen": 108095020,
      "step": 5014,
      "time_per_iteration": 2.580188035964966
    },
    {
      "auxiliary_loss_clip": 0.01148489,
      "auxiliary_loss_mlp": 0.01026996,
      "balance_loss_clip": 1.05526173,
      "balance_loss_mlp": 1.01934278,
      "epoch": 0.6030180965550412,
      "flos": 19644825018240.0,
      "grad_norm": 1.9904394804667211,
      "language_loss": 0.77557641,
      "learning_rate": 1.43748621455414e-06,
      "loss": 0.79733133,
      "num_input_tokens_seen": 108111455,
      "step": 5015,
      "time_per_iteration": 2.4913110733032227
    },
    {
      "auxiliary_loss_clip": 0.01144282,
      "auxiliary_loss_mlp": 0.01029333,
      "balance_loss_clip": 1.04798412,
      "balance_loss_mlp": 1.02138174,
      "epoch": 0.6031383394456803,
      "flos": 14458390289280.0,
      "grad_norm": 2.5235556294668817,
      "language_loss": 0.80881631,
      "learning_rate": 1.4367387307199082e-06,
      "loss": 0.83055246,
      "num_input_tokens_seen": 108128305,
      "step": 5016,
      "time_per_iteration": 2.4549930095672607
    },
    {
      "auxiliary_loss_clip": 0.01156434,
      "auxiliary_loss_mlp": 0.01030407,
      "balance_loss_clip": 1.0478915,
      "balance_loss_mlp": 1.02269471,
      "epoch": 0.6032585823363193,
      "flos": 13917791623680.0,
      "grad_norm": 2.02468675669203,
      "language_loss": 0.82572615,
      "learning_rate": 1.4359913323319632e-06,
      "loss": 0.84759456,
      "num_input_tokens_seen": 108145475,
      "step": 5017,
      "time_per_iteration": 2.4183881282806396
    },
    {
      "auxiliary_loss_clip": 0.01092142,
      "auxiliary_loss_mlp": 0.0102733,
      "balance_loss_clip": 1.04136455,
      "balance_loss_mlp": 1.01943231,
      "epoch": 0.6033788252269584,
      "flos": 24353252530560.0,
      "grad_norm": 1.5656940464598688,
      "language_loss": 0.77422225,
      "learning_rate": 1.4352440195036847e-06,
      "loss": 0.79541701,
      "num_input_tokens_seen": 108165650,
      "step": 5018,
      "time_per_iteration": 2.7103166580200195
    },
    {
      "auxiliary_loss_clip": 0.01096191,
      "auxiliary_loss_mlp": 0.01024006,
      "balance_loss_clip": 1.04015517,
      "balance_loss_mlp": 1.01623929,
      "epoch": 0.6034990681175976,
      "flos": 25521247077120.0,
      "grad_norm": 2.2470337703071093,
      "language_loss": 0.79701352,
      "learning_rate": 1.4344967923484395e-06,
      "loss": 0.81821549,
      "num_input_tokens_seen": 108187620,
      "step": 5019,
      "time_per_iteration": 2.645002603530884
    },
    {
      "auxiliary_loss_clip": 0.01157419,
      "auxiliary_loss_mlp": 0.01028951,
      "balance_loss_clip": 1.0487802,
      "balance_loss_mlp": 1.02129126,
      "epoch": 0.6036193110082366,
      "flos": 25958387594880.0,
      "grad_norm": 2.1305302587415147,
      "language_loss": 0.72435927,
      "learning_rate": 1.433749650979581e-06,
      "loss": 0.74622297,
      "num_input_tokens_seen": 108207605,
      "step": 5020,
      "time_per_iteration": 2.5045955181121826
    },
    {
      "auxiliary_loss_clip": 0.01137612,
      "auxiliary_loss_mlp": 0.01027896,
      "balance_loss_clip": 1.04668975,
      "balance_loss_mlp": 1.02015305,
      "epoch": 0.6037395538988757,
      "flos": 25593427457280.0,
      "grad_norm": 1.7935822766322709,
      "language_loss": 0.67929846,
      "learning_rate": 1.433002595510451e-06,
      "loss": 0.70095348,
      "num_input_tokens_seen": 108226385,
      "step": 5021,
      "time_per_iteration": 2.550509452819824
    },
    {
      "auxiliary_loss_clip": 0.01142626,
      "auxiliary_loss_mlp": 0.00763046,
      "balance_loss_clip": 1.04645419,
      "balance_loss_mlp": 1.00054348,
      "epoch": 0.6038597967895148,
      "flos": 17816253402240.0,
      "grad_norm": 1.8841670330695506,
      "language_loss": 0.72065091,
      "learning_rate": 1.4322556260543757e-06,
      "loss": 0.73970765,
      "num_input_tokens_seen": 108242960,
      "step": 5022,
      "time_per_iteration": 2.4696199893951416
    },
    {
      "auxiliary_loss_clip": 0.01042647,
      "auxiliary_loss_mlp": 0.01004339,
      "balance_loss_clip": 1.01380968,
      "balance_loss_mlp": 1.00323629,
      "epoch": 0.6039800396801539,
      "flos": 65169213235200.0,
      "grad_norm": 0.9004752528451603,
      "language_loss": 0.62714118,
      "learning_rate": 1.4315087427246703e-06,
      "loss": 0.64761102,
      "num_input_tokens_seen": 108296785,
      "step": 5023,
      "time_per_iteration": 2.988929271697998
    },
    {
      "auxiliary_loss_clip": 0.01074203,
      "auxiliary_loss_mlp": 0.01001746,
      "balance_loss_clip": 1.01607776,
      "balance_loss_mlp": 1.00070286,
      "epoch": 0.604100282570793,
      "flos": 67386409073280.0,
      "grad_norm": 0.8659566953079477,
      "language_loss": 0.58467013,
      "learning_rate": 1.4307619456346372e-06,
      "loss": 0.60542959,
      "num_input_tokens_seen": 108341090,
      "step": 5024,
      "time_per_iteration": 2.72853422164917
    },
    {
      "auxiliary_loss_clip": 0.01162675,
      "auxiliary_loss_mlp": 0.01027537,
      "balance_loss_clip": 1.04793739,
      "balance_loss_mlp": 1.01937151,
      "epoch": 0.6042205254614321,
      "flos": 35297495631360.0,
      "grad_norm": 2.4484646796305656,
      "language_loss": 0.74530387,
      "learning_rate": 1.430015234897564e-06,
      "loss": 0.76720595,
      "num_input_tokens_seen": 108364370,
      "step": 5025,
      "time_per_iteration": 2.5868020057678223
    },
    {
      "auxiliary_loss_clip": 0.01175461,
      "auxiliary_loss_mlp": 0.00762733,
      "balance_loss_clip": 1.05122185,
      "balance_loss_mlp": 1.00058103,
      "epoch": 0.6043407683520712,
      "flos": 45658262206080.0,
      "grad_norm": 1.9333595655394096,
      "language_loss": 0.66328311,
      "learning_rate": 1.4292686106267274e-06,
      "loss": 0.68266499,
      "num_input_tokens_seen": 108387220,
      "step": 5026,
      "time_per_iteration": 2.66455340385437
    },
    {
      "auxiliary_loss_clip": 0.01165658,
      "auxiliary_loss_mlp": 0.01031724,
      "balance_loss_clip": 1.050825,
      "balance_loss_mlp": 1.02364135,
      "epoch": 0.6044610112427102,
      "flos": 16180020138240.0,
      "grad_norm": 1.893756904251137,
      "language_loss": 0.77115452,
      "learning_rate": 1.4285220729353876e-06,
      "loss": 0.79312837,
      "num_input_tokens_seen": 108405760,
      "step": 5027,
      "time_per_iteration": 2.4467709064483643
    },
    {
      "auxiliary_loss_clip": 0.01143584,
      "auxiliary_loss_mlp": 0.01026576,
      "balance_loss_clip": 1.04542017,
      "balance_loss_mlp": 1.01868415,
      "epoch": 0.6045812541333494,
      "flos": 13804062186240.0,
      "grad_norm": 3.590940819406141,
      "language_loss": 0.78105903,
      "learning_rate": 1.4277756219367957e-06,
      "loss": 0.8027606,
      "num_input_tokens_seen": 108422785,
      "step": 5028,
      "time_per_iteration": 2.458397626876831
    },
    {
      "auxiliary_loss_clip": 0.01141256,
      "auxiliary_loss_mlp": 0.01025115,
      "balance_loss_clip": 1.04836535,
      "balance_loss_mlp": 1.01712227,
      "epoch": 0.6047014970239885,
      "flos": 19975059682560.0,
      "grad_norm": 2.1083199103529418,
      "language_loss": 0.79953861,
      "learning_rate": 1.4270292577441864e-06,
      "loss": 0.82120228,
      "num_input_tokens_seen": 108442290,
      "step": 5029,
      "time_per_iteration": 3.335294008255005
    },
    {
      "auxiliary_loss_clip": 0.01163883,
      "auxiliary_loss_mlp": 0.01026311,
      "balance_loss_clip": 1.04761863,
      "balance_loss_mlp": 1.01827013,
      "epoch": 0.6048217399146275,
      "flos": 25337097025920.0,
      "grad_norm": 1.554436458152638,
      "language_loss": 0.71815109,
      "learning_rate": 1.4262829804707836e-06,
      "loss": 0.74005306,
      "num_input_tokens_seen": 108464280,
      "step": 5030,
      "time_per_iteration": 2.478785514831543
    },
    {
      "auxiliary_loss_clip": 0.01162765,
      "auxiliary_loss_mlp": 0.010281,
      "balance_loss_clip": 1.04754078,
      "balance_loss_mlp": 1.01990485,
      "epoch": 0.6049419828052667,
      "flos": 26030819370240.0,
      "grad_norm": 1.5380958489320469,
      "language_loss": 0.69692987,
      "learning_rate": 1.4255367902297958e-06,
      "loss": 0.71883845,
      "num_input_tokens_seen": 108485610,
      "step": 5031,
      "time_per_iteration": 2.4898414611816406
    },
    {
      "auxiliary_loss_clip": 0.01172529,
      "auxiliary_loss_mlp": 0.01028998,
      "balance_loss_clip": 1.05104756,
      "balance_loss_mlp": 1.0216012,
      "epoch": 0.6050622256959057,
      "flos": 14648106948480.0,
      "grad_norm": 2.0205071211404957,
      "language_loss": 0.78258532,
      "learning_rate": 1.4247906871344215e-06,
      "loss": 0.8046006,
      "num_input_tokens_seen": 108501005,
      "step": 5032,
      "time_per_iteration": 3.2430365085601807
    },
    {
      "auxiliary_loss_clip": 0.01138409,
      "auxiliary_loss_mlp": 0.01021865,
      "balance_loss_clip": 1.04324579,
      "balance_loss_mlp": 1.01417613,
      "epoch": 0.6051824685865448,
      "flos": 23331450337920.0,
      "grad_norm": 1.9961447986959235,
      "language_loss": 0.75440133,
      "learning_rate": 1.4240446712978415e-06,
      "loss": 0.77600408,
      "num_input_tokens_seen": 108519990,
      "step": 5033,
      "time_per_iteration": 2.5093581676483154
    },
    {
      "auxiliary_loss_clip": 0.01166141,
      "auxiliary_loss_mlp": 0.01023443,
      "balance_loss_clip": 1.05149388,
      "balance_loss_mlp": 1.01497316,
      "epoch": 0.605302711477184,
      "flos": 27563307177600.0,
      "grad_norm": 1.944082708811842,
      "language_loss": 0.74390268,
      "learning_rate": 1.423298742833227e-06,
      "loss": 0.76579857,
      "num_input_tokens_seen": 108538650,
      "step": 5034,
      "time_per_iteration": 3.369309902191162
    },
    {
      "auxiliary_loss_clip": 0.011362,
      "auxiliary_loss_mlp": 0.01028501,
      "balance_loss_clip": 1.04437637,
      "balance_loss_mlp": 1.02083015,
      "epoch": 0.605422954367823,
      "flos": 15154698412800.0,
      "grad_norm": 1.9796431019349092,
      "language_loss": 0.71518171,
      "learning_rate": 1.4225529018537352e-06,
      "loss": 0.73682868,
      "num_input_tokens_seen": 108554155,
      "step": 5035,
      "time_per_iteration": 2.485997438430786
    },
    {
      "auxiliary_loss_clip": 0.01174463,
      "auxiliary_loss_mlp": 0.01027055,
      "balance_loss_clip": 1.05160427,
      "balance_loss_mlp": 1.0192349,
      "epoch": 0.6055431972584621,
      "flos": 27673912131840.0,
      "grad_norm": 1.5391614634909567,
      "language_loss": 0.77641916,
      "learning_rate": 1.4218071484725082e-06,
      "loss": 0.79843432,
      "num_input_tokens_seen": 108576275,
      "step": 5036,
      "time_per_iteration": 3.2165653705596924
    },
    {
      "auxiliary_loss_clip": 0.01143802,
      "auxiliary_loss_mlp": 0.01033921,
      "balance_loss_clip": 1.0498203,
      "balance_loss_mlp": 1.0262202,
      "epoch": 0.6056634401491012,
      "flos": 19387489006080.0,
      "grad_norm": 1.939184240632925,
      "language_loss": 0.76289034,
      "learning_rate": 1.4210614828026786e-06,
      "loss": 0.78466761,
      "num_input_tokens_seen": 108594125,
      "step": 5037,
      "time_per_iteration": 2.4745450019836426
    },
    {
      "auxiliary_loss_clip": 0.01173516,
      "auxiliary_loss_mlp": 0.01021558,
      "balance_loss_clip": 1.05024421,
      "balance_loss_mlp": 1.01408923,
      "epoch": 0.6057836830397403,
      "flos": 24789459294720.0,
      "grad_norm": 1.7997925081321913,
      "language_loss": 0.74395478,
      "learning_rate": 1.4203159049573605e-06,
      "loss": 0.7659055,
      "num_input_tokens_seen": 108615360,
      "step": 5038,
      "time_per_iteration": 2.4586880207061768
    },
    {
      "auxiliary_loss_clip": 0.01153919,
      "auxiliary_loss_mlp": 0.01028599,
      "balance_loss_clip": 1.0476861,
      "balance_loss_mlp": 1.02069545,
      "epoch": 0.6059039259303793,
      "flos": 20558248899840.0,
      "grad_norm": 2.0111601639072414,
      "language_loss": 0.86556637,
      "learning_rate": 1.4195704150496593e-06,
      "loss": 0.88739151,
      "num_input_tokens_seen": 108633075,
      "step": 5039,
      "time_per_iteration": 2.4744415283203125
    },
    {
      "auxiliary_loss_clip": 0.01146098,
      "auxiliary_loss_mlp": 0.01025455,
      "balance_loss_clip": 1.04911399,
      "balance_loss_mlp": 1.0177238,
      "epoch": 0.6060241688210185,
      "flos": 21069724613760.0,
      "grad_norm": 1.6540863857803458,
      "language_loss": 0.73883128,
      "learning_rate": 1.4188250131926639e-06,
      "loss": 0.7605468,
      "num_input_tokens_seen": 108651875,
      "step": 5040,
      "time_per_iteration": 2.480175018310547
    },
    {
      "auxiliary_loss_clip": 0.01148466,
      "auxiliary_loss_mlp": 0.01029292,
      "balance_loss_clip": 1.04756987,
      "balance_loss_mlp": 1.02094769,
      "epoch": 0.6061444117116576,
      "flos": 16361081619840.0,
      "grad_norm": 1.9325676538841574,
      "language_loss": 0.80827814,
      "learning_rate": 1.4180796994994525e-06,
      "loss": 0.83005571,
      "num_input_tokens_seen": 108669290,
      "step": 5041,
      "time_per_iteration": 2.4469096660614014
    },
    {
      "auxiliary_loss_clip": 0.01142959,
      "auxiliary_loss_mlp": 0.01021449,
      "balance_loss_clip": 1.04533744,
      "balance_loss_mlp": 1.01383805,
      "epoch": 0.6062646546022966,
      "flos": 21507296094720.0,
      "grad_norm": 1.7463168202069994,
      "language_loss": 0.72617656,
      "learning_rate": 1.4173344740830877e-06,
      "loss": 0.74782062,
      "num_input_tokens_seen": 108688420,
      "step": 5042,
      "time_per_iteration": 2.480771780014038
    },
    {
      "auxiliary_loss_clip": 0.01145253,
      "auxiliary_loss_mlp": 0.01031683,
      "balance_loss_clip": 1.05302942,
      "balance_loss_mlp": 1.0237143,
      "epoch": 0.6063848974929358,
      "flos": 38983151283840.0,
      "grad_norm": 1.9515909128149715,
      "language_loss": 0.70632088,
      "learning_rate": 1.4165893370566206e-06,
      "loss": 0.72809023,
      "num_input_tokens_seen": 108712175,
      "step": 5043,
      "time_per_iteration": 2.6273722648620605
    },
    {
      "auxiliary_loss_clip": 0.01154709,
      "auxiliary_loss_mlp": 0.01029925,
      "balance_loss_clip": 1.04581594,
      "balance_loss_mlp": 1.02180684,
      "epoch": 0.6065051403835748,
      "flos": 19646584784640.0,
      "grad_norm": 1.6974091260036241,
      "language_loss": 0.77468133,
      "learning_rate": 1.4158442885330865e-06,
      "loss": 0.79652762,
      "num_input_tokens_seen": 108730745,
      "step": 5044,
      "time_per_iteration": 2.4389431476593018
    },
    {
      "auxiliary_loss_clip": 0.01153892,
      "auxiliary_loss_mlp": 0.01031203,
      "balance_loss_clip": 1.04661655,
      "balance_loss_mlp": 1.02289975,
      "epoch": 0.6066253832742139,
      "flos": 23513086437120.0,
      "grad_norm": 1.9480080935240414,
      "language_loss": 0.78942889,
      "learning_rate": 1.4150993286255094e-06,
      "loss": 0.81127989,
      "num_input_tokens_seen": 108749995,
      "step": 5045,
      "time_per_iteration": 2.467611074447632
    },
    {
      "auxiliary_loss_clip": 0.0117245,
      "auxiliary_loss_mlp": 0.01025729,
      "balance_loss_clip": 1.04905248,
      "balance_loss_mlp": 1.01823974,
      "epoch": 0.6067456261648531,
      "flos": 19133708440320.0,
      "grad_norm": 2.471559180399533,
      "language_loss": 0.79588759,
      "learning_rate": 1.4143544574468993e-06,
      "loss": 0.81786942,
      "num_input_tokens_seen": 108768355,
      "step": 5046,
      "time_per_iteration": 2.4121909141540527
    },
    {
      "auxiliary_loss_clip": 0.01158391,
      "auxiliary_loss_mlp": 0.0102485,
      "balance_loss_clip": 1.04983783,
      "balance_loss_mlp": 1.01658869,
      "epoch": 0.6068658690554921,
      "flos": 20520614424960.0,
      "grad_norm": 1.722803665476089,
      "language_loss": 0.82395798,
      "learning_rate": 1.4136096751102523e-06,
      "loss": 0.84579033,
      "num_input_tokens_seen": 108786685,
      "step": 5047,
      "time_per_iteration": 2.447263479232788
    },
    {
      "auxiliary_loss_clip": 0.01149321,
      "auxiliary_loss_mlp": 0.0102625,
      "balance_loss_clip": 1.0499053,
      "balance_loss_mlp": 1.01854324,
      "epoch": 0.6069861119461312,
      "flos": 27374560185600.0,
      "grad_norm": 2.05435511508815,
      "language_loss": 0.8265512,
      "learning_rate": 1.4128649817285516e-06,
      "loss": 0.84830701,
      "num_input_tokens_seen": 108804820,
      "step": 5048,
      "time_per_iteration": 2.5240516662597656
    },
    {
      "auxiliary_loss_clip": 0.01149415,
      "auxiliary_loss_mlp": 0.01039071,
      "balance_loss_clip": 1.04676509,
      "balance_loss_mlp": 1.03107178,
      "epoch": 0.6071063548367702,
      "flos": 25626500904960.0,
      "grad_norm": 2.6059429650678747,
      "language_loss": 0.63015944,
      "learning_rate": 1.412120377414766e-06,
      "loss": 0.6520443,
      "num_input_tokens_seen": 108825010,
      "step": 5049,
      "time_per_iteration": 2.5339953899383545
    },
    {
      "auxiliary_loss_clip": 0.01176174,
      "auxiliary_loss_mlp": 0.01029646,
      "balance_loss_clip": 1.05325937,
      "balance_loss_mlp": 1.02182603,
      "epoch": 0.6072265977274094,
      "flos": 24460517520000.0,
      "grad_norm": 1.5043191827552223,
      "language_loss": 0.71129543,
      "learning_rate": 1.4113758622818522e-06,
      "loss": 0.73335361,
      "num_input_tokens_seen": 108845075,
      "step": 5050,
      "time_per_iteration": 2.453789710998535
    },
    {
      "auxiliary_loss_clip": 0.01151449,
      "auxiliary_loss_mlp": 0.00762229,
      "balance_loss_clip": 1.05008698,
      "balance_loss_mlp": 1.00051451,
      "epoch": 0.6073468406180484,
      "flos": 18149253413760.0,
      "grad_norm": 1.7414512563671498,
      "language_loss": 0.82665896,
      "learning_rate": 1.410631436442751e-06,
      "loss": 0.84579575,
      "num_input_tokens_seen": 108863870,
      "step": 5051,
      "time_per_iteration": 2.470334529876709
    },
    {
      "auxiliary_loss_clip": 0.011641,
      "auxiliary_loss_mlp": 0.01025167,
      "balance_loss_clip": 1.05019867,
      "balance_loss_mlp": 1.01738906,
      "epoch": 0.6074670835086875,
      "flos": 20697617669760.0,
      "grad_norm": 2.377530585480949,
      "language_loss": 0.86479461,
      "learning_rate": 1.4098871000103936e-06,
      "loss": 0.88668728,
      "num_input_tokens_seen": 108882470,
      "step": 5052,
      "time_per_iteration": 2.4396162033081055
    },
    {
      "auxiliary_loss_clip": 0.01145703,
      "auxiliary_loss_mlp": 0.01023482,
      "balance_loss_clip": 1.04607844,
      "balance_loss_mlp": 1.01618075,
      "epoch": 0.6075873263993267,
      "flos": 23769955572480.0,
      "grad_norm": 1.6721580359970072,
      "language_loss": 0.82650232,
      "learning_rate": 1.409142853097693e-06,
      "loss": 0.84819412,
      "num_input_tokens_seen": 108902710,
      "step": 5053,
      "time_per_iteration": 2.4990317821502686
    },
    {
      "auxiliary_loss_clip": 0.01147906,
      "auxiliary_loss_mlp": 0.01025782,
      "balance_loss_clip": 1.04817462,
      "balance_loss_mlp": 1.01827765,
      "epoch": 0.6077075692899657,
      "flos": 24454484035200.0,
      "grad_norm": 2.023040004575697,
      "language_loss": 0.79450166,
      "learning_rate": 1.408398695817553e-06,
      "loss": 0.81623852,
      "num_input_tokens_seen": 108919935,
      "step": 5054,
      "time_per_iteration": 2.5154056549072266
    },
    {
      "auxiliary_loss_clip": 0.01144591,
      "auxiliary_loss_mlp": 0.0103648,
      "balance_loss_clip": 1.04607213,
      "balance_loss_mlp": 1.02765226,
      "epoch": 0.6078278121806048,
      "flos": 27382102041600.0,
      "grad_norm": 1.5839432614312505,
      "language_loss": 0.70070207,
      "learning_rate": 1.4076546282828593e-06,
      "loss": 0.72251284,
      "num_input_tokens_seen": 108942790,
      "step": 5055,
      "time_per_iteration": 2.54927134513855
    },
    {
      "auxiliary_loss_clip": 0.0114762,
      "auxiliary_loss_mlp": 0.0102684,
      "balance_loss_clip": 1.04312754,
      "balance_loss_mlp": 1.01954412,
      "epoch": 0.6079480550712439,
      "flos": 38436447306240.0,
      "grad_norm": 2.9606739596897453,
      "language_loss": 0.65790534,
      "learning_rate": 1.4069106506064874e-06,
      "loss": 0.67964995,
      "num_input_tokens_seen": 108964215,
      "step": 5056,
      "time_per_iteration": 3.3528690338134766
    },
    {
      "auxiliary_loss_clip": 0.01141754,
      "auxiliary_loss_mlp": 0.01027874,
      "balance_loss_clip": 1.04757166,
      "balance_loss_mlp": 1.0200361,
      "epoch": 0.608068297961883,
      "flos": 25336271013120.0,
      "grad_norm": 2.5495524593936567,
      "language_loss": 0.78132248,
      "learning_rate": 1.4061667629012989e-06,
      "loss": 0.80301875,
      "num_input_tokens_seen": 108984885,
      "step": 5057,
      "time_per_iteration": 2.508213996887207
    },
    {
      "auxiliary_loss_clip": 0.01137808,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 1.04803705,
      "balance_loss_mlp": 1.01888633,
      "epoch": 0.608188540852522,
      "flos": 24202463235840.0,
      "grad_norm": 3.958121331552012,
      "language_loss": 0.83207607,
      "learning_rate": 1.40542296528014e-06,
      "loss": 0.85371894,
      "num_input_tokens_seen": 109004545,
      "step": 5058,
      "time_per_iteration": 2.5215721130371094
    },
    {
      "auxiliary_loss_clip": 0.01159327,
      "auxiliary_loss_mlp": 0.01032129,
      "balance_loss_clip": 1.04754245,
      "balance_loss_mlp": 1.02409744,
      "epoch": 0.6083087837431612,
      "flos": 21284146851840.0,
      "grad_norm": 2.121510698521352,
      "language_loss": 0.75962627,
      "learning_rate": 1.4046792578558452e-06,
      "loss": 0.78154087,
      "num_input_tokens_seen": 109022440,
      "step": 5059,
      "time_per_iteration": 3.316783905029297
    },
    {
      "auxiliary_loss_clip": 0.01141027,
      "auxiliary_loss_mlp": 0.0102892,
      "balance_loss_clip": 1.04591596,
      "balance_loss_mlp": 1.02095699,
      "epoch": 0.6084290266338003,
      "flos": 16471435178880.0,
      "grad_norm": 2.205578236021955,
      "language_loss": 0.76122385,
      "learning_rate": 1.4039356407412325e-06,
      "loss": 0.78292328,
      "num_input_tokens_seen": 109035680,
      "step": 5060,
      "time_per_iteration": 2.4452011585235596
    },
    {
      "auxiliary_loss_clip": 0.01066352,
      "auxiliary_loss_mlp": 0.0100103,
      "balance_loss_clip": 1.01757431,
      "balance_loss_mlp": 1.00011778,
      "epoch": 0.6085492695244393,
      "flos": 66443574931200.0,
      "grad_norm": 0.7881127065018196,
      "language_loss": 0.57123005,
      "learning_rate": 1.40319211404911e-06,
      "loss": 0.59190392,
      "num_input_tokens_seen": 109090680,
      "step": 5061,
      "time_per_iteration": 3.8978376388549805
    },
    {
      "auxiliary_loss_clip": 0.01174727,
      "auxiliary_loss_mlp": 0.01027046,
      "balance_loss_clip": 1.05089557,
      "balance_loss_mlp": 1.01908255,
      "epoch": 0.6086695124150785,
      "flos": 23618986709760.0,
      "grad_norm": 1.795289184050645,
      "language_loss": 0.9066298,
      "learning_rate": 1.4024486778922691e-06,
      "loss": 0.92864752,
      "num_input_tokens_seen": 109108995,
      "step": 5062,
      "time_per_iteration": 2.4421961307525635
    },
    {
      "auxiliary_loss_clip": 0.01149667,
      "auxiliary_loss_mlp": 0.01029942,
      "balance_loss_clip": 1.045959,
      "balance_loss_mlp": 1.02214026,
      "epoch": 0.6087897553057176,
      "flos": 20157054917760.0,
      "grad_norm": 1.8913959386274457,
      "language_loss": 0.77492893,
      "learning_rate": 1.4017053323834884e-06,
      "loss": 0.79672498,
      "num_input_tokens_seen": 109128825,
      "step": 5063,
      "time_per_iteration": 2.488363265991211
    },
    {
      "auxiliary_loss_clip": 0.01148366,
      "auxiliary_loss_mlp": 0.01025676,
      "balance_loss_clip": 1.04640567,
      "balance_loss_mlp": 1.01812387,
      "epoch": 0.6089099981963566,
      "flos": 25482535194240.0,
      "grad_norm": 1.933533855319585,
      "language_loss": 0.75814879,
      "learning_rate": 1.4009620776355333e-06,
      "loss": 0.77988923,
      "num_input_tokens_seen": 109150425,
      "step": 5064,
      "time_per_iteration": 3.2686140537261963
    },
    {
      "auxiliary_loss_clip": 0.01157879,
      "auxiliary_loss_mlp": 0.01022043,
      "balance_loss_clip": 1.04796982,
      "balance_loss_mlp": 1.01452076,
      "epoch": 0.6090302410869958,
      "flos": 25332895134720.0,
      "grad_norm": 1.763732955852719,
      "language_loss": 0.79360533,
      "learning_rate": 1.4002189137611553e-06,
      "loss": 0.81540453,
      "num_input_tokens_seen": 109169765,
      "step": 5065,
      "time_per_iteration": 2.4963266849517822
    },
    {
      "auxiliary_loss_clip": 0.01158074,
      "auxiliary_loss_mlp": 0.01024063,
      "balance_loss_clip": 1.04803658,
      "balance_loss_mlp": 1.01673794,
      "epoch": 0.6091504839776348,
      "flos": 23987358639360.0,
      "grad_norm": 1.6654030639640247,
      "language_loss": 0.69669372,
      "learning_rate": 1.3994758408730901e-06,
      "loss": 0.7185151,
      "num_input_tokens_seen": 109188950,
      "step": 5066,
      "time_per_iteration": 2.472099781036377
    },
    {
      "auxiliary_loss_clip": 0.01148095,
      "auxiliary_loss_mlp": 0.01024653,
      "balance_loss_clip": 1.0494163,
      "balance_loss_mlp": 1.0162847,
      "epoch": 0.6092707268682739,
      "flos": 29643037666560.0,
      "grad_norm": 2.7554103288725544,
      "language_loss": 0.76339543,
      "learning_rate": 1.3987328590840629e-06,
      "loss": 0.78512293,
      "num_input_tokens_seen": 109209895,
      "step": 5067,
      "time_per_iteration": 2.6054129600524902
    },
    {
      "auxiliary_loss_clip": 0.01156096,
      "auxiliary_loss_mlp": 0.01028028,
      "balance_loss_clip": 1.04747033,
      "balance_loss_mlp": 1.0207144,
      "epoch": 0.609390969758913,
      "flos": 24024957200640.0,
      "grad_norm": 2.075536064425432,
      "language_loss": 0.86368924,
      "learning_rate": 1.397989968506783e-06,
      "loss": 0.88553053,
      "num_input_tokens_seen": 109228905,
      "step": 5068,
      "time_per_iteration": 2.4662129878997803
    },
    {
      "auxiliary_loss_clip": 0.01179347,
      "auxiliary_loss_mlp": 0.01034572,
      "balance_loss_clip": 1.05269992,
      "balance_loss_mlp": 1.02661729,
      "epoch": 0.6095112126495521,
      "flos": 11102143288320.0,
      "grad_norm": 2.228408867935824,
      "language_loss": 0.72331738,
      "learning_rate": 1.3972471692539458e-06,
      "loss": 0.74545658,
      "num_input_tokens_seen": 109243620,
      "step": 5069,
      "time_per_iteration": 2.3966426849365234
    },
    {
      "auxiliary_loss_clip": 0.01142669,
      "auxiliary_loss_mlp": 0.01023374,
      "balance_loss_clip": 1.04760742,
      "balance_loss_mlp": 1.01561379,
      "epoch": 0.6096314555401912,
      "flos": 17265491187840.0,
      "grad_norm": 2.8268495863944274,
      "language_loss": 0.75498974,
      "learning_rate": 1.3965044614382348e-06,
      "loss": 0.77665019,
      "num_input_tokens_seen": 109259070,
      "step": 5070,
      "time_per_iteration": 2.445420742034912
    },
    {
      "auxiliary_loss_clip": 0.01177421,
      "auxiliary_loss_mlp": 0.0102623,
      "balance_loss_clip": 1.05192733,
      "balance_loss_mlp": 1.01843357,
      "epoch": 0.6097516984308303,
      "flos": 21645910679040.0,
      "grad_norm": 2.1283286011042972,
      "language_loss": 0.75576556,
      "learning_rate": 1.3957618451723162e-06,
      "loss": 0.77780211,
      "num_input_tokens_seen": 109275100,
      "step": 5071,
      "time_per_iteration": 2.426062822341919
    },
    {
      "auxiliary_loss_clip": 0.01146785,
      "auxiliary_loss_mlp": 0.01028045,
      "balance_loss_clip": 1.04809344,
      "balance_loss_mlp": 1.02062118,
      "epoch": 0.6098719413214694,
      "flos": 27199208966400.0,
      "grad_norm": 1.9461103588496604,
      "language_loss": 0.71213925,
      "learning_rate": 1.3950193205688457e-06,
      "loss": 0.73388755,
      "num_input_tokens_seen": 109294825,
      "step": 5072,
      "time_per_iteration": 2.5500707626342773
    },
    {
      "auxiliary_loss_clip": 0.01143449,
      "auxiliary_loss_mlp": 0.01025237,
      "balance_loss_clip": 1.04828525,
      "balance_loss_mlp": 1.01748812,
      "epoch": 0.6099921842121084,
      "flos": 20412954385920.0,
      "grad_norm": 1.839722369053823,
      "language_loss": 0.83545953,
      "learning_rate": 1.3942768877404627e-06,
      "loss": 0.85714644,
      "num_input_tokens_seen": 109313790,
      "step": 5073,
      "time_per_iteration": 2.476454973220825
    },
    {
      "auxiliary_loss_clip": 0.01172301,
      "auxiliary_loss_mlp": 0.01026507,
      "balance_loss_clip": 1.04937267,
      "balance_loss_mlp": 1.01946735,
      "epoch": 0.6101124271027476,
      "flos": 23366139897600.0,
      "grad_norm": 1.548529755196695,
      "language_loss": 0.73547733,
      "learning_rate": 1.393534546799795e-06,
      "loss": 0.75746548,
      "num_input_tokens_seen": 109333490,
      "step": 5074,
      "time_per_iteration": 2.4650776386260986
    },
    {
      "auxiliary_loss_clip": 0.01138827,
      "auxiliary_loss_mlp": 0.01030173,
      "balance_loss_clip": 1.04760671,
      "balance_loss_mlp": 1.0218401,
      "epoch": 0.6102326699933867,
      "flos": 26687840993280.0,
      "grad_norm": 1.7553216402234515,
      "language_loss": 0.67627013,
      "learning_rate": 1.3927922978594536e-06,
      "loss": 0.69796014,
      "num_input_tokens_seen": 109354575,
      "step": 5075,
      "time_per_iteration": 2.534072160720825
    },
    {
      "auxiliary_loss_clip": 0.01060701,
      "auxiliary_loss_mlp": 0.01001463,
      "balance_loss_clip": 1.01683462,
      "balance_loss_mlp": 1.00058126,
      "epoch": 0.6103529128840257,
      "flos": 60644612551680.0,
      "grad_norm": 0.771954237713556,
      "language_loss": 0.57446122,
      "learning_rate": 1.3920501410320387e-06,
      "loss": 0.59508288,
      "num_input_tokens_seen": 109410690,
      "step": 5076,
      "time_per_iteration": 2.9833619594573975
    },
    {
      "auxiliary_loss_clip": 0.01146146,
      "auxiliary_loss_mlp": 0.01026915,
      "balance_loss_clip": 1.04586792,
      "balance_loss_mlp": 1.01887441,
      "epoch": 0.6104731557746649,
      "flos": 19021307806080.0,
      "grad_norm": 2.046806922183719,
      "language_loss": 0.75950503,
      "learning_rate": 1.3913080764301333e-06,
      "loss": 0.78123569,
      "num_input_tokens_seen": 109427650,
      "step": 5077,
      "time_per_iteration": 2.460766553878784
    },
    {
      "auxiliary_loss_clip": 0.01126712,
      "auxiliary_loss_mlp": 0.01035234,
      "balance_loss_clip": 1.04381859,
      "balance_loss_mlp": 1.02750337,
      "epoch": 0.6105933986653039,
      "flos": 23366894083200.0,
      "grad_norm": 1.9358012460082523,
      "language_loss": 0.7128607,
      "learning_rate": 1.3905661041663085e-06,
      "loss": 0.73448014,
      "num_input_tokens_seen": 109448835,
      "step": 5078,
      "time_per_iteration": 2.560464859008789
    },
    {
      "auxiliary_loss_clip": 0.01161571,
      "auxiliary_loss_mlp": 0.01033974,
      "balance_loss_clip": 1.05051482,
      "balance_loss_mlp": 1.02599907,
      "epoch": 0.610713641555943,
      "flos": 34637565006720.0,
      "grad_norm": 2.2979424204514616,
      "language_loss": 0.65208471,
      "learning_rate": 1.389824224353122e-06,
      "loss": 0.67404014,
      "num_input_tokens_seen": 109470425,
      "step": 5079,
      "time_per_iteration": 2.5610344409942627
    },
    {
      "auxiliary_loss_clip": 0.0115995,
      "auxiliary_loss_mlp": 0.01023647,
      "balance_loss_clip": 1.05076504,
      "balance_loss_mlp": 1.0160296,
      "epoch": 0.610833884446582,
      "flos": 26646471504000.0,
      "grad_norm": 1.644420073547241,
      "language_loss": 0.76982391,
      "learning_rate": 1.389082437103115e-06,
      "loss": 0.79165983,
      "num_input_tokens_seen": 109489695,
      "step": 5080,
      "time_per_iteration": 2.4779303073883057
    },
    {
      "auxiliary_loss_clip": 0.01129828,
      "auxiliary_loss_mlp": 0.01025592,
      "balance_loss_clip": 1.04391813,
      "balance_loss_mlp": 1.01730132,
      "epoch": 0.6109541273372212,
      "flos": 21215126868480.0,
      "grad_norm": 2.9723896096586984,
      "language_loss": 0.77760744,
      "learning_rate": 1.3883407425288172e-06,
      "loss": 0.79916167,
      "num_input_tokens_seen": 109510030,
      "step": 5081,
      "time_per_iteration": 2.553284168243408
    },
    {
      "auxiliary_loss_clip": 0.01142196,
      "auxiliary_loss_mlp": 0.01027912,
      "balance_loss_clip": 1.04558182,
      "balance_loss_mlp": 1.02015734,
      "epoch": 0.6110743702278603,
      "flos": 20084084438400.0,
      "grad_norm": 2.133305995274337,
      "language_loss": 0.79975712,
      "learning_rate": 1.3875991407427417e-06,
      "loss": 0.82145822,
      "num_input_tokens_seen": 109528255,
      "step": 5082,
      "time_per_iteration": 3.2947733402252197
    },
    {
      "auxiliary_loss_clip": 0.01046629,
      "auxiliary_loss_mlp": 0.01002104,
      "balance_loss_clip": 1.01670015,
      "balance_loss_mlp": 1.00110853,
      "epoch": 0.6111946131184993,
      "flos": 68302957438080.0,
      "grad_norm": 0.7707187304039955,
      "language_loss": 0.58226377,
      "learning_rate": 1.38685763185739e-06,
      "loss": 0.60275108,
      "num_input_tokens_seen": 109581915,
      "step": 5083,
      "time_per_iteration": 3.105914831161499
    },
    {
      "auxiliary_loss_clip": 0.01173656,
      "auxiliary_loss_mlp": 0.01026081,
      "balance_loss_clip": 1.05030227,
      "balance_loss_mlp": 1.01804602,
      "epoch": 0.6113148560091385,
      "flos": 19937676602880.0,
      "grad_norm": 2.415949856770844,
      "language_loss": 0.67725444,
      "learning_rate": 1.3861162159852476e-06,
      "loss": 0.69925183,
      "num_input_tokens_seen": 109600050,
      "step": 5084,
      "time_per_iteration": 2.410493850708008
    },
    {
      "auxiliary_loss_clip": 0.01151265,
      "auxiliary_loss_mlp": 0.01026259,
      "balance_loss_clip": 1.04899645,
      "balance_loss_mlp": 1.01778293,
      "epoch": 0.6114350988997775,
      "flos": 23731854220800.0,
      "grad_norm": 2.6658033470219458,
      "language_loss": 0.80039108,
      "learning_rate": 1.3853748932387875e-06,
      "loss": 0.82216632,
      "num_input_tokens_seen": 109620690,
      "step": 5085,
      "time_per_iteration": 3.3655753135681152
    },
    {
      "auxiliary_loss_clip": 0.0113418,
      "auxiliary_loss_mlp": 0.01021389,
      "balance_loss_clip": 1.04547048,
      "balance_loss_mlp": 1.01365805,
      "epoch": 0.6115553417904166,
      "flos": 24023700224640.0,
      "grad_norm": 2.353728573836011,
      "language_loss": 0.75099081,
      "learning_rate": 1.3846336637304671e-06,
      "loss": 0.77254653,
      "num_input_tokens_seen": 109638960,
      "step": 5086,
      "time_per_iteration": 2.484530210494995
    },
    {
      "auxiliary_loss_clip": 0.01141938,
      "auxiliary_loss_mlp": 0.01023628,
      "balance_loss_clip": 1.04962564,
      "balance_loss_mlp": 1.01587987,
      "epoch": 0.6116755846810558,
      "flos": 23733542160000.0,
      "grad_norm": 1.8829038915104588,
      "language_loss": 0.83109522,
      "learning_rate": 1.3838925275727316e-06,
      "loss": 0.8527509,
      "num_input_tokens_seen": 109659700,
      "step": 5087,
      "time_per_iteration": 2.5030009746551514
    },
    {
      "auxiliary_loss_clip": 0.01175185,
      "auxiliary_loss_mlp": 0.01024154,
      "balance_loss_clip": 1.05197108,
      "balance_loss_mlp": 1.01677525,
      "epoch": 0.6117958275716948,
      "flos": 18661626967680.0,
      "grad_norm": 2.269638300784814,
      "language_loss": 0.79136729,
      "learning_rate": 1.3831514848780089e-06,
      "loss": 0.81336069,
      "num_input_tokens_seen": 109679275,
      "step": 5088,
      "time_per_iteration": 2.4074864387512207
    },
    {
      "auxiliary_loss_clip": 0.01154594,
      "auxiliary_loss_mlp": 0.01027564,
      "balance_loss_clip": 1.04809511,
      "balance_loss_mlp": 1.02032161,
      "epoch": 0.6119160704623339,
      "flos": 16471183783680.0,
      "grad_norm": 2.675359491424576,
      "language_loss": 0.91493845,
      "learning_rate": 1.3824105357587152e-06,
      "loss": 0.93676007,
      "num_input_tokens_seen": 109696380,
      "step": 5089,
      "time_per_iteration": 3.2116165161132812
    },
    {
      "auxiliary_loss_clip": 0.01140288,
      "auxiliary_loss_mlp": 0.01026302,
      "balance_loss_clip": 1.04482126,
      "balance_loss_mlp": 1.01857138,
      "epoch": 0.612036313352973,
      "flos": 23915465568000.0,
      "grad_norm": 1.5189165546542074,
      "language_loss": 0.82676542,
      "learning_rate": 1.381669680327253e-06,
      "loss": 0.84843129,
      "num_input_tokens_seen": 109718060,
      "step": 5090,
      "time_per_iteration": 3.291295051574707
    },
    {
      "auxiliary_loss_clip": 0.01141033,
      "auxiliary_loss_mlp": 0.0102603,
      "balance_loss_clip": 1.04912925,
      "balance_loss_mlp": 1.01788795,
      "epoch": 0.6121565562436121,
      "flos": 26974766833920.0,
      "grad_norm": 1.9626741935955292,
      "language_loss": 0.7062344,
      "learning_rate": 1.380928918696008e-06,
      "loss": 0.72790504,
      "num_input_tokens_seen": 109736830,
      "step": 5091,
      "time_per_iteration": 2.530230760574341
    },
    {
      "auxiliary_loss_clip": 0.011591,
      "auxiliary_loss_mlp": 0.01025017,
      "balance_loss_clip": 1.04774833,
      "balance_loss_mlp": 1.01716101,
      "epoch": 0.6122767991342511,
      "flos": 15668867646720.0,
      "grad_norm": 3.1379894187401,
      "language_loss": 0.71554208,
      "learning_rate": 1.3801882509773548e-06,
      "loss": 0.73738325,
      "num_input_tokens_seen": 109754690,
      "step": 5092,
      "time_per_iteration": 2.4204964637756348
    },
    {
      "auxiliary_loss_clip": 0.01154851,
      "auxiliary_loss_mlp": 0.01025703,
      "balance_loss_clip": 1.04695797,
      "balance_loss_mlp": 1.01766789,
      "epoch": 0.6123970420248903,
      "flos": 27964321591680.0,
      "grad_norm": 4.2475396185683,
      "language_loss": 0.81701678,
      "learning_rate": 1.3794476772836503e-06,
      "loss": 0.83882236,
      "num_input_tokens_seen": 109775790,
      "step": 5093,
      "time_per_iteration": 2.506866693496704
    },
    {
      "auxiliary_loss_clip": 0.01126149,
      "auxiliary_loss_mlp": 0.01029923,
      "balance_loss_clip": 1.04717636,
      "balance_loss_mlp": 1.02154219,
      "epoch": 0.6125172849155294,
      "flos": 21468727866240.0,
      "grad_norm": 1.6608983218580216,
      "language_loss": 0.84496176,
      "learning_rate": 1.3787071977272402e-06,
      "loss": 0.86652255,
      "num_input_tokens_seen": 109795050,
      "step": 5094,
      "time_per_iteration": 2.5703883171081543
    },
    {
      "auxiliary_loss_clip": 0.01112877,
      "auxiliary_loss_mlp": 0.01030748,
      "balance_loss_clip": 1.0462805,
      "balance_loss_mlp": 1.02293372,
      "epoch": 0.6126375278061684,
      "flos": 16248321849600.0,
      "grad_norm": 3.09254998674022,
      "language_loss": 0.71797681,
      "learning_rate": 1.3779668124204535e-06,
      "loss": 0.73941302,
      "num_input_tokens_seen": 109811465,
      "step": 5095,
      "time_per_iteration": 2.509732484817505
    },
    {
      "auxiliary_loss_clip": 0.01141317,
      "auxiliary_loss_mlp": 0.01027106,
      "balance_loss_clip": 1.04918694,
      "balance_loss_mlp": 1.01908934,
      "epoch": 0.6127577706968076,
      "flos": 20448865008000.0,
      "grad_norm": 1.501077716527644,
      "language_loss": 0.80910319,
      "learning_rate": 1.3772265214756074e-06,
      "loss": 0.83078742,
      "num_input_tokens_seen": 109831225,
      "step": 5096,
      "time_per_iteration": 2.4700565338134766
    },
    {
      "auxiliary_loss_clip": 0.01161292,
      "auxiliary_loss_mlp": 0.01028791,
      "balance_loss_clip": 1.04705644,
      "balance_loss_mlp": 1.0211854,
      "epoch": 0.6128780135874466,
      "flos": 18260397072000.0,
      "grad_norm": 1.8857198553529981,
      "language_loss": 0.75419867,
      "learning_rate": 1.3764863250050025e-06,
      "loss": 0.77609956,
      "num_input_tokens_seen": 109849465,
      "step": 5097,
      "time_per_iteration": 2.4436445236206055
    },
    {
      "auxiliary_loss_clip": 0.01132261,
      "auxiliary_loss_mlp": 0.01028296,
      "balance_loss_clip": 1.04572415,
      "balance_loss_mlp": 1.02087569,
      "epoch": 0.6129982564780857,
      "flos": 24937088192640.0,
      "grad_norm": 1.792018576268544,
      "language_loss": 0.80602854,
      "learning_rate": 1.3757462231209272e-06,
      "loss": 0.8276341,
      "num_input_tokens_seen": 109869770,
      "step": 5098,
      "time_per_iteration": 2.5587480068206787
    },
    {
      "auxiliary_loss_clip": 0.01139473,
      "auxiliary_loss_mlp": 0.01024067,
      "balance_loss_clip": 1.04609632,
      "balance_loss_mlp": 1.01593113,
      "epoch": 0.6131184993687249,
      "flos": 22492038430080.0,
      "grad_norm": 1.983903909425592,
      "language_loss": 0.88618672,
      "learning_rate": 1.3750062159356525e-06,
      "loss": 0.90782213,
      "num_input_tokens_seen": 109889120,
      "step": 5099,
      "time_per_iteration": 2.504131555557251
    },
    {
      "auxiliary_loss_clip": 0.0112095,
      "auxiliary_loss_mlp": 0.01027434,
      "balance_loss_clip": 1.04404151,
      "balance_loss_mlp": 1.01998353,
      "epoch": 0.6132387422593639,
      "flos": 15885839750400.0,
      "grad_norm": 1.6749461859328214,
      "language_loss": 0.83296037,
      "learning_rate": 1.3742663035614382e-06,
      "loss": 0.85444415,
      "num_input_tokens_seen": 109906490,
      "step": 5100,
      "time_per_iteration": 2.482245683670044
    },
    {
      "auxiliary_loss_clip": 0.01175817,
      "auxiliary_loss_mlp": 0.01030878,
      "balance_loss_clip": 1.05095983,
      "balance_loss_mlp": 1.0227834,
      "epoch": 0.613358985150003,
      "flos": 25411539962880.0,
      "grad_norm": 3.079416876949993,
      "language_loss": 0.80385959,
      "learning_rate": 1.3735264861105283e-06,
      "loss": 0.82592654,
      "num_input_tokens_seen": 109927130,
      "step": 5101,
      "time_per_iteration": 2.4548351764678955
    },
    {
      "auxiliary_loss_clip": 0.01133169,
      "auxiliary_loss_mlp": 0.01026517,
      "balance_loss_clip": 1.04508257,
      "balance_loss_mlp": 1.01901579,
      "epoch": 0.6134792280406421,
      "flos": 21361283308800.0,
      "grad_norm": 2.121434392117287,
      "language_loss": 0.78416133,
      "learning_rate": 1.372786763695152e-06,
      "loss": 0.80575818,
      "num_input_tokens_seen": 109945890,
      "step": 5102,
      "time_per_iteration": 2.516022205352783
    },
    {
      "auxiliary_loss_clip": 0.01160849,
      "auxiliary_loss_mlp": 0.01031982,
      "balance_loss_clip": 1.04775882,
      "balance_loss_mlp": 1.024001,
      "epoch": 0.6135994709312812,
      "flos": 21211248199680.0,
      "grad_norm": 3.0606831251367246,
      "language_loss": 0.77118242,
      "learning_rate": 1.3720471364275257e-06,
      "loss": 0.79311073,
      "num_input_tokens_seen": 109965535,
      "step": 5103,
      "time_per_iteration": 2.4676713943481445
    },
    {
      "auxiliary_loss_clip": 0.0112764,
      "auxiliary_loss_mlp": 0.00762904,
      "balance_loss_clip": 1.04507315,
      "balance_loss_mlp": 1.0007174,
      "epoch": 0.6137197138219203,
      "flos": 14794047907200.0,
      "grad_norm": 2.074752507963412,
      "language_loss": 0.78232431,
      "learning_rate": 1.3713076044198486e-06,
      "loss": 0.80122972,
      "num_input_tokens_seen": 109982345,
      "step": 5104,
      "time_per_iteration": 2.4856326580047607
    },
    {
      "auxiliary_loss_clip": 0.01140532,
      "auxiliary_loss_mlp": 0.01031838,
      "balance_loss_clip": 1.04648161,
      "balance_loss_mlp": 1.02377343,
      "epoch": 0.6138399567125594,
      "flos": 20084515401600.0,
      "grad_norm": 2.6463185183969746,
      "language_loss": 0.81177032,
      "learning_rate": 1.3705681677843086e-06,
      "loss": 0.83349407,
      "num_input_tokens_seen": 110000940,
      "step": 5105,
      "time_per_iteration": 2.4687111377716064
    },
    {
      "auxiliary_loss_clip": 0.01073599,
      "auxiliary_loss_mlp": 0.01001425,
      "balance_loss_clip": 1.01608825,
      "balance_loss_mlp": 1.00047779,
      "epoch": 0.6139601996031985,
      "flos": 60123838193280.0,
      "grad_norm": 0.8062958448978689,
      "language_loss": 0.60640794,
      "learning_rate": 1.3698288266330768e-06,
      "loss": 0.62715822,
      "num_input_tokens_seen": 110061565,
      "step": 5106,
      "time_per_iteration": 3.095419406890869
    },
    {
      "auxiliary_loss_clip": 0.01143363,
      "auxiliary_loss_mlp": 0.01023012,
      "balance_loss_clip": 1.05189347,
      "balance_loss_mlp": 1.01555526,
      "epoch": 0.6140804424938375,
      "flos": 23586703361280.0,
      "grad_norm": 3.761532254880022,
      "language_loss": 0.72768337,
      "learning_rate": 1.3690895810783113e-06,
      "loss": 0.74934709,
      "num_input_tokens_seen": 110080360,
      "step": 5107,
      "time_per_iteration": 2.5079920291900635
    },
    {
      "auxiliary_loss_clip": 0.01111024,
      "auxiliary_loss_mlp": 0.00762779,
      "balance_loss_clip": 1.0410949,
      "balance_loss_mlp": 1.00069571,
      "epoch": 0.6142006853844767,
      "flos": 21398199511680.0,
      "grad_norm": 2.859343515641965,
      "language_loss": 0.71429312,
      "learning_rate": 1.3683504312321543e-06,
      "loss": 0.73303109,
      "num_input_tokens_seen": 110100695,
      "step": 5108,
      "time_per_iteration": 3.3253512382507324
    },
    {
      "auxiliary_loss_clip": 0.01164364,
      "auxiliary_loss_mlp": 0.01027314,
      "balance_loss_clip": 1.04948795,
      "balance_loss_mlp": 1.01930869,
      "epoch": 0.6143209282751158,
      "flos": 12057367622400.0,
      "grad_norm": 2.2641694690977325,
      "language_loss": 0.80268037,
      "learning_rate": 1.3676113772067355e-06,
      "loss": 0.82459706,
      "num_input_tokens_seen": 110117750,
      "step": 5109,
      "time_per_iteration": 2.4397342205047607
    },
    {
      "auxiliary_loss_clip": 0.01122927,
      "auxiliary_loss_mlp": 0.01024662,
      "balance_loss_clip": 1.04479313,
      "balance_loss_mlp": 1.01677656,
      "epoch": 0.6144411711657548,
      "flos": 25082274965760.0,
      "grad_norm": 1.7993815904853432,
      "language_loss": 0.72608173,
      "learning_rate": 1.3668724191141671e-06,
      "loss": 0.74755764,
      "num_input_tokens_seen": 110137020,
      "step": 5110,
      "time_per_iteration": 2.574392557144165
    },
    {
      "auxiliary_loss_clip": 0.01131348,
      "auxiliary_loss_mlp": 0.01033588,
      "balance_loss_clip": 1.05282533,
      "balance_loss_mlp": 1.02523756,
      "epoch": 0.6145614140563939,
      "flos": 20114069316480.0,
      "grad_norm": 2.168741696093098,
      "language_loss": 0.6672402,
      "learning_rate": 1.3661335570665493e-06,
      "loss": 0.68888962,
      "num_input_tokens_seen": 110154930,
      "step": 5111,
      "time_per_iteration": 2.508439302444458
    },
    {
      "auxiliary_loss_clip": 0.01151295,
      "auxiliary_loss_mlp": 0.01028893,
      "balance_loss_clip": 1.05130553,
      "balance_loss_mlp": 1.02132928,
      "epoch": 0.614681656947033,
      "flos": 16800376953600.0,
      "grad_norm": 2.4082095880795484,
      "language_loss": 0.70047927,
      "learning_rate": 1.3653947911759676e-06,
      "loss": 0.72228116,
      "num_input_tokens_seen": 110172480,
      "step": 5112,
      "time_per_iteration": 3.3108434677124023
    },
    {
      "auxiliary_loss_clip": 0.01110389,
      "auxiliary_loss_mlp": 0.01032534,
      "balance_loss_clip": 1.04393744,
      "balance_loss_mlp": 1.0240581,
      "epoch": 0.6148018998376721,
      "flos": 38801587011840.0,
      "grad_norm": 1.9050274812869563,
      "language_loss": 0.74706751,
      "learning_rate": 1.3646561215544904e-06,
      "loss": 0.76849675,
      "num_input_tokens_seen": 110197120,
      "step": 5113,
      "time_per_iteration": 2.7064969539642334
    },
    {
      "auxiliary_loss_clip": 0.01161413,
      "auxiliary_loss_mlp": 0.01024166,
      "balance_loss_clip": 1.05038035,
      "balance_loss_mlp": 1.01647115,
      "epoch": 0.6149221427283111,
      "flos": 23327032965120.0,
      "grad_norm": 2.196896747359998,
      "language_loss": 0.79495418,
      "learning_rate": 1.363917548314176e-06,
      "loss": 0.81681001,
      "num_input_tokens_seen": 110216385,
      "step": 5114,
      "time_per_iteration": 2.507715940475464
    },
    {
      "auxiliary_loss_clip": 0.01167551,
      "auxiliary_loss_mlp": 0.01028041,
      "balance_loss_clip": 1.04987669,
      "balance_loss_mlp": 1.02024519,
      "epoch": 0.6150423856189503,
      "flos": 22379494141440.0,
      "grad_norm": 1.6859480971957332,
      "language_loss": 0.73069608,
      "learning_rate": 1.3631790715670626e-06,
      "loss": 0.75265205,
      "num_input_tokens_seen": 110234790,
      "step": 5115,
      "time_per_iteration": 3.320070266723633
    },
    {
      "auxiliary_loss_clip": 0.01080253,
      "auxiliary_loss_mlp": 0.01024134,
      "balance_loss_clip": 1.04433727,
      "balance_loss_mlp": 1.01688588,
      "epoch": 0.6151626285095894,
      "flos": 18692078722560.0,
      "grad_norm": 1.8132990081572578,
      "language_loss": 0.85411805,
      "learning_rate": 1.3624406914251783e-06,
      "loss": 0.87516189,
      "num_input_tokens_seen": 110251910,
      "step": 5116,
      "time_per_iteration": 2.614272356033325
    },
    {
      "auxiliary_loss_clip": 0.011616,
      "auxiliary_loss_mlp": 0.01028105,
      "balance_loss_clip": 1.04844487,
      "balance_loss_mlp": 1.0209291,
      "epoch": 0.6152828714002284,
      "flos": 15851688894720.0,
      "grad_norm": 1.875999404706754,
      "language_loss": 0.87926996,
      "learning_rate": 1.3617024080005335e-06,
      "loss": 0.90116704,
      "num_input_tokens_seen": 110268810,
      "step": 5117,
      "time_per_iteration": 3.205634593963623
    },
    {
      "auxiliary_loss_clip": 0.01148334,
      "auxiliary_loss_mlp": 0.00762388,
      "balance_loss_clip": 1.04618704,
      "balance_loss_mlp": 1.00068903,
      "epoch": 0.6154031142908676,
      "flos": 24869792062080.0,
      "grad_norm": 1.5118296874236248,
      "language_loss": 0.74606073,
      "learning_rate": 1.3609642214051266e-06,
      "loss": 0.76516789,
      "num_input_tokens_seen": 110293035,
      "step": 5118,
      "time_per_iteration": 2.586886405944824
    },
    {
      "auxiliary_loss_clip": 0.01143076,
      "auxiliary_loss_mlp": 0.01031343,
      "balance_loss_clip": 1.05111957,
      "balance_loss_mlp": 1.02277803,
      "epoch": 0.6155233571815066,
      "flos": 19244744357760.0,
      "grad_norm": 1.8796245593984653,
      "language_loss": 0.66211236,
      "learning_rate": 1.3602261317509385e-06,
      "loss": 0.68385661,
      "num_input_tokens_seen": 110309695,
      "step": 5119,
      "time_per_iteration": 2.4761598110198975
    },
    {
      "auxiliary_loss_clip": 0.0116176,
      "auxiliary_loss_mlp": 0.01025005,
      "balance_loss_clip": 1.04947948,
      "balance_loss_mlp": 1.01654756,
      "epoch": 0.6156436000721457,
      "flos": 18770077105920.0,
      "grad_norm": 2.692538032462891,
      "language_loss": 0.82466644,
      "learning_rate": 1.3594881391499387e-06,
      "loss": 0.84653413,
      "num_input_tokens_seen": 110328610,
      "step": 5120,
      "time_per_iteration": 2.4384398460388184
    },
    {
      "auxiliary_loss_clip": 0.01149388,
      "auxiliary_loss_mlp": 0.01026951,
      "balance_loss_clip": 1.04891694,
      "balance_loss_mlp": 1.0190537,
      "epoch": 0.6157638429627849,
      "flos": 18041198325120.0,
      "grad_norm": 2.0166243919535636,
      "language_loss": 0.79189581,
      "learning_rate": 1.3587502437140778e-06,
      "loss": 0.81365919,
      "num_input_tokens_seen": 110346775,
      "step": 5121,
      "time_per_iteration": 2.4741883277893066
    },
    {
      "auxiliary_loss_clip": 0.01149189,
      "auxiliary_loss_mlp": 0.01029567,
      "balance_loss_clip": 1.04741859,
      "balance_loss_mlp": 1.02150834,
      "epoch": 0.6158840858534239,
      "flos": 25556726736000.0,
      "grad_norm": 2.2288144063697803,
      "language_loss": 0.85067892,
      "learning_rate": 1.3580124455552952e-06,
      "loss": 0.87246656,
      "num_input_tokens_seen": 110366140,
      "step": 5122,
      "time_per_iteration": 2.5107157230377197
    },
    {
      "auxiliary_loss_clip": 0.01161623,
      "auxiliary_loss_mlp": 0.00761794,
      "balance_loss_clip": 1.0498805,
      "balance_loss_mlp": 1.00073385,
      "epoch": 0.616004328744063,
      "flos": 24640788902400.0,
      "grad_norm": 1.6461498777306336,
      "language_loss": 0.87382662,
      "learning_rate": 1.3572747447855148e-06,
      "loss": 0.8930608,
      "num_input_tokens_seen": 110386550,
      "step": 5123,
      "time_per_iteration": 2.4893503189086914
    },
    {
      "auxiliary_loss_clip": 0.01178223,
      "auxiliary_loss_mlp": 0.0102792,
      "balance_loss_clip": 1.05366778,
      "balance_loss_mlp": 1.01996326,
      "epoch": 0.6161245716347021,
      "flos": 21689686379520.0,
      "grad_norm": 2.2205169980477075,
      "language_loss": 0.69211447,
      "learning_rate": 1.356537141516644e-06,
      "loss": 0.71417594,
      "num_input_tokens_seen": 110403970,
      "step": 5124,
      "time_per_iteration": 2.4197683334350586
    },
    {
      "auxiliary_loss_clip": 0.01162462,
      "auxiliary_loss_mlp": 0.01026927,
      "balance_loss_clip": 1.05254543,
      "balance_loss_mlp": 1.01904678,
      "epoch": 0.6162448145253412,
      "flos": 35189225061120.0,
      "grad_norm": 2.1713747739713054,
      "language_loss": 0.61966944,
      "learning_rate": 1.3557996358605775e-06,
      "loss": 0.6415633,
      "num_input_tokens_seen": 110423890,
      "step": 5125,
      "time_per_iteration": 2.5521202087402344
    },
    {
      "auxiliary_loss_clip": 0.01159368,
      "auxiliary_loss_mlp": 0.01031449,
      "balance_loss_clip": 1.04893374,
      "balance_loss_mlp": 1.02369475,
      "epoch": 0.6163650574159802,
      "flos": 21615279356160.0,
      "grad_norm": 2.137293296134422,
      "language_loss": 0.69733453,
      "learning_rate": 1.3550622279291941e-06,
      "loss": 0.71924269,
      "num_input_tokens_seen": 110442035,
      "step": 5126,
      "time_per_iteration": 2.4518954753875732
    },
    {
      "auxiliary_loss_clip": 0.01108061,
      "auxiliary_loss_mlp": 0.01026371,
      "balance_loss_clip": 1.04277468,
      "balance_loss_mlp": 1.01847303,
      "epoch": 0.6164853003066194,
      "flos": 24572163968640.0,
      "grad_norm": 1.4525271136953573,
      "language_loss": 0.83256698,
      "learning_rate": 1.354324917834358e-06,
      "loss": 0.85391128,
      "num_input_tokens_seen": 110463280,
      "step": 5127,
      "time_per_iteration": 2.5621418952941895
    },
    {
      "auxiliary_loss_clip": 0.01102617,
      "auxiliary_loss_mlp": 0.00762848,
      "balance_loss_clip": 1.04464948,
      "balance_loss_mlp": 1.00073767,
      "epoch": 0.6166055431972585,
      "flos": 21835986474240.0,
      "grad_norm": 1.740837685526935,
      "language_loss": 0.7687853,
      "learning_rate": 1.353587705687918e-06,
      "loss": 0.78744,
      "num_input_tokens_seen": 110481455,
      "step": 5128,
      "time_per_iteration": 2.589176654815674
    },
    {
      "auxiliary_loss_clip": 0.01152411,
      "auxiliary_loss_mlp": 0.01026174,
      "balance_loss_clip": 1.05061007,
      "balance_loss_mlp": 1.01831222,
      "epoch": 0.6167257860878975,
      "flos": 17785262943360.0,
      "grad_norm": 18.805756662928104,
      "language_loss": 0.71780175,
      "learning_rate": 1.3528505916017096e-06,
      "loss": 0.73958755,
      "num_input_tokens_seen": 110499155,
      "step": 5129,
      "time_per_iteration": 2.4418556690216064
    },
    {
      "auxiliary_loss_clip": 0.01160673,
      "auxiliary_loss_mlp": 0.01030876,
      "balance_loss_clip": 1.04801965,
      "balance_loss_mlp": 1.02291274,
      "epoch": 0.6168460289785367,
      "flos": 23214811898880.0,
      "grad_norm": 1.9956938296100017,
      "language_loss": 0.88457108,
      "learning_rate": 1.3521135756875514e-06,
      "loss": 0.90648663,
      "num_input_tokens_seen": 110515470,
      "step": 5130,
      "time_per_iteration": 2.4279496669769287
    },
    {
      "auxiliary_loss_clip": 0.01096143,
      "auxiliary_loss_mlp": 0.01027447,
      "balance_loss_clip": 1.04391456,
      "balance_loss_mlp": 1.02001441,
      "epoch": 0.6169662718691757,
      "flos": 26213281482240.0,
      "grad_norm": 1.5922104159147648,
      "language_loss": 0.8646912,
      "learning_rate": 1.3513766580572496e-06,
      "loss": 0.88592708,
      "num_input_tokens_seen": 110538290,
      "step": 5131,
      "time_per_iteration": 2.64207124710083
    },
    {
      "auxiliary_loss_clip": 0.01158315,
      "auxiliary_loss_mlp": 0.01025482,
      "balance_loss_clip": 1.0488596,
      "balance_loss_mlp": 1.01830602,
      "epoch": 0.6170865147598148,
      "flos": 19026120228480.0,
      "grad_norm": 2.001301953903268,
      "language_loss": 0.77359492,
      "learning_rate": 1.3506398388225924e-06,
      "loss": 0.79543287,
      "num_input_tokens_seen": 110555610,
      "step": 5132,
      "time_per_iteration": 2.4383535385131836
    },
    {
      "auxiliary_loss_clip": 0.0117407,
      "auxiliary_loss_mlp": 0.0102423,
      "balance_loss_clip": 1.05286646,
      "balance_loss_mlp": 1.01690412,
      "epoch": 0.617206757650454,
      "flos": 18260361158400.0,
      "grad_norm": 1.8237794699146703,
      "language_loss": 0.71642679,
      "learning_rate": 1.349903118095355e-06,
      "loss": 0.73840976,
      "num_input_tokens_seen": 110574745,
      "step": 5133,
      "time_per_iteration": 2.411007881164551
    },
    {
      "auxiliary_loss_clip": 0.0116385,
      "auxiliary_loss_mlp": 0.01029153,
      "balance_loss_clip": 1.04898012,
      "balance_loss_mlp": 1.02131474,
      "epoch": 0.617327000541093,
      "flos": 18186959715840.0,
      "grad_norm": 1.634229230515609,
      "language_loss": 0.73205948,
      "learning_rate": 1.349166495987298e-06,
      "loss": 0.75398946,
      "num_input_tokens_seen": 110593310,
      "step": 5134,
      "time_per_iteration": 2.4359846115112305
    },
    {
      "auxiliary_loss_clip": 0.01061904,
      "auxiliary_loss_mlp": 0.01021128,
      "balance_loss_clip": 1.02895141,
      "balance_loss_mlp": 1.02000737,
      "epoch": 0.6174472434317321,
      "flos": 61833796122240.0,
      "grad_norm": 0.8233008660393084,
      "language_loss": 0.60909635,
      "learning_rate": 1.348429972610166e-06,
      "loss": 0.62992668,
      "num_input_tokens_seen": 110657615,
      "step": 5135,
      "time_per_iteration": 3.934006929397583
    },
    {
      "auxiliary_loss_clip": 0.01037221,
      "auxiliary_loss_mlp": 0.01008393,
      "balance_loss_clip": 1.03129709,
      "balance_loss_mlp": 1.00718272,
      "epoch": 0.6175674863223712,
      "flos": 71230970494080.0,
      "grad_norm": 0.8490553402599413,
      "language_loss": 0.57908899,
      "learning_rate": 1.3476935480756897e-06,
      "loss": 0.59954512,
      "num_input_tokens_seen": 110714365,
      "step": 5136,
      "time_per_iteration": 3.005431890487671
    },
    {
      "auxiliary_loss_clip": 0.01121309,
      "auxiliary_loss_mlp": 0.01033753,
      "balance_loss_clip": 1.0434444,
      "balance_loss_mlp": 1.0258677,
      "epoch": 0.6176877292130103,
      "flos": 21835447770240.0,
      "grad_norm": 2.03787729210879,
      "language_loss": 0.756091,
      "learning_rate": 1.346957222495583e-06,
      "loss": 0.77764165,
      "num_input_tokens_seen": 110732160,
      "step": 5137,
      "time_per_iteration": 2.5312812328338623
    },
    {
      "auxiliary_loss_clip": 0.01151661,
      "auxiliary_loss_mlp": 0.00762518,
      "balance_loss_clip": 1.04998922,
      "balance_loss_mlp": 1.00070214,
      "epoch": 0.6178079721036493,
      "flos": 17741738638080.0,
      "grad_norm": 3.0990128599154794,
      "language_loss": 0.71393383,
      "learning_rate": 1.3462209959815466e-06,
      "loss": 0.73307556,
      "num_input_tokens_seen": 110746900,
      "step": 5138,
      "time_per_iteration": 2.4329543113708496
    },
    {
      "auxiliary_loss_clip": 0.01151018,
      "auxiliary_loss_mlp": 0.01027787,
      "balance_loss_clip": 1.05056465,
      "balance_loss_mlp": 1.01985025,
      "epoch": 0.6179282149942885,
      "flos": 22633131052800.0,
      "grad_norm": 1.7254070638110282,
      "language_loss": 0.74007356,
      "learning_rate": 1.345484868645265e-06,
      "loss": 0.76186162,
      "num_input_tokens_seen": 110765710,
      "step": 5139,
      "time_per_iteration": 3.324608087539673
    },
    {
      "auxiliary_loss_clip": 0.01140696,
      "auxiliary_loss_mlp": 0.01030782,
      "balance_loss_clip": 1.04719186,
      "balance_loss_mlp": 1.02263427,
      "epoch": 0.6180484578849276,
      "flos": 22310330503680.0,
      "grad_norm": 1.93353122568057,
      "language_loss": 0.78708959,
      "learning_rate": 1.3447488405984088e-06,
      "loss": 0.80880433,
      "num_input_tokens_seen": 110783970,
      "step": 5140,
      "time_per_iteration": 2.5369367599487305
    },
    {
      "auxiliary_loss_clip": 0.01144588,
      "auxiliary_loss_mlp": 0.01024222,
      "balance_loss_clip": 1.04718387,
      "balance_loss_mlp": 1.01670551,
      "epoch": 0.6181687007755666,
      "flos": 35225458905600.0,
      "grad_norm": 2.1289467633489254,
      "language_loss": 0.69880098,
      "learning_rate": 1.3440129119526322e-06,
      "loss": 0.72048903,
      "num_input_tokens_seen": 110806395,
      "step": 5141,
      "time_per_iteration": 3.3777740001678467
    },
    {
      "auxiliary_loss_clip": 0.0107402,
      "auxiliary_loss_mlp": 0.01004253,
      "balance_loss_clip": 1.01679921,
      "balance_loss_mlp": 1.003335,
      "epoch": 0.6182889436662057,
      "flos": 61547370094080.0,
      "grad_norm": 0.8001682396038746,
      "language_loss": 0.51233035,
      "learning_rate": 1.3432770828195762e-06,
      "loss": 0.53311312,
      "num_input_tokens_seen": 110867380,
      "step": 5142,
      "time_per_iteration": 3.17407488822937
    },
    {
      "auxiliary_loss_clip": 0.01120848,
      "auxiliary_loss_mlp": 0.01022626,
      "balance_loss_clip": 1.04271865,
      "balance_loss_mlp": 1.01443088,
      "epoch": 0.6184091865568448,
      "flos": 19609991804160.0,
      "grad_norm": 2.3179526312180276,
      "language_loss": 0.70428991,
      "learning_rate": 1.3425413533108635e-06,
      "loss": 0.72572464,
      "num_input_tokens_seen": 110885980,
      "step": 5143,
      "time_per_iteration": 3.317586898803711
    },
    {
      "auxiliary_loss_clip": 0.01120033,
      "auxiliary_loss_mlp": 0.01026638,
      "balance_loss_clip": 1.04843521,
      "balance_loss_mlp": 1.01929438,
      "epoch": 0.6185294294474839,
      "flos": 23586882929280.0,
      "grad_norm": 2.0891966194110334,
      "language_loss": 0.70644093,
      "learning_rate": 1.341805723538105e-06,
      "loss": 0.72790766,
      "num_input_tokens_seen": 110906085,
      "step": 5144,
      "time_per_iteration": 2.6144511699676514
    },
    {
      "auxiliary_loss_clip": 0.01155755,
      "auxiliary_loss_mlp": 0.01031882,
      "balance_loss_clip": 1.05108666,
      "balance_loss_mlp": 1.02363276,
      "epoch": 0.618649672338123,
      "flos": 26762032535040.0,
      "grad_norm": 1.6604391015252626,
      "language_loss": 0.77731955,
      "learning_rate": 1.3410701936128948e-06,
      "loss": 0.79919589,
      "num_input_tokens_seen": 110928865,
      "step": 5145,
      "time_per_iteration": 2.5302317142486572
    },
    {
      "auxiliary_loss_clip": 0.01163699,
      "auxiliary_loss_mlp": 0.01027024,
      "balance_loss_clip": 1.05389094,
      "balance_loss_mlp": 1.01953161,
      "epoch": 0.6187699152287621,
      "flos": 14456630522880.0,
      "grad_norm": 2.5091876659540846,
      "language_loss": 0.84897012,
      "learning_rate": 1.340334763646812e-06,
      "loss": 0.87087739,
      "num_input_tokens_seen": 110943000,
      "step": 5146,
      "time_per_iteration": 2.3902246952056885
    },
    {
      "auxiliary_loss_clip": 0.01177569,
      "auxiliary_loss_mlp": 0.01030709,
      "balance_loss_clip": 1.05235517,
      "balance_loss_mlp": 1.02191734,
      "epoch": 0.6188901581194012,
      "flos": 20084766796800.0,
      "grad_norm": 1.8313339041881738,
      "language_loss": 0.74298638,
      "learning_rate": 1.3395994337514218e-06,
      "loss": 0.76506919,
      "num_input_tokens_seen": 110963170,
      "step": 5147,
      "time_per_iteration": 2.4604387283325195
    },
    {
      "auxiliary_loss_clip": 0.01153079,
      "auxiliary_loss_mlp": 0.01028197,
      "balance_loss_clip": 1.04697967,
      "balance_loss_mlp": 1.02035332,
      "epoch": 0.6190104010100402,
      "flos": 25700728360320.0,
      "grad_norm": 1.6864189167042607,
      "language_loss": 0.78778219,
      "learning_rate": 1.3388642040382725e-06,
      "loss": 0.80959487,
      "num_input_tokens_seen": 110983595,
      "step": 5148,
      "time_per_iteration": 2.4791977405548096
    },
    {
      "auxiliary_loss_clip": 0.01133196,
      "auxiliary_loss_mlp": 0.01025399,
      "balance_loss_clip": 1.04170585,
      "balance_loss_mlp": 1.01743031,
      "epoch": 0.6191306439006794,
      "flos": 30442372974720.0,
      "grad_norm": 1.7344759551081834,
      "language_loss": 0.84180474,
      "learning_rate": 1.3381290746188975e-06,
      "loss": 0.86339074,
      "num_input_tokens_seen": 111002965,
      "step": 5149,
      "time_per_iteration": 2.581310272216797
    },
    {
      "auxiliary_loss_clip": 0.01162987,
      "auxiliary_loss_mlp": 0.01032321,
      "balance_loss_clip": 1.05254984,
      "balance_loss_mlp": 1.02409577,
      "epoch": 0.6192508867913185,
      "flos": 26685793918080.0,
      "grad_norm": 1.6903521483416424,
      "language_loss": 0.67075741,
      "learning_rate": 1.3373940456048152e-06,
      "loss": 0.69271052,
      "num_input_tokens_seen": 111022990,
      "step": 5150,
      "time_per_iteration": 2.4894511699676514
    },
    {
      "auxiliary_loss_clip": 0.01174787,
      "auxiliary_loss_mlp": 0.01024873,
      "balance_loss_clip": 1.05245173,
      "balance_loss_mlp": 1.01749969,
      "epoch": 0.6193711296819575,
      "flos": 36722036090880.0,
      "grad_norm": 1.8015590074361678,
      "language_loss": 0.5909031,
      "learning_rate": 1.3366591171075299e-06,
      "loss": 0.61289972,
      "num_input_tokens_seen": 111046495,
      "step": 5151,
      "time_per_iteration": 2.550067186355591
    },
    {
      "auxiliary_loss_clip": 0.01145976,
      "auxiliary_loss_mlp": 0.0102376,
      "balance_loss_clip": 1.04841852,
      "balance_loss_mlp": 1.01637459,
      "epoch": 0.6194913725725967,
      "flos": 25192556697600.0,
      "grad_norm": 1.9518248568930159,
      "language_loss": 0.91008627,
      "learning_rate": 1.335924289238529e-06,
      "loss": 0.93178362,
      "num_input_tokens_seen": 111065705,
      "step": 5152,
      "time_per_iteration": 2.516597270965576
    },
    {
      "auxiliary_loss_clip": 0.0114438,
      "auxiliary_loss_mlp": 0.00763011,
      "balance_loss_clip": 1.05101037,
      "balance_loss_mlp": 1.00079656,
      "epoch": 0.6196116154632357,
      "flos": 21178821196800.0,
      "grad_norm": 1.5269647474180734,
      "language_loss": 0.76932395,
      "learning_rate": 1.3351895621092859e-06,
      "loss": 0.78839779,
      "num_input_tokens_seen": 111086050,
      "step": 5153,
      "time_per_iteration": 2.5211715698242188
    },
    {
      "auxiliary_loss_clip": 0.01081732,
      "auxiliary_loss_mlp": 0.01029174,
      "balance_loss_clip": 1.03751802,
      "balance_loss_mlp": 1.02149105,
      "epoch": 0.6197318583538748,
      "flos": 16253744803200.0,
      "grad_norm": 1.8713889016813556,
      "language_loss": 0.76723945,
      "learning_rate": 1.3344549358312567e-06,
      "loss": 0.7883485,
      "num_input_tokens_seen": 111104450,
      "step": 5154,
      "time_per_iteration": 2.616731643676758
    },
    {
      "auxiliary_loss_clip": 0.01165772,
      "auxiliary_loss_mlp": 0.01023722,
      "balance_loss_clip": 1.05164266,
      "balance_loss_mlp": 1.01562738,
      "epoch": 0.619852101244514,
      "flos": 24425612478720.0,
      "grad_norm": 1.8889001754231425,
      "language_loss": 0.78362107,
      "learning_rate": 1.3337204105158852e-06,
      "loss": 0.805516,
      "num_input_tokens_seen": 111123320,
      "step": 5155,
      "time_per_iteration": 2.4758620262145996
    },
    {
      "auxiliary_loss_clip": 0.01115153,
      "auxiliary_loss_mlp": 0.01027215,
      "balance_loss_clip": 1.03721404,
      "balance_loss_mlp": 1.01937151,
      "epoch": 0.619972344135153,
      "flos": 16727298733440.0,
      "grad_norm": 1.8317029600533525,
      "language_loss": 0.72757632,
      "learning_rate": 1.332985986274597e-06,
      "loss": 0.74899995,
      "num_input_tokens_seen": 111140950,
      "step": 5156,
      "time_per_iteration": 2.491074323654175
    },
    {
      "auxiliary_loss_clip": 0.01095943,
      "auxiliary_loss_mlp": 0.00762421,
      "balance_loss_clip": 1.04548001,
      "balance_loss_mlp": 1.00068665,
      "epoch": 0.6200925870257921,
      "flos": 12495190498560.0,
      "grad_norm": 1.9978033146800827,
      "language_loss": 0.75580359,
      "learning_rate": 1.3322516632188047e-06,
      "loss": 0.77438724,
      "num_input_tokens_seen": 111157845,
      "step": 5157,
      "time_per_iteration": 2.5521092414855957
    },
    {
      "auxiliary_loss_clip": 0.01129246,
      "auxiliary_loss_mlp": 0.01027925,
      "balance_loss_clip": 1.0455339,
      "balance_loss_mlp": 1.01991391,
      "epoch": 0.6202128299164312,
      "flos": 26539350168960.0,
      "grad_norm": 1.7963965760158007,
      "language_loss": 0.6719082,
      "learning_rate": 1.3315174414599045e-06,
      "loss": 0.69347996,
      "num_input_tokens_seen": 111179165,
      "step": 5158,
      "time_per_iteration": 2.5751895904541016
    },
    {
      "auxiliary_loss_clip": 0.01156214,
      "auxiliary_loss_mlp": 0.01024233,
      "balance_loss_clip": 1.04734278,
      "balance_loss_mlp": 1.01568592,
      "epoch": 0.6203330728070703,
      "flos": 18770508069120.0,
      "grad_norm": 2.1245506726345535,
      "language_loss": 0.75202847,
      "learning_rate": 1.3307833211092768e-06,
      "loss": 0.77383298,
      "num_input_tokens_seen": 111197830,
      "step": 5159,
      "time_per_iteration": 2.4460315704345703
    },
    {
      "auxiliary_loss_clip": 0.01177647,
      "auxiliary_loss_mlp": 0.01031116,
      "balance_loss_clip": 1.05416012,
      "balance_loss_mlp": 1.02321851,
      "epoch": 0.6204533156977093,
      "flos": 20629782835200.0,
      "grad_norm": 1.6359629451403146,
      "language_loss": 0.75242364,
      "learning_rate": 1.3300493022782873e-06,
      "loss": 0.77451134,
      "num_input_tokens_seen": 111218400,
      "step": 5160,
      "time_per_iteration": 2.437129020690918
    },
    {
      "auxiliary_loss_clip": 0.01109571,
      "auxiliary_loss_mlp": 0.00763293,
      "balance_loss_clip": 1.04406416,
      "balance_loss_mlp": 1.00076747,
      "epoch": 0.6205735585883485,
      "flos": 17348050598400.0,
      "grad_norm": 1.789987078371122,
      "language_loss": 0.72415805,
      "learning_rate": 1.3293153850782855e-06,
      "loss": 0.74288672,
      "num_input_tokens_seen": 111236720,
      "step": 5161,
      "time_per_iteration": 2.529670000076294
    },
    {
      "auxiliary_loss_clip": 0.01123569,
      "auxiliary_loss_mlp": 0.01028339,
      "balance_loss_clip": 1.04506731,
      "balance_loss_mlp": 1.0192728,
      "epoch": 0.6206938014789876,
      "flos": 22965017742720.0,
      "grad_norm": 2.394727344070964,
      "language_loss": 0.71178114,
      "learning_rate": 1.3285815696206069e-06,
      "loss": 0.73330021,
      "num_input_tokens_seen": 111258265,
      "step": 5162,
      "time_per_iteration": 3.2981951236724854
    },
    {
      "auxiliary_loss_clip": 0.01133436,
      "auxiliary_loss_mlp": 0.01032915,
      "balance_loss_clip": 1.04450345,
      "balance_loss_mlp": 1.02477264,
      "epoch": 0.6208140443696266,
      "flos": 23983192661760.0,
      "grad_norm": 2.69615125241396,
      "language_loss": 0.7683441,
      "learning_rate": 1.32784785601657e-06,
      "loss": 0.79000771,
      "num_input_tokens_seen": 111277675,
      "step": 5163,
      "time_per_iteration": 2.548370599746704
    },
    {
      "auxiliary_loss_clip": 0.01148839,
      "auxiliary_loss_mlp": 0.01024809,
      "balance_loss_clip": 1.0469408,
      "balance_loss_mlp": 1.01693249,
      "epoch": 0.6209342872602658,
      "flos": 35077291303680.0,
      "grad_norm": 1.7211830320276993,
      "language_loss": 0.73686028,
      "learning_rate": 1.3271142443774798e-06,
      "loss": 0.75859672,
      "num_input_tokens_seen": 111299910,
      "step": 5164,
      "time_per_iteration": 2.629284143447876
    },
    {
      "auxiliary_loss_clip": 0.01144833,
      "auxiliary_loss_mlp": 0.01022473,
      "balance_loss_clip": 1.04928827,
      "balance_loss_mlp": 1.01459968,
      "epoch": 0.6210545301509048,
      "flos": 26979327861120.0,
      "grad_norm": 1.766327599393501,
      "language_loss": 0.81522989,
      "learning_rate": 1.3263807348146228e-06,
      "loss": 0.83690292,
      "num_input_tokens_seen": 111319765,
      "step": 5165,
      "time_per_iteration": 2.5535645484924316
    },
    {
      "auxiliary_loss_clip": 0.0114433,
      "auxiliary_loss_mlp": 0.01036337,
      "balance_loss_clip": 1.04550493,
      "balance_loss_mlp": 1.0276525,
      "epoch": 0.6211747730415439,
      "flos": 33618240852480.0,
      "grad_norm": 2.192007811063716,
      "language_loss": 0.73259747,
      "learning_rate": 1.3256473274392733e-06,
      "loss": 0.75440407,
      "num_input_tokens_seen": 111341110,
      "step": 5166,
      "time_per_iteration": 3.4649343490600586
    },
    {
      "auxiliary_loss_clip": 0.01174752,
      "auxiliary_loss_mlp": 0.01029226,
      "balance_loss_clip": 1.05107534,
      "balance_loss_mlp": 1.02116108,
      "epoch": 0.6212950159321831,
      "flos": 34167099646080.0,
      "grad_norm": 1.9716200427603439,
      "language_loss": 0.69928497,
      "learning_rate": 1.3249140223626873e-06,
      "loss": 0.72132474,
      "num_input_tokens_seen": 111362730,
      "step": 5167,
      "time_per_iteration": 3.3117260932922363
    },
    {
      "auxiliary_loss_clip": 0.01159025,
      "auxiliary_loss_mlp": 0.01025647,
      "balance_loss_clip": 1.05048549,
      "balance_loss_mlp": 1.01794565,
      "epoch": 0.6214152588228221,
      "flos": 27965758135680.0,
      "grad_norm": 1.8686629737140226,
      "language_loss": 0.75584412,
      "learning_rate": 1.3241808196961077e-06,
      "loss": 0.77769083,
      "num_input_tokens_seen": 111383855,
      "step": 5168,
      "time_per_iteration": 2.499868392944336
    },
    {
      "auxiliary_loss_clip": 0.01133113,
      "auxiliary_loss_mlp": 0.01024034,
      "balance_loss_clip": 1.04511786,
      "balance_loss_mlp": 1.01642799,
      "epoch": 0.6215355017134612,
      "flos": 20230204965120.0,
      "grad_norm": 1.7888571058056566,
      "language_loss": 0.71031201,
      "learning_rate": 1.3234477195507608e-06,
      "loss": 0.73188341,
      "num_input_tokens_seen": 111402685,
      "step": 5169,
      "time_per_iteration": 2.4794108867645264
    },
    {
      "auxiliary_loss_clip": 0.01132301,
      "auxiliary_loss_mlp": 0.01027922,
      "balance_loss_clip": 1.04677236,
      "balance_loss_mlp": 1.02076328,
      "epoch": 0.6216557446041003,
      "flos": 41428129219200.0,
      "grad_norm": 2.3187113633964214,
      "language_loss": 0.62189484,
      "learning_rate": 1.322714722037857e-06,
      "loss": 0.64349705,
      "num_input_tokens_seen": 111424130,
      "step": 5170,
      "time_per_iteration": 3.42864727973938
    },
    {
      "auxiliary_loss_clip": 0.01140212,
      "auxiliary_loss_mlp": 0.01030152,
      "balance_loss_clip": 1.04759312,
      "balance_loss_mlp": 1.02231979,
      "epoch": 0.6217759874947394,
      "flos": 27928770105600.0,
      "grad_norm": 2.0451137869749543,
      "language_loss": 0.77578139,
      "learning_rate": 1.321981827268591e-06,
      "loss": 0.79748499,
      "num_input_tokens_seen": 111444785,
      "step": 5171,
      "time_per_iteration": 2.5806195735931396
    },
    {
      "auxiliary_loss_clip": 0.01150253,
      "auxiliary_loss_mlp": 0.01028771,
      "balance_loss_clip": 1.04932022,
      "balance_loss_mlp": 1.02086735,
      "epoch": 0.6218962303853784,
      "flos": 21765673601280.0,
      "grad_norm": 1.697017083153132,
      "language_loss": 0.81457758,
      "learning_rate": 1.3212490353541426e-06,
      "loss": 0.83636785,
      "num_input_tokens_seen": 111467045,
      "step": 5172,
      "time_per_iteration": 2.5358362197875977
    },
    {
      "auxiliary_loss_clip": 0.01176365,
      "auxiliary_loss_mlp": 0.01024585,
      "balance_loss_clip": 1.05082786,
      "balance_loss_mlp": 1.01658583,
      "epoch": 0.6220164732760175,
      "flos": 21246260981760.0,
      "grad_norm": 1.9215630578600953,
      "language_loss": 0.80180526,
      "learning_rate": 1.3205163464056762e-06,
      "loss": 0.82381475,
      "num_input_tokens_seen": 111483650,
      "step": 5173,
      "time_per_iteration": 2.428027868270874
    },
    {
      "auxiliary_loss_clip": 0.01158613,
      "auxiliary_loss_mlp": 0.0102875,
      "balance_loss_clip": 1.04830134,
      "balance_loss_mlp": 1.0210371,
      "epoch": 0.6221367161666567,
      "flos": 26136360506880.0,
      "grad_norm": 1.883102375572845,
      "language_loss": 0.72920758,
      "learning_rate": 1.319783760534339e-06,
      "loss": 0.75108123,
      "num_input_tokens_seen": 111502895,
      "step": 5174,
      "time_per_iteration": 2.4963786602020264
    },
    {
      "auxiliary_loss_clip": 0.01161858,
      "auxiliary_loss_mlp": 0.01030575,
      "balance_loss_clip": 1.052176,
      "balance_loss_mlp": 1.02232587,
      "epoch": 0.6222569590572957,
      "flos": 16284196558080.0,
      "grad_norm": 2.248326752209355,
      "language_loss": 0.75213224,
      "learning_rate": 1.319051277851266e-06,
      "loss": 0.77405655,
      "num_input_tokens_seen": 111519180,
      "step": 5175,
      "time_per_iteration": 2.431265115737915
    },
    {
      "auxiliary_loss_clip": 0.01162926,
      "auxiliary_loss_mlp": 0.01030796,
      "balance_loss_clip": 1.04964113,
      "balance_loss_mlp": 1.02309537,
      "epoch": 0.6223772019479348,
      "flos": 18223840005120.0,
      "grad_norm": 1.8167662182747357,
      "language_loss": 0.842345,
      "learning_rate": 1.3183188984675716e-06,
      "loss": 0.86428225,
      "num_input_tokens_seen": 111537545,
      "step": 5176,
      "time_per_iteration": 2.429539203643799
    },
    {
      "auxiliary_loss_clip": 0.0114653,
      "auxiliary_loss_mlp": 0.01033007,
      "balance_loss_clip": 1.05012107,
      "balance_loss_mlp": 1.02555668,
      "epoch": 0.6224974448385739,
      "flos": 27489797994240.0,
      "grad_norm": 2.4275330939966455,
      "language_loss": 0.71225476,
      "learning_rate": 1.3175866224943586e-06,
      "loss": 0.73405015,
      "num_input_tokens_seen": 111556265,
      "step": 5177,
      "time_per_iteration": 2.5202741622924805
    },
    {
      "auxiliary_loss_clip": 0.01150551,
      "auxiliary_loss_mlp": 0.01029362,
      "balance_loss_clip": 1.04937506,
      "balance_loss_mlp": 1.02128005,
      "epoch": 0.622617687729213,
      "flos": 19791951125760.0,
      "grad_norm": 2.5902245639624097,
      "language_loss": 0.73468375,
      "learning_rate": 1.316854450042712e-06,
      "loss": 0.75648284,
      "num_input_tokens_seen": 111574205,
      "step": 5178,
      "time_per_iteration": 2.468618392944336
    },
    {
      "auxiliary_loss_clip": 0.01166919,
      "auxiliary_loss_mlp": 0.01025371,
      "balance_loss_clip": 1.05129933,
      "balance_loss_mlp": 1.01751196,
      "epoch": 0.622737930619852,
      "flos": 23038886062080.0,
      "grad_norm": 1.9047970956836242,
      "language_loss": 0.74362719,
      "learning_rate": 1.3161223812237024e-06,
      "loss": 0.76555008,
      "num_input_tokens_seen": 111593560,
      "step": 5179,
      "time_per_iteration": 2.4668078422546387
    },
    {
      "auxiliary_loss_clip": 0.01174002,
      "auxiliary_loss_mlp": 0.01031181,
      "balance_loss_clip": 1.04919124,
      "balance_loss_mlp": 1.02327764,
      "epoch": 0.6228581735104912,
      "flos": 12634271959680.0,
      "grad_norm": 2.937125254931495,
      "language_loss": 0.84914672,
      "learning_rate": 1.3153904161483842e-06,
      "loss": 0.87119859,
      "num_input_tokens_seen": 111608860,
      "step": 5180,
      "time_per_iteration": 2.3765153884887695
    },
    {
      "auxiliary_loss_clip": 0.01127725,
      "auxiliary_loss_mlp": 0.01026438,
      "balance_loss_clip": 1.04467797,
      "balance_loss_mlp": 1.01800966,
      "epoch": 0.6229784164011303,
      "flos": 23802813538560.0,
      "grad_norm": 2.0340228026532836,
      "language_loss": 0.85037589,
      "learning_rate": 1.3146585549277953e-06,
      "loss": 0.87191749,
      "num_input_tokens_seen": 111627500,
      "step": 5181,
      "time_per_iteration": 2.5239310264587402
    },
    {
      "auxiliary_loss_clip": 0.01157476,
      "auxiliary_loss_mlp": 0.01032754,
      "balance_loss_clip": 1.05071926,
      "balance_loss_mlp": 1.02495217,
      "epoch": 0.6230986592917693,
      "flos": 22414219614720.0,
      "grad_norm": 2.0717912476862907,
      "language_loss": 0.78470969,
      "learning_rate": 1.3139267976729591e-06,
      "loss": 0.80661196,
      "num_input_tokens_seen": 111647690,
      "step": 5182,
      "time_per_iteration": 2.4967501163482666
    },
    {
      "auxiliary_loss_clip": 0.01165023,
      "auxiliary_loss_mlp": 0.01029867,
      "balance_loss_clip": 1.05192065,
      "balance_loss_mlp": 1.02140307,
      "epoch": 0.6232189021824085,
      "flos": 34528217028480.0,
      "grad_norm": 1.642552819933219,
      "language_loss": 0.71445298,
      "learning_rate": 1.3131951444948815e-06,
      "loss": 0.73640186,
      "num_input_tokens_seen": 111667090,
      "step": 5183,
      "time_per_iteration": 2.5647177696228027
    },
    {
      "auxiliary_loss_clip": 0.01149273,
      "auxiliary_loss_mlp": 0.01032329,
      "balance_loss_clip": 1.04965508,
      "balance_loss_mlp": 1.02431846,
      "epoch": 0.6233391450730476,
      "flos": 22237000888320.0,
      "grad_norm": 2.1184798783130776,
      "language_loss": 0.76420403,
      "learning_rate": 1.3124635955045546e-06,
      "loss": 0.78602004,
      "num_input_tokens_seen": 111686905,
      "step": 5184,
      "time_per_iteration": 2.4844810962677
    },
    {
      "auxiliary_loss_clip": 0.01107052,
      "auxiliary_loss_mlp": 0.00763131,
      "balance_loss_clip": 1.04104447,
      "balance_loss_mlp": 1.00061691,
      "epoch": 0.6234593879636866,
      "flos": 20332693445760.0,
      "grad_norm": 1.832222907773644,
      "language_loss": 0.83983433,
      "learning_rate": 1.3117321508129537e-06,
      "loss": 0.85853612,
      "num_input_tokens_seen": 111704985,
      "step": 5185,
      "time_per_iteration": 2.5574467182159424
    },
    {
      "auxiliary_loss_clip": 0.0114947,
      "auxiliary_loss_mlp": 0.01022647,
      "balance_loss_clip": 1.04886055,
      "balance_loss_mlp": 1.01532722,
      "epoch": 0.6235796308543258,
      "flos": 20664903358080.0,
      "grad_norm": 1.5358608182006221,
      "language_loss": 0.76228631,
      "learning_rate": 1.3110008105310388e-06,
      "loss": 0.78400743,
      "num_input_tokens_seen": 111724805,
      "step": 5186,
      "time_per_iteration": 2.502291440963745
    },
    {
      "auxiliary_loss_clip": 0.01176002,
      "auxiliary_loss_mlp": 0.01033589,
      "balance_loss_clip": 1.0499531,
      "balance_loss_mlp": 1.02547073,
      "epoch": 0.6236998737449648,
      "flos": 26618641441920.0,
      "grad_norm": 1.620482699143074,
      "language_loss": 0.78040814,
      "learning_rate": 1.3102695747697526e-06,
      "loss": 0.80250406,
      "num_input_tokens_seen": 111747675,
      "step": 5187,
      "time_per_iteration": 2.491316318511963
    },
    {
      "auxiliary_loss_clip": 0.01104596,
      "auxiliary_loss_mlp": 0.01027368,
      "balance_loss_clip": 1.04617989,
      "balance_loss_mlp": 1.01916695,
      "epoch": 0.6238201166356039,
      "flos": 12674599954560.0,
      "grad_norm": 3.164098414209472,
      "language_loss": 0.90217841,
      "learning_rate": 1.3095384436400237e-06,
      "loss": 0.92349803,
      "num_input_tokens_seen": 111759205,
      "step": 5188,
      "time_per_iteration": 2.5018911361694336
    },
    {
      "auxiliary_loss_clip": 0.01153973,
      "auxiliary_loss_mlp": 0.01023299,
      "balance_loss_clip": 1.04810059,
      "balance_loss_mlp": 1.01557386,
      "epoch": 0.623940359526243,
      "flos": 10452160730880.0,
      "grad_norm": 2.1649389303604223,
      "language_loss": 0.81824613,
      "learning_rate": 1.3088074172527633e-06,
      "loss": 0.84001887,
      "num_input_tokens_seen": 111776335,
      "step": 5189,
      "time_per_iteration": 3.333324432373047
    },
    {
      "auxiliary_loss_clip": 0.01148948,
      "auxiliary_loss_mlp": 0.01023697,
      "balance_loss_clip": 1.04586041,
      "balance_loss_mlp": 1.0154295,
      "epoch": 0.6240606024168821,
      "flos": 29059525226880.0,
      "grad_norm": 1.7440212301044016,
      "language_loss": 0.71493834,
      "learning_rate": 1.3080764957188684e-06,
      "loss": 0.73666477,
      "num_input_tokens_seen": 111796580,
      "step": 5190,
      "time_per_iteration": 2.5504536628723145
    },
    {
      "auxiliary_loss_clip": 0.01122099,
      "auxiliary_loss_mlp": 0.0102464,
      "balance_loss_clip": 1.04511976,
      "balance_loss_mlp": 1.01633668,
      "epoch": 0.6241808453075212,
      "flos": 22018089450240.0,
      "grad_norm": 1.7427760892535424,
      "language_loss": 0.70954829,
      "learning_rate": 1.3073456791492192e-06,
      "loss": 0.73101568,
      "num_input_tokens_seen": 111816290,
      "step": 5191,
      "time_per_iteration": 2.557647943496704
    },
    {
      "auxiliary_loss_clip": 0.01148211,
      "auxiliary_loss_mlp": 0.01025333,
      "balance_loss_clip": 1.04606092,
      "balance_loss_mlp": 1.01770949,
      "epoch": 0.6243010881981603,
      "flos": 21138708683520.0,
      "grad_norm": 1.7562718852180146,
      "language_loss": 0.78606039,
      "learning_rate": 1.3066149676546801e-06,
      "loss": 0.80779582,
      "num_input_tokens_seen": 111834470,
      "step": 5192,
      "time_per_iteration": 2.4930386543273926
    },
    {
      "auxiliary_loss_clip": 0.01148658,
      "auxiliary_loss_mlp": 0.01028748,
      "balance_loss_clip": 1.05349004,
      "balance_loss_mlp": 1.02068949,
      "epoch": 0.6244213310887994,
      "flos": 22344948236160.0,
      "grad_norm": 1.6920099490156852,
      "language_loss": 0.6629023,
      "learning_rate": 1.3058843613460985e-06,
      "loss": 0.68467641,
      "num_input_tokens_seen": 111852410,
      "step": 5193,
      "time_per_iteration": 3.344407081604004
    },
    {
      "auxiliary_loss_clip": 0.01140214,
      "auxiliary_loss_mlp": 0.01027518,
      "balance_loss_clip": 1.04753375,
      "balance_loss_mlp": 1.01972127,
      "epoch": 0.6245415739794384,
      "flos": 15231978524160.0,
      "grad_norm": 1.9263940141314557,
      "language_loss": 0.74089432,
      "learning_rate": 1.3051538603343075e-06,
      "loss": 0.76257157,
      "num_input_tokens_seen": 111870340,
      "step": 5194,
      "time_per_iteration": 3.365662097930908
    },
    {
      "auxiliary_loss_clip": 0.01161555,
      "auxiliary_loss_mlp": 0.01036097,
      "balance_loss_clip": 1.0514853,
      "balance_loss_mlp": 1.02819633,
      "epoch": 0.6246618168700776,
      "flos": 18879891960960.0,
      "grad_norm": 1.9666510575706388,
      "language_loss": 0.67418522,
      "learning_rate": 1.3044234647301235e-06,
      "loss": 0.69616175,
      "num_input_tokens_seen": 111888365,
      "step": 5195,
      "time_per_iteration": 2.442220449447632
    },
    {
      "auxiliary_loss_clip": 0.01155336,
      "auxiliary_loss_mlp": 0.01026757,
      "balance_loss_clip": 1.04876482,
      "balance_loss_mlp": 1.01986647,
      "epoch": 0.6247820597607167,
      "flos": 14319201087360.0,
      "grad_norm": 1.9101932011054747,
      "language_loss": 0.72601497,
      "learning_rate": 1.303693174644347e-06,
      "loss": 0.74783587,
      "num_input_tokens_seen": 111905840,
      "step": 5196,
      "time_per_iteration": 3.16748046875
    },
    {
      "auxiliary_loss_clip": 0.0114268,
      "auxiliary_loss_mlp": 0.0103079,
      "balance_loss_clip": 1.04643488,
      "balance_loss_mlp": 1.02220666,
      "epoch": 0.6249023026513557,
      "flos": 22637979388800.0,
      "grad_norm": 2.0862632571542226,
      "language_loss": 0.80467904,
      "learning_rate": 1.3029629901877625e-06,
      "loss": 0.82641375,
      "num_input_tokens_seen": 111925215,
      "step": 5197,
      "time_per_iteration": 2.4859719276428223
    },
    {
      "auxiliary_loss_clip": 0.01168862,
      "auxiliary_loss_mlp": 0.01027572,
      "balance_loss_clip": 1.05253983,
      "balance_loss_mlp": 1.01925683,
      "epoch": 0.6250225455419949,
      "flos": 20266690204800.0,
      "grad_norm": 3.0725458616635564,
      "language_loss": 0.77077407,
      "learning_rate": 1.3022329114711376e-06,
      "loss": 0.79273844,
      "num_input_tokens_seen": 111943925,
      "step": 5198,
      "time_per_iteration": 2.4443161487579346
    },
    {
      "auxiliary_loss_clip": 0.01143222,
      "auxiliary_loss_mlp": 0.0102402,
      "balance_loss_clip": 1.04749584,
      "balance_loss_mlp": 1.016361,
      "epoch": 0.6251427884326339,
      "flos": 23437853400960.0,
      "grad_norm": 2.4790473496002225,
      "language_loss": 0.69548547,
      "learning_rate": 1.3015029386052256e-06,
      "loss": 0.71715784,
      "num_input_tokens_seen": 111964095,
      "step": 5199,
      "time_per_iteration": 2.512484550476074
    },
    {
      "auxiliary_loss_clip": 0.01144577,
      "auxiliary_loss_mlp": 0.01030605,
      "balance_loss_clip": 1.04931223,
      "balance_loss_mlp": 1.02269578,
      "epoch": 0.625263031323273,
      "flos": 31723055464320.0,
      "grad_norm": 2.127825540730026,
      "language_loss": 0.72924012,
      "learning_rate": 1.3007730717007622e-06,
      "loss": 0.75099194,
      "num_input_tokens_seen": 111984910,
      "step": 5200,
      "time_per_iteration": 2.603738784790039
    },
    {
      "auxiliary_loss_clip": 0.01177455,
      "auxiliary_loss_mlp": 0.01029186,
      "balance_loss_clip": 1.05234516,
      "balance_loss_mlp": 1.02088344,
      "epoch": 0.6253832742139122,
      "flos": 24134341092480.0,
      "grad_norm": 1.7274660524267338,
      "language_loss": 0.75600898,
      "learning_rate": 1.3000433108684676e-06,
      "loss": 0.77807534,
      "num_input_tokens_seen": 112005410,
      "step": 5201,
      "time_per_iteration": 2.450544834136963
    },
    {
      "auxiliary_loss_clip": 0.01157304,
      "auxiliary_loss_mlp": 0.01022565,
      "balance_loss_clip": 1.04908442,
      "balance_loss_mlp": 1.01484585,
      "epoch": 0.6255035171045512,
      "flos": 27668812400640.0,
      "grad_norm": 9.456925047273225,
      "language_loss": 0.8065123,
      "learning_rate": 1.2993136562190467e-06,
      "loss": 0.82831097,
      "num_input_tokens_seen": 112024530,
      "step": 5202,
      "time_per_iteration": 2.4952661991119385
    },
    {
      "auxiliary_loss_clip": 0.01150399,
      "auxiliary_loss_mlp": 0.01028074,
      "balance_loss_clip": 1.04889143,
      "balance_loss_mlp": 1.01988471,
      "epoch": 0.6256237599951903,
      "flos": 20227798753920.0,
      "grad_norm": 1.5608837175036179,
      "language_loss": 0.70270973,
      "learning_rate": 1.2985841078631871e-06,
      "loss": 0.72449446,
      "num_input_tokens_seen": 112043850,
      "step": 5203,
      "time_per_iteration": 2.4833059310913086
    },
    {
      "auxiliary_loss_clip": 0.01099335,
      "auxiliary_loss_mlp": 0.01031528,
      "balance_loss_clip": 1.03912997,
      "balance_loss_mlp": 1.02352262,
      "epoch": 0.6257440028858293,
      "flos": 24170574936960.0,
      "grad_norm": 1.8296164103154822,
      "language_loss": 0.78451347,
      "learning_rate": 1.2978546659115608e-06,
      "loss": 0.80582213,
      "num_input_tokens_seen": 112061930,
      "step": 5204,
      "time_per_iteration": 2.5853633880615234
    },
    {
      "auxiliary_loss_clip": 0.01149213,
      "auxiliary_loss_mlp": 0.01030503,
      "balance_loss_clip": 1.0490036,
      "balance_loss_mlp": 1.02269483,
      "epoch": 0.6258642457764685,
      "flos": 15851940289920.0,
      "grad_norm": 1.9241035156068131,
      "language_loss": 0.85477406,
      "learning_rate": 1.2971253304748228e-06,
      "loss": 0.87657118,
      "num_input_tokens_seen": 112079645,
      "step": 5205,
      "time_per_iteration": 2.4536120891571045
    },
    {
      "auxiliary_loss_clip": 0.01166588,
      "auxiliary_loss_mlp": 0.01029086,
      "balance_loss_clip": 1.05259645,
      "balance_loss_mlp": 1.02108145,
      "epoch": 0.6259844886671075,
      "flos": 11911354836480.0,
      "grad_norm": 1.588159424865464,
      "language_loss": 0.75170666,
      "learning_rate": 1.296396101663614e-06,
      "loss": 0.7736634,
      "num_input_tokens_seen": 112096205,
      "step": 5206,
      "time_per_iteration": 2.429076671600342
    },
    {
      "auxiliary_loss_clip": 0.01164154,
      "auxiliary_loss_mlp": 0.01026794,
      "balance_loss_clip": 1.05111194,
      "balance_loss_mlp": 1.01894987,
      "epoch": 0.6261047315577466,
      "flos": 15887958652800.0,
      "grad_norm": 3.574489457448891,
      "language_loss": 0.8394323,
      "learning_rate": 1.2956669795885565e-06,
      "loss": 0.86134183,
      "num_input_tokens_seen": 112112835,
      "step": 5207,
      "time_per_iteration": 2.4251773357391357
    },
    {
      "auxiliary_loss_clip": 0.01127823,
      "auxiliary_loss_mlp": 0.01034217,
      "balance_loss_clip": 1.04628062,
      "balance_loss_mlp": 1.0260694,
      "epoch": 0.6262249744483858,
      "flos": 31248926916480.0,
      "grad_norm": 2.1633989928415707,
      "language_loss": 0.6804316,
      "learning_rate": 1.294937964360259e-06,
      "loss": 0.702052,
      "num_input_tokens_seen": 112133105,
      "step": 5208,
      "time_per_iteration": 2.5997345447540283
    },
    {
      "auxiliary_loss_clip": 0.01152985,
      "auxiliary_loss_mlp": 0.01030884,
      "balance_loss_clip": 1.04851007,
      "balance_loss_mlp": 1.02177668,
      "epoch": 0.6263452173390248,
      "flos": 27198598435200.0,
      "grad_norm": 2.202788255060346,
      "language_loss": 0.71013522,
      "learning_rate": 1.2942090560893108e-06,
      "loss": 0.73197389,
      "num_input_tokens_seen": 112152510,
      "step": 5209,
      "time_per_iteration": 2.5261213779449463
    },
    {
      "auxiliary_loss_clip": 0.01173905,
      "auxiliary_loss_mlp": 0.01024677,
      "balance_loss_clip": 1.05083227,
      "balance_loss_mlp": 1.01739907,
      "epoch": 0.6264654602296639,
      "flos": 37342069683840.0,
      "grad_norm": 2.0449967530183635,
      "language_loss": 0.60759819,
      "learning_rate": 1.2934802548862882e-06,
      "loss": 0.62958395,
      "num_input_tokens_seen": 112175295,
      "step": 5210,
      "time_per_iteration": 2.5536608695983887
    },
    {
      "auxiliary_loss_clip": 0.01142308,
      "auxiliary_loss_mlp": 0.01029336,
      "balance_loss_clip": 1.04583013,
      "balance_loss_mlp": 1.02179551,
      "epoch": 0.626585703120303,
      "flos": 14756952136320.0,
      "grad_norm": 1.7741220645081541,
      "language_loss": 0.83058244,
      "learning_rate": 1.292751560861749e-06,
      "loss": 0.85229892,
      "num_input_tokens_seen": 112190200,
      "step": 5211,
      "time_per_iteration": 2.4346988201141357
    },
    {
      "auxiliary_loss_clip": 0.01175801,
      "auxiliary_loss_mlp": 0.01028444,
      "balance_loss_clip": 1.0510031,
      "balance_loss_mlp": 1.02000999,
      "epoch": 0.6267059460109421,
      "flos": 22347318533760.0,
      "grad_norm": 1.8700318560239715,
      "language_loss": 0.79694223,
      "learning_rate": 1.2920229741262354e-06,
      "loss": 0.81898469,
      "num_input_tokens_seen": 112208205,
      "step": 5212,
      "time_per_iteration": 2.4382312297821045
    },
    {
      "auxiliary_loss_clip": 0.01147328,
      "auxiliary_loss_mlp": 0.01026899,
      "balance_loss_clip": 1.04774451,
      "balance_loss_mlp": 1.01924598,
      "epoch": 0.6268261889015811,
      "flos": 17748813617280.0,
      "grad_norm": 2.7344753534121433,
      "language_loss": 0.75384033,
      "learning_rate": 1.2912944947902739e-06,
      "loss": 0.77558255,
      "num_input_tokens_seen": 112224690,
      "step": 5213,
      "time_per_iteration": 2.4599359035491943
    },
    {
      "auxiliary_loss_clip": 0.01152422,
      "auxiliary_loss_mlp": 0.01024836,
      "balance_loss_clip": 1.0487715,
      "balance_loss_mlp": 1.01639569,
      "epoch": 0.6269464317922203,
      "flos": 32846484211200.0,
      "grad_norm": 1.9549116769146608,
      "language_loss": 0.71654594,
      "learning_rate": 1.2905661229643742e-06,
      "loss": 0.73831856,
      "num_input_tokens_seen": 112244450,
      "step": 5214,
      "time_per_iteration": 2.630216360092163
    },
    {
      "auxiliary_loss_clip": 0.01176062,
      "auxiliary_loss_mlp": 0.01029642,
      "balance_loss_clip": 1.05081916,
      "balance_loss_mlp": 1.02159548,
      "epoch": 0.6270666746828594,
      "flos": 17929192740480.0,
      "grad_norm": 2.2214229842556663,
      "language_loss": 0.84309161,
      "learning_rate": 1.2898378587590299e-06,
      "loss": 0.86514866,
      "num_input_tokens_seen": 112261050,
      "step": 5215,
      "time_per_iteration": 3.1176860332489014
    },
    {
      "auxiliary_loss_clip": 0.0115838,
      "auxiliary_loss_mlp": 0.0102373,
      "balance_loss_clip": 1.0502367,
      "balance_loss_mlp": 1.01613069,
      "epoch": 0.6271869175734984,
      "flos": 17457326749440.0,
      "grad_norm": 1.8758552555966743,
      "language_loss": 0.87420404,
      "learning_rate": 1.2891097022847173e-06,
      "loss": 0.89602506,
      "num_input_tokens_seen": 112278395,
      "step": 5216,
      "time_per_iteration": 2.455867290496826
    },
    {
      "auxiliary_loss_clip": 0.01146082,
      "auxiliary_loss_mlp": 0.01033469,
      "balance_loss_clip": 1.0476253,
      "balance_loss_mlp": 1.0246172,
      "epoch": 0.6273071604641376,
      "flos": 26868615166080.0,
      "grad_norm": 2.3621541623132916,
      "language_loss": 0.66724181,
      "learning_rate": 1.2883816536518978e-06,
      "loss": 0.68903726,
      "num_input_tokens_seen": 112299535,
      "step": 5217,
      "time_per_iteration": 2.5225913524627686
    },
    {
      "auxiliary_loss_clip": 0.0115751,
      "auxiliary_loss_mlp": 0.01026678,
      "balance_loss_clip": 1.04755771,
      "balance_loss_mlp": 1.01952553,
      "epoch": 0.6274274033547766,
      "flos": 26062384446720.0,
      "grad_norm": 1.71003600763185,
      "language_loss": 0.81872934,
      "learning_rate": 1.2876537129710155e-06,
      "loss": 0.84057128,
      "num_input_tokens_seen": 112317265,
      "step": 5218,
      "time_per_iteration": 2.475831985473633
    },
    {
      "auxiliary_loss_clip": 0.0114661,
      "auxiliary_loss_mlp": 0.01031142,
      "balance_loss_clip": 1.05319691,
      "balance_loss_mlp": 1.02334619,
      "epoch": 0.6275476462454157,
      "flos": 20266259241600.0,
      "grad_norm": 1.970132464259165,
      "language_loss": 0.7524184,
      "learning_rate": 1.286925880352499e-06,
      "loss": 0.77419591,
      "num_input_tokens_seen": 112336125,
      "step": 5219,
      "time_per_iteration": 2.473417043685913
    },
    {
      "auxiliary_loss_clip": 0.01145305,
      "auxiliary_loss_mlp": 0.01021292,
      "balance_loss_clip": 1.04897714,
      "balance_loss_mlp": 1.0139488,
      "epoch": 0.6276678891360549,
      "flos": 26320402817280.0,
      "grad_norm": 1.7128010599842847,
      "language_loss": 0.71219057,
      "learning_rate": 1.2861981559067592e-06,
      "loss": 0.7338565,
      "num_input_tokens_seen": 112356730,
      "step": 5220,
      "time_per_iteration": 3.3524868488311768
    },
    {
      "auxiliary_loss_clip": 0.0110738,
      "auxiliary_loss_mlp": 0.01024401,
      "balance_loss_clip": 1.04274035,
      "balance_loss_mlp": 1.0168134,
      "epoch": 0.6277881320266939,
      "flos": 13912512324480.0,
      "grad_norm": 2.0379167900975714,
      "language_loss": 0.80072564,
      "learning_rate": 1.2854705397441917e-06,
      "loss": 0.82204348,
      "num_input_tokens_seen": 112372270,
      "step": 5221,
      "time_per_iteration": 3.384284734725952
    },
    {
      "auxiliary_loss_clip": 0.01125709,
      "auxiliary_loss_mlp": 0.01026377,
      "balance_loss_clip": 1.04305279,
      "balance_loss_mlp": 1.01877129,
      "epoch": 0.627908374917333,
      "flos": 27048922462080.0,
      "grad_norm": 2.2332600876185844,
      "language_loss": 0.7759856,
      "learning_rate": 1.2847430319751747e-06,
      "loss": 0.79750645,
      "num_input_tokens_seen": 112390365,
      "step": 5222,
      "time_per_iteration": 2.5690886974334717
    },
    {
      "auxiliary_loss_clip": 0.01155425,
      "auxiliary_loss_mlp": 0.01024738,
      "balance_loss_clip": 1.05053329,
      "balance_loss_mlp": 1.0174545,
      "epoch": 0.6280286178079721,
      "flos": 23769201386880.0,
      "grad_norm": 2.288177421563343,
      "language_loss": 0.67311609,
      "learning_rate": 1.2840156327100712e-06,
      "loss": 0.69491774,
      "num_input_tokens_seen": 112407490,
      "step": 5223,
      "time_per_iteration": 3.172185182571411
    },
    {
      "auxiliary_loss_clip": 0.01172709,
      "auxiliary_loss_mlp": 0.01023958,
      "balance_loss_clip": 1.05046487,
      "balance_loss_mlp": 1.01622117,
      "epoch": 0.6281488606986112,
      "flos": 26359150613760.0,
      "grad_norm": 1.8129402269948693,
      "language_loss": 0.72260547,
      "learning_rate": 1.2832883420592272e-06,
      "loss": 0.7445721,
      "num_input_tokens_seen": 112426385,
      "step": 5224,
      "time_per_iteration": 2.4578473567962646
    },
    {
      "auxiliary_loss_clip": 0.01142726,
      "auxiliary_loss_mlp": 0.01028501,
      "balance_loss_clip": 1.0477531,
      "balance_loss_mlp": 1.02057946,
      "epoch": 0.6282691035892503,
      "flos": 36137194848000.0,
      "grad_norm": 3.5637560289815746,
      "language_loss": 0.64467061,
      "learning_rate": 1.282561160132972e-06,
      "loss": 0.66638285,
      "num_input_tokens_seen": 112446905,
      "step": 5225,
      "time_per_iteration": 2.6068241596221924
    },
    {
      "auxiliary_loss_clip": 0.01151054,
      "auxiliary_loss_mlp": 0.01037466,
      "balance_loss_clip": 1.04585648,
      "balance_loss_mlp": 1.02968717,
      "epoch": 0.6283893464798894,
      "flos": 26537231266560.0,
      "grad_norm": 1.9507870436213395,
      "language_loss": 0.80939525,
      "learning_rate": 1.2818340870416186e-06,
      "loss": 0.83128047,
      "num_input_tokens_seen": 112468040,
      "step": 5226,
      "time_per_iteration": 2.5375113487243652
    },
    {
      "auxiliary_loss_clip": 0.01138921,
      "auxiliary_loss_mlp": 0.01029503,
      "balance_loss_clip": 1.04562402,
      "balance_loss_mlp": 1.02097321,
      "epoch": 0.6285095893705285,
      "flos": 22237216369920.0,
      "grad_norm": 2.411895185398057,
      "language_loss": 0.75849283,
      "learning_rate": 1.2811071228954626e-06,
      "loss": 0.780177,
      "num_input_tokens_seen": 112486675,
      "step": 5227,
      "time_per_iteration": 2.5346691608428955
    },
    {
      "auxiliary_loss_clip": 0.01146403,
      "auxiliary_loss_mlp": 0.01026043,
      "balance_loss_clip": 1.04913485,
      "balance_loss_mlp": 1.01816332,
      "epoch": 0.6286298322611675,
      "flos": 26542259170560.0,
      "grad_norm": 2.116746544027814,
      "language_loss": 0.81267941,
      "learning_rate": 1.2803802678047846e-06,
      "loss": 0.83440387,
      "num_input_tokens_seen": 112506825,
      "step": 5228,
      "time_per_iteration": 2.5236423015594482
    },
    {
      "auxiliary_loss_clip": 0.0115161,
      "auxiliary_loss_mlp": 0.01035118,
      "balance_loss_clip": 1.04968143,
      "balance_loss_mlp": 1.02688074,
      "epoch": 0.6287500751518067,
      "flos": 21795227516160.0,
      "grad_norm": 1.9220122988730943,
      "language_loss": 0.74154377,
      "learning_rate": 1.279653521879848e-06,
      "loss": 0.76341105,
      "num_input_tokens_seen": 112526890,
      "step": 5229,
      "time_per_iteration": 2.489473819732666
    },
    {
      "auxiliary_loss_clip": 0.01080833,
      "auxiliary_loss_mlp": 0.01026129,
      "balance_loss_clip": 1.03982747,
      "balance_loss_mlp": 1.01892579,
      "epoch": 0.6288703180424458,
      "flos": 20009605587840.0,
      "grad_norm": 1.9858378219753012,
      "language_loss": 0.83831459,
      "learning_rate": 1.2789268852308997e-06,
      "loss": 0.85938418,
      "num_input_tokens_seen": 112542100,
      "step": 5230,
      "time_per_iteration": 2.578523874282837
    },
    {
      "auxiliary_loss_clip": 0.01152263,
      "auxiliary_loss_mlp": 0.0103009,
      "balance_loss_clip": 1.04742026,
      "balance_loss_mlp": 1.02191186,
      "epoch": 0.6289905609330848,
      "flos": 22124923476480.0,
      "grad_norm": 2.1981195811225214,
      "language_loss": 0.71028095,
      "learning_rate": 1.2782003579681688e-06,
      "loss": 0.73210454,
      "num_input_tokens_seen": 112561630,
      "step": 5231,
      "time_per_iteration": 2.470848798751831
    },
    {
      "auxiliary_loss_clip": 0.01177175,
      "auxiliary_loss_mlp": 0.01029849,
      "balance_loss_clip": 1.05290663,
      "balance_loss_mlp": 1.02168894,
      "epoch": 0.629110803823724,
      "flos": 25518481729920.0,
      "grad_norm": 1.5388962269697406,
      "language_loss": 0.74302125,
      "learning_rate": 1.2774739402018701e-06,
      "loss": 0.76509148,
      "num_input_tokens_seen": 112582465,
      "step": 5232,
      "time_per_iteration": 2.4598615169525146
    },
    {
      "auxiliary_loss_clip": 0.01162717,
      "auxiliary_loss_mlp": 0.01030345,
      "balance_loss_clip": 1.05331588,
      "balance_loss_mlp": 1.02197623,
      "epoch": 0.629231046714363,
      "flos": 20886616056960.0,
      "grad_norm": 2.1229375345960295,
      "language_loss": 0.73529476,
      "learning_rate": 1.2767476320422002e-06,
      "loss": 0.75722539,
      "num_input_tokens_seen": 112602390,
      "step": 5233,
      "time_per_iteration": 2.467738628387451
    },
    {
      "auxiliary_loss_clip": 0.01048812,
      "auxiliary_loss_mlp": 0.01001011,
      "balance_loss_clip": 1.01820767,
      "balance_loss_mlp": 1.00014102,
      "epoch": 0.6293512896050021,
      "flos": 65050027908480.0,
      "grad_norm": 0.6788227917817856,
      "language_loss": 0.57197392,
      "learning_rate": 1.2760214335993392e-06,
      "loss": 0.59247214,
      "num_input_tokens_seen": 112669035,
      "step": 5234,
      "time_per_iteration": 3.1789321899414062
    },
    {
      "auxiliary_loss_clip": 0.01152627,
      "auxiliary_loss_mlp": 0.01025423,
      "balance_loss_clip": 1.04704154,
      "balance_loss_mlp": 1.01815128,
      "epoch": 0.6294715324956413,
      "flos": 34677857088000.0,
      "grad_norm": 1.9001739370799604,
      "language_loss": 0.59017414,
      "learning_rate": 1.2752953449834514e-06,
      "loss": 0.61195463,
      "num_input_tokens_seen": 112691485,
      "step": 5235,
      "time_per_iteration": 2.579751968383789
    },
    {
      "auxiliary_loss_clip": 0.01175512,
      "auxiliary_loss_mlp": 0.01027767,
      "balance_loss_clip": 1.05241036,
      "balance_loss_mlp": 1.02036381,
      "epoch": 0.6295917753862803,
      "flos": 22784207656320.0,
      "grad_norm": 1.5564747156146077,
      "language_loss": 0.7999087,
      "learning_rate": 1.2745693663046836e-06,
      "loss": 0.82194149,
      "num_input_tokens_seen": 112710555,
      "step": 5236,
      "time_per_iteration": 2.432197093963623
    },
    {
      "auxiliary_loss_clip": 0.01154967,
      "auxiliary_loss_mlp": 0.01023839,
      "balance_loss_clip": 1.04776597,
      "balance_loss_mlp": 1.01674283,
      "epoch": 0.6297120182769194,
      "flos": 20850454039680.0,
      "grad_norm": 2.415094296415789,
      "language_loss": 0.80901217,
      "learning_rate": 1.2738434976731662e-06,
      "loss": 0.83080018,
      "num_input_tokens_seen": 112728740,
      "step": 5237,
      "time_per_iteration": 2.446061611175537
    },
    {
      "auxiliary_loss_clip": 0.01147329,
      "auxiliary_loss_mlp": 0.01036582,
      "balance_loss_clip": 1.04914629,
      "balance_loss_mlp": 1.028023,
      "epoch": 0.6298322611675584,
      "flos": 19497662997120.0,
      "grad_norm": 1.511658108858409,
      "language_loss": 0.75341427,
      "learning_rate": 1.2731177391990125e-06,
      "loss": 0.77525342,
      "num_input_tokens_seen": 112748665,
      "step": 5238,
      "time_per_iteration": 2.495856285095215
    },
    {
      "auxiliary_loss_clip": 0.01146229,
      "auxiliary_loss_mlp": 0.01024142,
      "balance_loss_clip": 1.04514074,
      "balance_loss_mlp": 1.01671517,
      "epoch": 0.6299525040581976,
      "flos": 12604466649600.0,
      "grad_norm": 1.9377019127844985,
      "language_loss": 0.81650442,
      "learning_rate": 1.2723920909923203e-06,
      "loss": 0.83820814,
      "num_input_tokens_seen": 112764410,
      "step": 5239,
      "time_per_iteration": 2.4648613929748535
    },
    {
      "auxiliary_loss_clip": 0.01074429,
      "auxiliary_loss_mlp": 0.01004566,
      "balance_loss_clip": 1.0169102,
      "balance_loss_mlp": 1.00367188,
      "epoch": 0.6300727469488366,
      "flos": 57725685636480.0,
      "grad_norm": 0.8612994398266367,
      "language_loss": 0.60413402,
      "learning_rate": 1.2716665531631688e-06,
      "loss": 0.62492394,
      "num_input_tokens_seen": 112818695,
      "step": 5240,
      "time_per_iteration": 2.967538595199585
    },
    {
      "auxiliary_loss_clip": 0.01164447,
      "auxiliary_loss_mlp": 0.01023709,
      "balance_loss_clip": 1.04732883,
      "balance_loss_mlp": 1.01520991,
      "epoch": 0.6301929898394757,
      "flos": 22527302607360.0,
      "grad_norm": 1.7266647771960117,
      "language_loss": 0.77568197,
      "learning_rate": 1.270941125821623e-06,
      "loss": 0.79756355,
      "num_input_tokens_seen": 112839120,
      "step": 5241,
      "time_per_iteration": 2.4679481983184814
    },
    {
      "auxiliary_loss_clip": 0.01152948,
      "auxiliary_loss_mlp": 0.01030813,
      "balance_loss_clip": 1.0445534,
      "balance_loss_mlp": 1.02295089,
      "epoch": 0.6303132327301149,
      "flos": 28293550675200.0,
      "grad_norm": 1.7122368213958923,
      "language_loss": 0.75129187,
      "learning_rate": 1.2702158090777278e-06,
      "loss": 0.77312946,
      "num_input_tokens_seen": 112860210,
      "step": 5242,
      "time_per_iteration": 3.3576436042785645
    },
    {
      "auxiliary_loss_clip": 0.01127615,
      "auxiliary_loss_mlp": 0.01025928,
      "balance_loss_clip": 1.0440861,
      "balance_loss_mlp": 1.01821852,
      "epoch": 0.6304334756207539,
      "flos": 25264521596160.0,
      "grad_norm": 1.906634468510918,
      "language_loss": 0.74925578,
      "learning_rate": 1.2694906030415148e-06,
      "loss": 0.77079117,
      "num_input_tokens_seen": 112877955,
      "step": 5243,
      "time_per_iteration": 2.5397820472717285
    },
    {
      "auxiliary_loss_clip": 0.01154239,
      "auxiliary_loss_mlp": 0.01028966,
      "balance_loss_clip": 1.04855275,
      "balance_loss_mlp": 1.02098536,
      "epoch": 0.630553718511393,
      "flos": 18033548728320.0,
      "grad_norm": 2.7027013290048454,
      "language_loss": 0.82178998,
      "learning_rate": 1.2687655078229958e-06,
      "loss": 0.84362197,
      "num_input_tokens_seen": 112892285,
      "step": 5244,
      "time_per_iteration": 2.6105520725250244
    },
    {
      "auxiliary_loss_clip": 0.01143502,
      "auxiliary_loss_mlp": 0.01025547,
      "balance_loss_clip": 1.04837537,
      "balance_loss_mlp": 1.01791167,
      "epoch": 0.6306739614020321,
      "flos": 27304103658240.0,
      "grad_norm": 2.2809779005503863,
      "language_loss": 0.68951988,
      "learning_rate": 1.2680405235321678e-06,
      "loss": 0.71121031,
      "num_input_tokens_seen": 112913620,
      "step": 5245,
      "time_per_iteration": 2.5363271236419678
    },
    {
      "auxiliary_loss_clip": 0.01148288,
      "auxiliary_loss_mlp": 0.00762787,
      "balance_loss_clip": 1.05169559,
      "balance_loss_mlp": 1.0003823,
      "epoch": 0.6307942042926712,
      "flos": 15341434243200.0,
      "grad_norm": 2.0723166847212067,
      "language_loss": 0.78863728,
      "learning_rate": 1.267315650279011e-06,
      "loss": 0.80774796,
      "num_input_tokens_seen": 112932090,
      "step": 5246,
      "time_per_iteration": 3.3224740028381348
    },
    {
      "auxiliary_loss_clip": 0.01126687,
      "auxiliary_loss_mlp": 0.01024207,
      "balance_loss_clip": 1.04773426,
      "balance_loss_mlp": 1.01682222,
      "epoch": 0.6309144471833102,
      "flos": 19606400444160.0,
      "grad_norm": 1.7387648797215294,
      "language_loss": 0.73967999,
      "learning_rate": 1.2665908881734874e-06,
      "loss": 0.76118898,
      "num_input_tokens_seen": 112950925,
      "step": 5247,
      "time_per_iteration": 3.412727117538452
    },
    {
      "auxiliary_loss_clip": 0.01159651,
      "auxiliary_loss_mlp": 0.01025458,
      "balance_loss_clip": 1.04971337,
      "balance_loss_mlp": 1.01848137,
      "epoch": 0.6310346900739494,
      "flos": 17493345112320.0,
      "grad_norm": 2.1975669661306965,
      "language_loss": 0.84660661,
      "learning_rate": 1.2658662373255432e-06,
      "loss": 0.86845767,
      "num_input_tokens_seen": 112969315,
      "step": 5248,
      "time_per_iteration": 2.5256261825561523
    },
    {
      "auxiliary_loss_clip": 0.01053738,
      "auxiliary_loss_mlp": 0.01002163,
      "balance_loss_clip": 1.01614547,
      "balance_loss_mlp": 1.00128686,
      "epoch": 0.6311549329645885,
      "flos": 55070164131840.0,
      "grad_norm": 0.7103915134079878,
      "language_loss": 0.52294707,
      "learning_rate": 1.2651416978451063e-06,
      "loss": 0.54350609,
      "num_input_tokens_seen": 113034700,
      "step": 5249,
      "time_per_iteration": 3.1693356037139893
    },
    {
      "auxiliary_loss_clip": 0.01175721,
      "auxiliary_loss_mlp": 0.01024664,
      "balance_loss_clip": 1.05043006,
      "balance_loss_mlp": 1.01685023,
      "epoch": 0.6312751758552275,
      "flos": 41902545075840.0,
      "grad_norm": 2.2658316028768093,
      "language_loss": 0.65159422,
      "learning_rate": 1.2644172698420903e-06,
      "loss": 0.67359805,
      "num_input_tokens_seen": 113056805,
      "step": 5250,
      "time_per_iteration": 3.312739372253418
    },
    {
      "auxiliary_loss_clip": 0.01131324,
      "auxiliary_loss_mlp": 0.0103052,
      "balance_loss_clip": 1.04585195,
      "balance_loss_mlp": 1.02242541,
      "epoch": 0.6313954187458667,
      "flos": 19646800266240.0,
      "grad_norm": 1.796287760389685,
      "language_loss": 0.84842646,
      "learning_rate": 1.2636929534263892e-06,
      "loss": 0.87004495,
      "num_input_tokens_seen": 113075790,
      "step": 5251,
      "time_per_iteration": 2.5325424671173096
    },
    {
      "auxiliary_loss_clip": 0.01131442,
      "auxiliary_loss_mlp": 0.01023171,
      "balance_loss_clip": 1.04186857,
      "balance_loss_mlp": 1.01551795,
      "epoch": 0.6315156616365057,
      "flos": 22894273906560.0,
      "grad_norm": 1.7231176928805898,
      "language_loss": 0.77580416,
      "learning_rate": 1.2629687487078821e-06,
      "loss": 0.79735029,
      "num_input_tokens_seen": 113094600,
      "step": 5252,
      "time_per_iteration": 2.551456928253174
    },
    {
      "auxiliary_loss_clip": 0.01162681,
      "auxiliary_loss_mlp": 0.01028536,
      "balance_loss_clip": 1.04751658,
      "balance_loss_mlp": 1.02010226,
      "epoch": 0.6316359045271448,
      "flos": 23726251699200.0,
      "grad_norm": 2.164319552760073,
      "language_loss": 0.76402354,
      "learning_rate": 1.2622446557964293e-06,
      "loss": 0.78593576,
      "num_input_tokens_seen": 113112605,
      "step": 5253,
      "time_per_iteration": 2.5192368030548096
    },
    {
      "auxiliary_loss_clip": 0.01143661,
      "auxiliary_loss_mlp": 0.01027515,
      "balance_loss_clip": 1.04325616,
      "balance_loss_mlp": 1.02047586,
      "epoch": 0.631756147417784,
      "flos": 33108417164160.0,
      "grad_norm": 1.6950993714776457,
      "language_loss": 0.71356583,
      "learning_rate": 1.261520674801876e-06,
      "loss": 0.73527753,
      "num_input_tokens_seen": 113133200,
      "step": 5254,
      "time_per_iteration": 2.6268553733825684
    },
    {
      "auxiliary_loss_clip": 0.01146707,
      "auxiliary_loss_mlp": 0.01026586,
      "balance_loss_clip": 1.05171251,
      "balance_loss_mlp": 1.01870036,
      "epoch": 0.631876390308423,
      "flos": 31248424126080.0,
      "grad_norm": 1.892572863066902,
      "language_loss": 0.72500235,
      "learning_rate": 1.2607968058340488e-06,
      "loss": 0.74673533,
      "num_input_tokens_seen": 113152895,
      "step": 5255,
      "time_per_iteration": 2.5809309482574463
    },
    {
      "auxiliary_loss_clip": 0.01140263,
      "auxiliary_loss_mlp": 0.01029561,
      "balance_loss_clip": 1.04521251,
      "balance_loss_mlp": 1.0217886,
      "epoch": 0.6319966331990621,
      "flos": 24681152810880.0,
      "grad_norm": 1.7664565370931442,
      "language_loss": 0.73207307,
      "learning_rate": 1.2600730490027583e-06,
      "loss": 0.75377136,
      "num_input_tokens_seen": 113173135,
      "step": 5256,
      "time_per_iteration": 2.557288408279419
    },
    {
      "auxiliary_loss_clip": 0.01131318,
      "auxiliary_loss_mlp": 0.01027553,
      "balance_loss_clip": 1.04668927,
      "balance_loss_mlp": 1.0197866,
      "epoch": 0.6321168760897012,
      "flos": 17491764913920.0,
      "grad_norm": 1.584690559047317,
      "language_loss": 0.80507481,
      "learning_rate": 1.2593494044177984e-06,
      "loss": 0.82666349,
      "num_input_tokens_seen": 113191440,
      "step": 5257,
      "time_per_iteration": 2.5556273460388184
    },
    {
      "auxiliary_loss_clip": 0.011768,
      "auxiliary_loss_mlp": 0.01026408,
      "balance_loss_clip": 1.04890287,
      "balance_loss_mlp": 1.01830173,
      "epoch": 0.6322371189803403,
      "flos": 18295373940480.0,
      "grad_norm": 2.251341422288516,
      "language_loss": 0.80602717,
      "learning_rate": 1.2586258721889448e-06,
      "loss": 0.8280592,
      "num_input_tokens_seen": 113208790,
      "step": 5258,
      "time_per_iteration": 2.4100778102874756
    },
    {
      "auxiliary_loss_clip": 0.01111202,
      "auxiliary_loss_mlp": 0.0102647,
      "balance_loss_clip": 1.0454185,
      "balance_loss_mlp": 1.01876307,
      "epoch": 0.6323573618709794,
      "flos": 20157270399360.0,
      "grad_norm": 1.9190900136527766,
      "language_loss": 0.81956846,
      "learning_rate": 1.2579024524259573e-06,
      "loss": 0.84094512,
      "num_input_tokens_seen": 113225050,
      "step": 5259,
      "time_per_iteration": 2.5287811756134033
    },
    {
      "auxiliary_loss_clip": 0.0113967,
      "auxiliary_loss_mlp": 0.01025239,
      "balance_loss_clip": 1.04355669,
      "balance_loss_mlp": 1.01740742,
      "epoch": 0.6324776047616185,
      "flos": 20042391726720.0,
      "grad_norm": 2.081989733055051,
      "language_loss": 0.91321421,
      "learning_rate": 1.2571791452385768e-06,
      "loss": 0.93486321,
      "num_input_tokens_seen": 113242315,
      "step": 5260,
      "time_per_iteration": 2.480090618133545
    },
    {
      "auxiliary_loss_clip": 0.01146337,
      "auxiliary_loss_mlp": 0.01031223,
      "balance_loss_clip": 1.04780018,
      "balance_loss_mlp": 1.02333784,
      "epoch": 0.6325978476522576,
      "flos": 30848235724800.0,
      "grad_norm": 1.567618844513022,
      "language_loss": 0.77061307,
      "learning_rate": 1.2564559507365301e-06,
      "loss": 0.79238868,
      "num_input_tokens_seen": 113264720,
      "step": 5261,
      "time_per_iteration": 2.5631837844848633
    },
    {
      "auxiliary_loss_clip": 0.01145494,
      "auxiliary_loss_mlp": 0.01025971,
      "balance_loss_clip": 1.04648447,
      "balance_loss_mlp": 1.01751292,
      "epoch": 0.6327180905428966,
      "flos": 24535104111360.0,
      "grad_norm": 2.611555615167692,
      "language_loss": 0.78617263,
      "learning_rate": 1.2557328690295244e-06,
      "loss": 0.80788732,
      "num_input_tokens_seen": 113282910,
      "step": 5262,
      "time_per_iteration": 2.5441393852233887
    },
    {
      "auxiliary_loss_clip": 0.01137522,
      "auxiliary_loss_mlp": 0.01026415,
      "balance_loss_clip": 1.04886937,
      "balance_loss_mlp": 1.01872611,
      "epoch": 0.6328383334335358,
      "flos": 21575274583680.0,
      "grad_norm": 1.863550005022295,
      "language_loss": 0.76022011,
      "learning_rate": 1.255009900227251e-06,
      "loss": 0.78185952,
      "num_input_tokens_seen": 113301935,
      "step": 5263,
      "time_per_iteration": 2.5330796241760254
    },
    {
      "auxiliary_loss_clip": 0.01170232,
      "auxiliary_loss_mlp": 0.01024154,
      "balance_loss_clip": 1.05040181,
      "balance_loss_mlp": 1.01716805,
      "epoch": 0.6329585763241748,
      "flos": 22929861306240.0,
      "grad_norm": 1.7881547881459787,
      "language_loss": 0.79132324,
      "learning_rate": 1.254287044439383e-06,
      "loss": 0.81326705,
      "num_input_tokens_seen": 113321540,
      "step": 5264,
      "time_per_iteration": 2.437756299972534
    },
    {
      "auxiliary_loss_clip": 0.0107299,
      "auxiliary_loss_mlp": 0.01001186,
      "balance_loss_clip": 1.01553082,
      "balance_loss_mlp": 1.00032794,
      "epoch": 0.6330788192148139,
      "flos": 70936897847040.0,
      "grad_norm": 0.779024323654976,
      "language_loss": 0.54449677,
      "learning_rate": 1.2535643017755776e-06,
      "loss": 0.56523854,
      "num_input_tokens_seen": 113383730,
      "step": 5265,
      "time_per_iteration": 3.1355056762695312
    },
    {
      "auxiliary_loss_clip": 0.01130683,
      "auxiliary_loss_mlp": 0.01032311,
      "balance_loss_clip": 1.04431105,
      "balance_loss_mlp": 1.02402544,
      "epoch": 0.6331990621054531,
      "flos": 21244501215360.0,
      "grad_norm": 2.595743669949413,
      "language_loss": 0.72268641,
      "learning_rate": 1.2528416723454737e-06,
      "loss": 0.74431634,
      "num_input_tokens_seen": 113400400,
      "step": 5266,
      "time_per_iteration": 2.511258602142334
    },
    {
      "auxiliary_loss_clip": 0.01171116,
      "auxiliary_loss_mlp": 0.01023696,
      "balance_loss_clip": 1.05074954,
      "balance_loss_mlp": 1.01669884,
      "epoch": 0.6333193049960921,
      "flos": 34459412526720.0,
      "grad_norm": 1.4988909305434233,
      "language_loss": 0.70961404,
      "learning_rate": 1.2521191562586945e-06,
      "loss": 0.73156214,
      "num_input_tokens_seen": 113424050,
      "step": 5267,
      "time_per_iteration": 2.527148485183716
    },
    {
      "auxiliary_loss_clip": 0.01174584,
      "auxiliary_loss_mlp": 0.007623,
      "balance_loss_clip": 1.05160499,
      "balance_loss_mlp": 1.00047827,
      "epoch": 0.6334395478867312,
      "flos": 18329883932160.0,
      "grad_norm": 2.3017376266098637,
      "language_loss": 0.76833642,
      "learning_rate": 1.2513967536248445e-06,
      "loss": 0.78770524,
      "num_input_tokens_seen": 113440370,
      "step": 5268,
      "time_per_iteration": 2.4046616554260254
    },
    {
      "auxiliary_loss_clip": 0.01154731,
      "auxiliary_loss_mlp": 0.01028181,
      "balance_loss_clip": 1.04934096,
      "balance_loss_mlp": 1.02093577,
      "epoch": 0.6335597907773702,
      "flos": 23623152687360.0,
      "grad_norm": 1.5746507331262924,
      "language_loss": 0.80809134,
      "learning_rate": 1.2506744645535117e-06,
      "loss": 0.82992047,
      "num_input_tokens_seen": 113460800,
      "step": 5269,
      "time_per_iteration": 3.2198593616485596
    },
    {
      "auxiliary_loss_clip": 0.01136489,
      "auxiliary_loss_mlp": 0.01021628,
      "balance_loss_clip": 1.04161441,
      "balance_loss_mlp": 1.01389718,
      "epoch": 0.6336800336680094,
      "flos": 22710913954560.0,
      "grad_norm": 7.694002005438443,
      "language_loss": 0.604617,
      "learning_rate": 1.249952289154267e-06,
      "loss": 0.62619817,
      "num_input_tokens_seen": 113480840,
      "step": 5270,
      "time_per_iteration": 2.5100300312042236
    },
    {
      "auxiliary_loss_clip": 0.01090882,
      "auxiliary_loss_mlp": 0.01028785,
      "balance_loss_clip": 1.04044759,
      "balance_loss_mlp": 1.0215013,
      "epoch": 0.6338002765586485,
      "flos": 23622757637760.0,
      "grad_norm": 2.075392247291788,
      "language_loss": 0.76413989,
      "learning_rate": 1.2492302275366635e-06,
      "loss": 0.78533655,
      "num_input_tokens_seen": 113500515,
      "step": 5271,
      "time_per_iteration": 2.5960912704467773
    },
    {
      "auxiliary_loss_clip": 0.01152743,
      "auxiliary_loss_mlp": 0.01027776,
      "balance_loss_clip": 1.04660082,
      "balance_loss_mlp": 1.02003384,
      "epoch": 0.6339205194492875,
      "flos": 26505450708480.0,
      "grad_norm": 2.546027301746887,
      "language_loss": 0.65856743,
      "learning_rate": 1.2485082798102377e-06,
      "loss": 0.6803726,
      "num_input_tokens_seen": 113520930,
      "step": 5272,
      "time_per_iteration": 2.497006416320801
    },
    {
      "auxiliary_loss_clip": 0.01135857,
      "auxiliary_loss_mlp": 0.01024821,
      "balance_loss_clip": 1.04543376,
      "balance_loss_mlp": 1.01626766,
      "epoch": 0.6340407623399267,
      "flos": 18544306170240.0,
      "grad_norm": 2.237009596368795,
      "language_loss": 0.68699163,
      "learning_rate": 1.2477864460845084e-06,
      "loss": 0.70859838,
      "num_input_tokens_seen": 113537330,
      "step": 5273,
      "time_per_iteration": 3.34025239944458
    },
    {
      "auxiliary_loss_clip": 0.01143964,
      "auxiliary_loss_mlp": 0.0102555,
      "balance_loss_clip": 1.0454905,
      "balance_loss_mlp": 1.01764035,
      "epoch": 0.6341610052305657,
      "flos": 17712579772800.0,
      "grad_norm": 2.7675679732899767,
      "language_loss": 0.7315557,
      "learning_rate": 1.2470647264689776e-06,
      "loss": 0.75325084,
      "num_input_tokens_seen": 113555810,
      "step": 5274,
      "time_per_iteration": 3.3289694786071777
    },
    {
      "auxiliary_loss_clip": 0.01110216,
      "auxiliary_loss_mlp": 0.01026458,
      "balance_loss_clip": 1.04107213,
      "balance_loss_mlp": 1.01845288,
      "epoch": 0.6342812481212048,
      "flos": 23587026583680.0,
      "grad_norm": 15.072758610497422,
      "language_loss": 0.70952404,
      "learning_rate": 1.2463431210731282e-06,
      "loss": 0.73089075,
      "num_input_tokens_seen": 113575395,
      "step": 5275,
      "time_per_iteration": 2.6223363876342773
    },
    {
      "auxiliary_loss_clip": 0.01125579,
      "auxiliary_loss_mlp": 0.01022951,
      "balance_loss_clip": 1.04233372,
      "balance_loss_mlp": 1.01567364,
      "epoch": 0.634401491011844,
      "flos": 17821927751040.0,
      "grad_norm": 2.4653105526294388,
      "language_loss": 0.76252979,
      "learning_rate": 1.2456216300064289e-06,
      "loss": 0.78401506,
      "num_input_tokens_seen": 113592945,
      "step": 5276,
      "time_per_iteration": 3.3060522079467773
    },
    {
      "auxiliary_loss_clip": 0.01128194,
      "auxiliary_loss_mlp": 0.01025351,
      "balance_loss_clip": 1.04329205,
      "balance_loss_mlp": 1.01741791,
      "epoch": 0.634521733902483,
      "flos": 21358158825600.0,
      "grad_norm": 1.6074230619354897,
      "language_loss": 0.78452903,
      "learning_rate": 1.244900253378328e-06,
      "loss": 0.80606449,
      "num_input_tokens_seen": 113613000,
      "step": 5277,
      "time_per_iteration": 2.549111843109131
    },
    {
      "auxiliary_loss_clip": 0.01086331,
      "auxiliary_loss_mlp": 0.0102657,
      "balance_loss_clip": 1.04237294,
      "balance_loss_mlp": 1.01922965,
      "epoch": 0.6346419767931221,
      "flos": 16545052103040.0,
      "grad_norm": 1.8858344114320909,
      "language_loss": 0.69168562,
      "learning_rate": 1.2441789912982583e-06,
      "loss": 0.71281463,
      "num_input_tokens_seen": 113630085,
      "step": 5278,
      "time_per_iteration": 2.5837655067443848
    },
    {
      "auxiliary_loss_clip": 0.01162283,
      "auxiliary_loss_mlp": 0.01022699,
      "balance_loss_clip": 1.04936278,
      "balance_loss_mlp": 1.0149262,
      "epoch": 0.6347622196837612,
      "flos": 24350989973760.0,
      "grad_norm": 1.8158614743689814,
      "language_loss": 0.64959836,
      "learning_rate": 1.2434578438756346e-06,
      "loss": 0.67144823,
      "num_input_tokens_seen": 113650515,
      "step": 5279,
      "time_per_iteration": 2.507667064666748
    },
    {
      "auxiliary_loss_clip": 0.01157929,
      "auxiliary_loss_mlp": 0.01022055,
      "balance_loss_clip": 1.04682398,
      "balance_loss_mlp": 1.01469946,
      "epoch": 0.6348824625744003,
      "flos": 64523178195840.0,
      "grad_norm": 1.9204991286030741,
      "language_loss": 0.77853382,
      "learning_rate": 1.242736811219855e-06,
      "loss": 0.80033362,
      "num_input_tokens_seen": 113676475,
      "step": 5280,
      "time_per_iteration": 2.8520243167877197
    },
    {
      "auxiliary_loss_clip": 0.01154301,
      "auxiliary_loss_mlp": 0.01024287,
      "balance_loss_clip": 1.04774368,
      "balance_loss_mlp": 1.01656866,
      "epoch": 0.6350027054650393,
      "flos": 28622133313920.0,
      "grad_norm": 2.0580047871682092,
      "language_loss": 0.81554079,
      "learning_rate": 1.2420158934402988e-06,
      "loss": 0.83732677,
      "num_input_tokens_seen": 113697090,
      "step": 5281,
      "time_per_iteration": 2.518745183944702
    },
    {
      "auxiliary_loss_clip": 0.011162,
      "auxiliary_loss_mlp": 0.01024988,
      "balance_loss_clip": 1.04131126,
      "balance_loss_mlp": 1.01732886,
      "epoch": 0.6351229483556785,
      "flos": 23002544476800.0,
      "grad_norm": 1.8350585422268777,
      "language_loss": 0.84928542,
      "learning_rate": 1.2412950906463286e-06,
      "loss": 0.87069726,
      "num_input_tokens_seen": 113714395,
      "step": 5282,
      "time_per_iteration": 2.5169124603271484
    },
    {
      "auxiliary_loss_clip": 0.01115645,
      "auxiliary_loss_mlp": 0.01026344,
      "balance_loss_clip": 1.0443542,
      "balance_loss_mlp": 1.01885462,
      "epoch": 0.6352431912463176,
      "flos": 21939300967680.0,
      "grad_norm": 1.7217629415082754,
      "language_loss": 0.89771909,
      "learning_rate": 1.2405744029472902e-06,
      "loss": 0.91913891,
      "num_input_tokens_seen": 113733880,
      "step": 5283,
      "time_per_iteration": 2.570225715637207
    },
    {
      "auxiliary_loss_clip": 0.01141719,
      "auxiliary_loss_mlp": 0.01024448,
      "balance_loss_clip": 1.04499722,
      "balance_loss_mlp": 1.01726615,
      "epoch": 0.6353634341369566,
      "flos": 13735257684480.0,
      "grad_norm": 2.3437238379443532,
      "language_loss": 0.76385826,
      "learning_rate": 1.2398538304525108e-06,
      "loss": 0.78551996,
      "num_input_tokens_seen": 113752505,
      "step": 5284,
      "time_per_iteration": 2.468639612197876
    },
    {
      "auxiliary_loss_clip": 0.01117207,
      "auxiliary_loss_mlp": 0.01034064,
      "balance_loss_clip": 1.04538131,
      "balance_loss_mlp": 1.0256952,
      "epoch": 0.6354836770275958,
      "flos": 19316170552320.0,
      "grad_norm": 1.973459835833161,
      "language_loss": 0.75322002,
      "learning_rate": 1.2391333732713016e-06,
      "loss": 0.77473271,
      "num_input_tokens_seen": 113770310,
      "step": 5285,
      "time_per_iteration": 2.5357022285461426
    },
    {
      "auxiliary_loss_clip": 0.01121697,
      "auxiliary_loss_mlp": 0.01037953,
      "balance_loss_clip": 1.04391575,
      "balance_loss_mlp": 1.02874374,
      "epoch": 0.6356039199182348,
      "flos": 21613375935360.0,
      "grad_norm": 2.091643414321461,
      "language_loss": 0.78670728,
      "learning_rate": 1.2384130315129543e-06,
      "loss": 0.80830371,
      "num_input_tokens_seen": 113788635,
      "step": 5286,
      "time_per_iteration": 2.5537593364715576
    },
    {
      "auxiliary_loss_clip": 0.01091519,
      "auxiliary_loss_mlp": 0.0102581,
      "balance_loss_clip": 1.04289269,
      "balance_loss_mlp": 1.01763833,
      "epoch": 0.6357241628088739,
      "flos": 18111978074880.0,
      "grad_norm": 2.5820126966692167,
      "language_loss": 0.73408329,
      "learning_rate": 1.2376928052867447e-06,
      "loss": 0.75525659,
      "num_input_tokens_seen": 113807755,
      "step": 5287,
      "time_per_iteration": 2.60221266746521
    },
    {
      "auxiliary_loss_clip": 0.01145728,
      "auxiliary_loss_mlp": 0.01026239,
      "balance_loss_clip": 1.04870617,
      "balance_loss_mlp": 1.01874113,
      "epoch": 0.6358444056995131,
      "flos": 24935256599040.0,
      "grad_norm": 1.900953781513786,
      "language_loss": 0.77475965,
      "learning_rate": 1.2369726947019299e-06,
      "loss": 0.79647923,
      "num_input_tokens_seen": 113828230,
      "step": 5288,
      "time_per_iteration": 2.5208442211151123
    },
    {
      "auxiliary_loss_clip": 0.01156505,
      "auxiliary_loss_mlp": 0.01020329,
      "balance_loss_clip": 1.04632115,
      "balance_loss_mlp": 1.01276457,
      "epoch": 0.6359646485901521,
      "flos": 23293348986240.0,
      "grad_norm": 2.162948496529071,
      "language_loss": 0.67297655,
      "learning_rate": 1.2362526998677511e-06,
      "loss": 0.69474483,
      "num_input_tokens_seen": 113844595,
      "step": 5289,
      "time_per_iteration": 2.457869529724121
    },
    {
      "auxiliary_loss_clip": 0.01148012,
      "auxiliary_loss_mlp": 0.01027002,
      "balance_loss_clip": 1.04748273,
      "balance_loss_mlp": 1.01988482,
      "epoch": 0.6360848914807912,
      "flos": 20887442069760.0,
      "grad_norm": 1.8536267063473426,
      "language_loss": 0.84292012,
      "learning_rate": 1.2355328208934301e-06,
      "loss": 0.86467028,
      "num_input_tokens_seen": 113863470,
      "step": 5290,
      "time_per_iteration": 2.492821216583252
    },
    {
      "auxiliary_loss_clip": 0.01156999,
      "auxiliary_loss_mlp": 0.00762547,
      "balance_loss_clip": 1.04509509,
      "balance_loss_mlp": 1.00057101,
      "epoch": 0.6362051343714303,
      "flos": 18479775386880.0,
      "grad_norm": 1.6114396935860047,
      "language_loss": 0.7220962,
      "learning_rate": 1.2348130578881728e-06,
      "loss": 0.74129164,
      "num_input_tokens_seen": 113881690,
      "step": 5291,
      "time_per_iteration": 2.4520437717437744
    },
    {
      "auxiliary_loss_clip": 0.01174524,
      "auxiliary_loss_mlp": 0.0102957,
      "balance_loss_clip": 1.04988742,
      "balance_loss_mlp": 1.02157152,
      "epoch": 0.6363253772620694,
      "flos": 24389594115840.0,
      "grad_norm": 2.519250035934543,
      "language_loss": 0.76353866,
      "learning_rate": 1.2340934109611664e-06,
      "loss": 0.78557956,
      "num_input_tokens_seen": 113902450,
      "step": 5292,
      "time_per_iteration": 2.452216148376465
    },
    {
      "auxiliary_loss_clip": 0.01150948,
      "auxiliary_loss_mlp": 0.01026443,
      "balance_loss_clip": 1.04919553,
      "balance_loss_mlp": 1.01784182,
      "epoch": 0.6364456201527084,
      "flos": 25958243940480.0,
      "grad_norm": 1.9950303360599584,
      "language_loss": 0.68935609,
      "learning_rate": 1.2333738802215798e-06,
      "loss": 0.71112996,
      "num_input_tokens_seen": 113922670,
      "step": 5293,
      "time_per_iteration": 2.5406911373138428
    },
    {
      "auxiliary_loss_clip": 0.01108759,
      "auxiliary_loss_mlp": 0.01025938,
      "balance_loss_clip": 1.04078245,
      "balance_loss_mlp": 1.01828742,
      "epoch": 0.6365658630433476,
      "flos": 20740711011840.0,
      "grad_norm": 1.8296366015077776,
      "language_loss": 0.80890852,
      "learning_rate": 1.2326544657785668e-06,
      "loss": 0.83025551,
      "num_input_tokens_seen": 113942360,
      "step": 5294,
      "time_per_iteration": 2.544131278991699
    },
    {
      "auxiliary_loss_clip": 0.01122527,
      "auxiliary_loss_mlp": 0.01033007,
      "balance_loss_clip": 1.04422414,
      "balance_loss_mlp": 1.02450192,
      "epoch": 0.6366861059339867,
      "flos": 21434146047360.0,
      "grad_norm": 2.83681433898636,
      "language_loss": 0.74706316,
      "learning_rate": 1.2319351677412608e-06,
      "loss": 0.76861858,
      "num_input_tokens_seen": 113959405,
      "step": 5295,
      "time_per_iteration": 3.3298044204711914
    },
    {
      "auxiliary_loss_clip": 0.01140489,
      "auxiliary_loss_mlp": 0.01026381,
      "balance_loss_clip": 1.04899931,
      "balance_loss_mlp": 1.01842988,
      "epoch": 0.6368063488246257,
      "flos": 22267093507200.0,
      "grad_norm": 2.28397664650324,
      "language_loss": 0.74407566,
      "learning_rate": 1.2312159862187796e-06,
      "loss": 0.76574433,
      "num_input_tokens_seen": 113977815,
      "step": 5296,
      "time_per_iteration": 2.5113227367401123
    },
    {
      "auxiliary_loss_clip": 0.01177998,
      "auxiliary_loss_mlp": 0.01034236,
      "balance_loss_clip": 1.05291986,
      "balance_loss_mlp": 1.02645445,
      "epoch": 0.6369265917152649,
      "flos": 22420719976320.0,
      "grad_norm": 2.688298999957881,
      "language_loss": 0.75958943,
      "learning_rate": 1.2304969213202217e-06,
      "loss": 0.78171176,
      "num_input_tokens_seen": 113999075,
      "step": 5297,
      "time_per_iteration": 2.453690767288208
    },
    {
      "auxiliary_loss_clip": 0.01139792,
      "auxiliary_loss_mlp": 0.01024813,
      "balance_loss_clip": 1.0453496,
      "balance_loss_mlp": 1.01769006,
      "epoch": 0.6370468346059039,
      "flos": 24718176754560.0,
      "grad_norm": 2.2552513326488963,
      "language_loss": 0.79033393,
      "learning_rate": 1.2297779731546692e-06,
      "loss": 0.81198001,
      "num_input_tokens_seen": 114018170,
      "step": 5298,
      "time_per_iteration": 2.5117056369781494
    },
    {
      "auxiliary_loss_clip": 0.01143514,
      "auxiliary_loss_mlp": 0.01029705,
      "balance_loss_clip": 1.04846466,
      "balance_loss_mlp": 1.02180135,
      "epoch": 0.637167077496543,
      "flos": 25296589463040.0,
      "grad_norm": 1.9199288917648691,
      "language_loss": 0.78064179,
      "learning_rate": 1.2290591418311853e-06,
      "loss": 0.80237395,
      "num_input_tokens_seen": 114035565,
      "step": 5299,
      "time_per_iteration": 3.3526906967163086
    },
    {
      "auxiliary_loss_clip": 0.01161287,
      "auxiliary_loss_mlp": 0.01027273,
      "balance_loss_clip": 1.05219555,
      "balance_loss_mlp": 1.01969123,
      "epoch": 0.637287320387182,
      "flos": 27671110871040.0,
      "grad_norm": 1.6197946487048385,
      "language_loss": 0.72045326,
      "learning_rate": 1.2283404274588172e-06,
      "loss": 0.7423389,
      "num_input_tokens_seen": 114054510,
      "step": 5300,
      "time_per_iteration": 3.367604970932007
    },
    {
      "auxiliary_loss_clip": 0.00996388,
      "auxiliary_loss_mlp": 0.01004192,
      "balance_loss_clip": 1.01321888,
      "balance_loss_mlp": 1.00309527,
      "epoch": 0.6374075632778212,
      "flos": 63173406873600.0,
      "grad_norm": 0.7427944534174411,
      "language_loss": 0.52826333,
      "learning_rate": 1.227621830146592e-06,
      "loss": 0.54826909,
      "num_input_tokens_seen": 114109875,
      "step": 5301,
      "time_per_iteration": 3.1422998905181885
    },
    {
      "auxiliary_loss_clip": 0.01139577,
      "auxiliary_loss_mlp": 0.01030294,
      "balance_loss_clip": 1.05139172,
      "balance_loss_mlp": 1.02269101,
      "epoch": 0.6375278061684603,
      "flos": 25558127366400.0,
      "grad_norm": 1.9973624382908963,
      "language_loss": 0.79488832,
      "learning_rate": 1.2269033500035217e-06,
      "loss": 0.81658697,
      "num_input_tokens_seen": 114130010,
      "step": 5302,
      "time_per_iteration": 2.84183406829834
    },
    {
      "auxiliary_loss_clip": 0.011305,
      "auxiliary_loss_mlp": 0.01030061,
      "balance_loss_clip": 1.04673529,
      "balance_loss_mlp": 1.02270937,
      "epoch": 0.6376480490590993,
      "flos": 25666362023040.0,
      "grad_norm": 1.6599732459490695,
      "language_loss": 0.73446679,
      "learning_rate": 1.2261849871385988e-06,
      "loss": 0.7560724,
      "num_input_tokens_seen": 114151115,
      "step": 5303,
      "time_per_iteration": 3.293503522872925
    },
    {
      "auxiliary_loss_clip": 0.01174423,
      "auxiliary_loss_mlp": 0.01023777,
      "balance_loss_clip": 1.04949498,
      "balance_loss_mlp": 1.01572454,
      "epoch": 0.6377682919497385,
      "flos": 31537684350720.0,
      "grad_norm": 2.2285008761150786,
      "language_loss": 0.6242066,
      "learning_rate": 1.2254667416607972e-06,
      "loss": 0.64618862,
      "num_input_tokens_seen": 114172715,
      "step": 5304,
      "time_per_iteration": 2.506847858428955
    },
    {
      "auxiliary_loss_clip": 0.01157954,
      "auxiliary_loss_mlp": 0.01022167,
      "balance_loss_clip": 1.0504241,
      "balance_loss_mlp": 1.01438594,
      "epoch": 0.6378885348403776,
      "flos": 23039209284480.0,
      "grad_norm": 1.8430296024005839,
      "language_loss": 0.82953912,
      "learning_rate": 1.2247486136790756e-06,
      "loss": 0.85134029,
      "num_input_tokens_seen": 114192195,
      "step": 5305,
      "time_per_iteration": 2.461240291595459
    },
    {
      "auxiliary_loss_clip": 0.01163064,
      "auxiliary_loss_mlp": 0.01032924,
      "balance_loss_clip": 1.05182195,
      "balance_loss_mlp": 1.02541375,
      "epoch": 0.6380087777310166,
      "flos": 18697070712960.0,
      "grad_norm": 2.080045351034289,
      "language_loss": 0.80872786,
      "learning_rate": 1.2240306033023726e-06,
      "loss": 0.83068776,
      "num_input_tokens_seen": 114210020,
      "step": 5306,
      "time_per_iteration": 2.432070255279541
    },
    {
      "auxiliary_loss_clip": 0.01131885,
      "auxiliary_loss_mlp": 0.01028584,
      "balance_loss_clip": 1.04243898,
      "balance_loss_mlp": 1.02046561,
      "epoch": 0.6381290206216558,
      "flos": 23331558078720.0,
      "grad_norm": 1.8907804918699203,
      "language_loss": 0.72097391,
      "learning_rate": 1.223312710639611e-06,
      "loss": 0.74257857,
      "num_input_tokens_seen": 114228740,
      "step": 5307,
      "time_per_iteration": 2.52185320854187
    },
    {
      "auxiliary_loss_clip": 0.01144887,
      "auxiliary_loss_mlp": 0.01026909,
      "balance_loss_clip": 1.04798126,
      "balance_loss_mlp": 1.01893127,
      "epoch": 0.6382492635122948,
      "flos": 18880466578560.0,
      "grad_norm": 2.1403445603410773,
      "language_loss": 0.86983752,
      "learning_rate": 1.2225949357996928e-06,
      "loss": 0.89155549,
      "num_input_tokens_seen": 114246865,
      "step": 5308,
      "time_per_iteration": 2.453169107437134
    },
    {
      "auxiliary_loss_clip": 0.01156393,
      "auxiliary_loss_mlp": 0.01026432,
      "balance_loss_clip": 1.049945,
      "balance_loss_mlp": 1.01925528,
      "epoch": 0.6383695064029339,
      "flos": 27819134818560.0,
      "grad_norm": 1.4945758287424313,
      "language_loss": 0.80474615,
      "learning_rate": 1.221877278891505e-06,
      "loss": 0.82657439,
      "num_input_tokens_seen": 114266120,
      "step": 5309,
      "time_per_iteration": 2.493582248687744
    },
    {
      "auxiliary_loss_clip": 0.01166133,
      "auxiliary_loss_mlp": 0.0103186,
      "balance_loss_clip": 1.05068755,
      "balance_loss_mlp": 1.0235033,
      "epoch": 0.638489749293573,
      "flos": 26395635853440.0,
      "grad_norm": 1.9957006960726937,
      "language_loss": 0.71371841,
      "learning_rate": 1.221159740023915e-06,
      "loss": 0.7356984,
      "num_input_tokens_seen": 114285950,
      "step": 5310,
      "time_per_iteration": 2.477654218673706
    },
    {
      "auxiliary_loss_clip": 0.01140549,
      "auxiliary_loss_mlp": 0.00762534,
      "balance_loss_clip": 1.04776287,
      "balance_loss_mlp": 1.00055432,
      "epoch": 0.6386099921842121,
      "flos": 23988328306560.0,
      "grad_norm": 1.8415529101352488,
      "language_loss": 0.72419107,
      "learning_rate": 1.2204423193057735e-06,
      "loss": 0.74322188,
      "num_input_tokens_seen": 114304780,
      "step": 5311,
      "time_per_iteration": 2.5925655364990234
    },
    {
      "auxiliary_loss_clip": 0.01052044,
      "auxiliary_loss_mlp": 0.0100202,
      "balance_loss_clip": 1.01491857,
      "balance_loss_mlp": 1.00102472,
      "epoch": 0.6387302350748512,
      "flos": 71731169337600.0,
      "grad_norm": 0.8919463888173281,
      "language_loss": 0.63398248,
      "learning_rate": 1.2197250168459122e-06,
      "loss": 0.65452307,
      "num_input_tokens_seen": 114361180,
      "step": 5312,
      "time_per_iteration": 3.088937997817993
    },
    {
      "auxiliary_loss_clip": 0.01161212,
      "auxiliary_loss_mlp": 0.01025727,
      "balance_loss_clip": 1.04964519,
      "balance_loss_mlp": 1.01815724,
      "epoch": 0.6388504779654903,
      "flos": 14535778141440.0,
      "grad_norm": 1.9262403540987723,
      "language_loss": 0.74465376,
      "learning_rate": 1.2190078327531454e-06,
      "loss": 0.76652312,
      "num_input_tokens_seen": 114377425,
      "step": 5313,
      "time_per_iteration": 2.45031476020813
    },
    {
      "auxiliary_loss_clip": 0.01159214,
      "auxiliary_loss_mlp": 0.01029459,
      "balance_loss_clip": 1.04774022,
      "balance_loss_mlp": 1.02214575,
      "epoch": 0.6389707208561294,
      "flos": 22346133384960.0,
      "grad_norm": 1.4173791068018777,
      "language_loss": 0.7263236,
      "learning_rate": 1.2182907671362697e-06,
      "loss": 0.74821031,
      "num_input_tokens_seen": 114398120,
      "step": 5314,
      "time_per_iteration": 2.528268814086914
    },
    {
      "auxiliary_loss_clip": 0.0115957,
      "auxiliary_loss_mlp": 0.01025745,
      "balance_loss_clip": 1.05124092,
      "balance_loss_mlp": 1.01796103,
      "epoch": 0.6390909637467684,
      "flos": 19426883247360.0,
      "grad_norm": 1.7790856853570518,
      "language_loss": 0.78865641,
      "learning_rate": 1.2175738201040626e-06,
      "loss": 0.81050956,
      "num_input_tokens_seen": 114415160,
      "step": 5315,
      "time_per_iteration": 2.4518160820007324
    },
    {
      "auxiliary_loss_clip": 0.01157315,
      "auxiliary_loss_mlp": 0.01029092,
      "balance_loss_clip": 1.04750371,
      "balance_loss_mlp": 1.02150989,
      "epoch": 0.6392112066374076,
      "flos": 24090852700800.0,
      "grad_norm": 1.7995876809726286,
      "language_loss": 0.78570986,
      "learning_rate": 1.2168569917652855e-06,
      "loss": 0.80757391,
      "num_input_tokens_seen": 114435015,
      "step": 5316,
      "time_per_iteration": 2.4894604682922363
    },
    {
      "auxiliary_loss_clip": 0.01161029,
      "auxiliary_loss_mlp": 0.0102335,
      "balance_loss_clip": 1.05093455,
      "balance_loss_mlp": 1.01529157,
      "epoch": 0.6393314495280467,
      "flos": 26795141896320.0,
      "grad_norm": 1.5124650012420744,
      "language_loss": 0.63884765,
      "learning_rate": 1.2161402822286797e-06,
      "loss": 0.66069144,
      "num_input_tokens_seen": 114455700,
      "step": 5317,
      "time_per_iteration": 2.4946131706237793
    },
    {
      "auxiliary_loss_clip": 0.01127592,
      "auxiliary_loss_mlp": 0.01021914,
      "balance_loss_clip": 1.04453397,
      "balance_loss_mlp": 1.01433849,
      "epoch": 0.6394516924186857,
      "flos": 20260692633600.0,
      "grad_norm": 2.1216247270816617,
      "language_loss": 0.78773034,
      "learning_rate": 1.2154236916029703e-06,
      "loss": 0.80922544,
      "num_input_tokens_seen": 114473675,
      "step": 5318,
      "time_per_iteration": 2.497843027114868
    },
    {
      "auxiliary_loss_clip": 0.01115721,
      "auxiliary_loss_mlp": 0.01024901,
      "balance_loss_clip": 1.04071355,
      "balance_loss_mlp": 1.01742077,
      "epoch": 0.6395719353093249,
      "flos": 18368847210240.0,
      "grad_norm": 2.534705224720435,
      "language_loss": 0.73448962,
      "learning_rate": 1.2147072199968627e-06,
      "loss": 0.75589585,
      "num_input_tokens_seen": 114492310,
      "step": 5319,
      "time_per_iteration": 2.5192885398864746
    },
    {
      "auxiliary_loss_clip": 0.01156808,
      "auxiliary_loss_mlp": 0.01026179,
      "balance_loss_clip": 1.04859042,
      "balance_loss_mlp": 1.01899624,
      "epoch": 0.6396921781999639,
      "flos": 17566315591680.0,
      "grad_norm": 1.687268839794007,
      "language_loss": 0.71678048,
      "learning_rate": 1.2139908675190454e-06,
      "loss": 0.73861033,
      "num_input_tokens_seen": 114511520,
      "step": 5320,
      "time_per_iteration": 2.4200022220611572
    },
    {
      "auxiliary_loss_clip": 0.01091807,
      "auxiliary_loss_mlp": 0.01025279,
      "balance_loss_clip": 1.03890812,
      "balance_loss_mlp": 1.01778078,
      "epoch": 0.639812421090603,
      "flos": 21251252972160.0,
      "grad_norm": 2.1242980023188798,
      "language_loss": 0.75461257,
      "learning_rate": 1.2132746342781883e-06,
      "loss": 0.77578342,
      "num_input_tokens_seen": 114532680,
      "step": 5321,
      "time_per_iteration": 2.600229024887085
    },
    {
      "auxiliary_loss_clip": 0.0117499,
      "auxiliary_loss_mlp": 0.01026994,
      "balance_loss_clip": 1.05134273,
      "balance_loss_mlp": 1.0186435,
      "epoch": 0.6399326639812422,
      "flos": 11180967684480.0,
      "grad_norm": 2.4183223498876187,
      "language_loss": 0.79890311,
      "learning_rate": 1.2125585203829442e-06,
      "loss": 0.82092297,
      "num_input_tokens_seen": 114548320,
      "step": 5322,
      "time_per_iteration": 3.1619529724121094
    },
    {
      "auxiliary_loss_clip": 0.01118718,
      "auxiliary_loss_mlp": 0.01030233,
      "balance_loss_clip": 1.04556656,
      "balance_loss_mlp": 1.02222228,
      "epoch": 0.6400529068718812,
      "flos": 23911048195200.0,
      "grad_norm": 4.753229500658536,
      "language_loss": 0.74411368,
      "learning_rate": 1.211842525941946e-06,
      "loss": 0.76560318,
      "num_input_tokens_seen": 114568115,
      "step": 5323,
      "time_per_iteration": 2.5193850994110107
    },
    {
      "auxiliary_loss_clip": 0.0111287,
      "auxiliary_loss_mlp": 0.01025994,
      "balance_loss_clip": 1.04585505,
      "balance_loss_mlp": 1.01857877,
      "epoch": 0.6401731497625203,
      "flos": 44018724890880.0,
      "grad_norm": 1.8372634232212812,
      "language_loss": 0.78586113,
      "learning_rate": 1.2111266510638105e-06,
      "loss": 0.80724972,
      "num_input_tokens_seen": 114591040,
      "step": 5324,
      "time_per_iteration": 2.7600488662719727
    },
    {
      "auxiliary_loss_clip": 0.01094668,
      "auxiliary_loss_mlp": 0.01026398,
      "balance_loss_clip": 1.04182291,
      "balance_loss_mlp": 1.01886964,
      "epoch": 0.6402933926531594,
      "flos": 20662209838080.0,
      "grad_norm": 1.7788410479294903,
      "language_loss": 0.80138856,
      "learning_rate": 1.2104108958571346e-06,
      "loss": 0.82259917,
      "num_input_tokens_seen": 114609310,
      "step": 5325,
      "time_per_iteration": 2.6203114986419678
    },
    {
      "auxiliary_loss_clip": 0.01155702,
      "auxiliary_loss_mlp": 0.01028062,
      "balance_loss_clip": 1.04929554,
      "balance_loss_mlp": 1.02068889,
      "epoch": 0.6404136355437985,
      "flos": 24863327614080.0,
      "grad_norm": 1.4649566621947558,
      "language_loss": 0.75913548,
      "learning_rate": 1.2096952604304975e-06,
      "loss": 0.7809732,
      "num_input_tokens_seen": 114629740,
      "step": 5326,
      "time_per_iteration": 3.3330068588256836
    },
    {
      "auxiliary_loss_clip": 0.01160867,
      "auxiliary_loss_mlp": 0.01026281,
      "balance_loss_clip": 1.04821277,
      "balance_loss_mlp": 1.01840734,
      "epoch": 0.6405338784344375,
      "flos": 40479548901120.0,
      "grad_norm": 2.3478994769926786,
      "language_loss": 0.705194,
      "learning_rate": 1.2089797448924616e-06,
      "loss": 0.72706544,
      "num_input_tokens_seen": 114653615,
      "step": 5327,
      "time_per_iteration": 3.415560007095337
    },
    {
      "auxiliary_loss_clip": 0.01122666,
      "auxiliary_loss_mlp": 0.01026401,
      "balance_loss_clip": 1.0423646,
      "balance_loss_mlp": 1.01864362,
      "epoch": 0.6406541213250767,
      "flos": 20886041439360.0,
      "grad_norm": 2.0119672809734164,
      "language_loss": 0.65710902,
      "learning_rate": 1.2082643493515692e-06,
      "loss": 0.67859972,
      "num_input_tokens_seen": 114671935,
      "step": 5328,
      "time_per_iteration": 2.5431039333343506
    },
    {
      "auxiliary_loss_clip": 0.01157258,
      "auxiliary_loss_mlp": 0.01024726,
      "balance_loss_clip": 1.04804981,
      "balance_loss_mlp": 1.01707888,
      "epoch": 0.6407743642157158,
      "flos": 23295970679040.0,
      "grad_norm": 2.0107220421596996,
      "language_loss": 0.82066506,
      "learning_rate": 1.207549073916346e-06,
      "loss": 0.84248489,
      "num_input_tokens_seen": 114692870,
      "step": 5329,
      "time_per_iteration": 3.2290544509887695
    },
    {
      "auxiliary_loss_clip": 0.01134585,
      "auxiliary_loss_mlp": 0.01022905,
      "balance_loss_clip": 1.04556584,
      "balance_loss_mlp": 1.01579404,
      "epoch": 0.6408946071063548,
      "flos": 15012636122880.0,
      "grad_norm": 1.8958550359033812,
      "language_loss": 0.77789593,
      "learning_rate": 1.2068339186952976e-06,
      "loss": 0.7994709,
      "num_input_tokens_seen": 114710410,
      "step": 5330,
      "time_per_iteration": 2.5155410766601562
    },
    {
      "auxiliary_loss_clip": 0.011611,
      "auxiliary_loss_mlp": 0.01029178,
      "balance_loss_clip": 1.04946446,
      "balance_loss_mlp": 1.02127123,
      "epoch": 0.6410148499969939,
      "flos": 22528595496960.0,
      "grad_norm": 2.0442715750028824,
      "language_loss": 0.73243988,
      "learning_rate": 1.2061188837969136e-06,
      "loss": 0.75434262,
      "num_input_tokens_seen": 114730020,
      "step": 5331,
      "time_per_iteration": 2.4662926197052
    },
    {
      "auxiliary_loss_clip": 0.01122873,
      "auxiliary_loss_mlp": 0.01025999,
      "balance_loss_clip": 1.04150641,
      "balance_loss_mlp": 1.01804209,
      "epoch": 0.641135092887633,
      "flos": 12422004537600.0,
      "grad_norm": 2.480355896826845,
      "language_loss": 0.84041226,
      "learning_rate": 1.2054039693296631e-06,
      "loss": 0.86190093,
      "num_input_tokens_seen": 114748015,
      "step": 5332,
      "time_per_iteration": 2.5016486644744873
    },
    {
      "auxiliary_loss_clip": 0.01125519,
      "auxiliary_loss_mlp": 0.01029453,
      "balance_loss_clip": 1.0445025,
      "balance_loss_mlp": 1.02235389,
      "epoch": 0.6412553357782721,
      "flos": 22127329687680.0,
      "grad_norm": 1.8420310628330967,
      "language_loss": 0.81520379,
      "learning_rate": 1.2046891754019992e-06,
      "loss": 0.83675349,
      "num_input_tokens_seen": 114768625,
      "step": 5333,
      "time_per_iteration": 2.536783218383789
    },
    {
      "auxiliary_loss_clip": 0.01161449,
      "auxiliary_loss_mlp": 0.01025574,
      "balance_loss_clip": 1.04948533,
      "balance_loss_mlp": 1.0180223,
      "epoch": 0.6413755786689112,
      "flos": 15888605097600.0,
      "grad_norm": 1.9282467162316852,
      "language_loss": 0.82951975,
      "learning_rate": 1.2039745021223548e-06,
      "loss": 0.85139,
      "num_input_tokens_seen": 114786045,
      "step": 5334,
      "time_per_iteration": 2.4252753257751465
    },
    {
      "auxiliary_loss_clip": 0.01023782,
      "auxiliary_loss_mlp": 0.01004669,
      "balance_loss_clip": 1.0148356,
      "balance_loss_mlp": 1.0038408,
      "epoch": 0.6414958215595503,
      "flos": 68039159955840.0,
      "grad_norm": 0.7914621700395528,
      "language_loss": 0.57056165,
      "learning_rate": 1.2032599495991456e-06,
      "loss": 0.59084612,
      "num_input_tokens_seen": 114850785,
      "step": 5335,
      "time_per_iteration": 3.2243902683258057
    },
    {
      "auxiliary_loss_clip": 0.01157959,
      "auxiliary_loss_mlp": 0.01023383,
      "balance_loss_clip": 1.04883182,
      "balance_loss_mlp": 1.01552749,
      "epoch": 0.6416160644501894,
      "flos": 44091300320640.0,
      "grad_norm": 2.070151705950607,
      "language_loss": 0.70034564,
      "learning_rate": 1.2025455179407685e-06,
      "loss": 0.72215903,
      "num_input_tokens_seen": 114871945,
      "step": 5336,
      "time_per_iteration": 2.6606969833374023
    },
    {
      "auxiliary_loss_clip": 0.01155571,
      "auxiliary_loss_mlp": 0.0076235,
      "balance_loss_clip": 1.04840839,
      "balance_loss_mlp": 1.00055385,
      "epoch": 0.6417363073408284,
      "flos": 20959837931520.0,
      "grad_norm": 2.0420737735067864,
      "language_loss": 0.73590958,
      "learning_rate": 1.2018312072556022e-06,
      "loss": 0.75508875,
      "num_input_tokens_seen": 114890445,
      "step": 5337,
      "time_per_iteration": 2.4405548572540283
    },
    {
      "auxiliary_loss_clip": 0.01168738,
      "auxiliary_loss_mlp": 0.00762265,
      "balance_loss_clip": 1.04844379,
      "balance_loss_mlp": 1.00057149,
      "epoch": 0.6418565502314676,
      "flos": 22455122227200.0,
      "grad_norm": 1.669480341017247,
      "language_loss": 0.74216533,
      "learning_rate": 1.2011170176520077e-06,
      "loss": 0.76147532,
      "num_input_tokens_seen": 114911360,
      "step": 5338,
      "time_per_iteration": 2.431079387664795
    },
    {
      "auxiliary_loss_clip": 0.01084319,
      "auxiliary_loss_mlp": 0.01023944,
      "balance_loss_clip": 1.03932023,
      "balance_loss_mlp": 1.01627302,
      "epoch": 0.6419767931221066,
      "flos": 25045502417280.0,
      "grad_norm": 1.6519615999822461,
      "language_loss": 0.81233078,
      "learning_rate": 1.2004029492383256e-06,
      "loss": 0.83341342,
      "num_input_tokens_seen": 114932700,
      "step": 5339,
      "time_per_iteration": 2.6286911964416504
    },
    {
      "auxiliary_loss_clip": 0.01156269,
      "auxiliary_loss_mlp": 0.01026485,
      "balance_loss_clip": 1.04970431,
      "balance_loss_mlp": 1.01866508,
      "epoch": 0.6420970360127457,
      "flos": 19463691709440.0,
      "grad_norm": 1.9546883748002868,
      "language_loss": 0.73321879,
      "learning_rate": 1.1996890021228814e-06,
      "loss": 0.75504637,
      "num_input_tokens_seen": 114949475,
      "step": 5340,
      "time_per_iteration": 2.434872627258301
    },
    {
      "auxiliary_loss_clip": 0.01140382,
      "auxiliary_loss_mlp": 0.01024294,
      "balance_loss_clip": 1.04528308,
      "balance_loss_mlp": 1.01681995,
      "epoch": 0.6422172789033849,
      "flos": 40406147458560.0,
      "grad_norm": 1.5912915364760845,
      "language_loss": 0.69944096,
      "learning_rate": 1.1989751764139785e-06,
      "loss": 0.72108781,
      "num_input_tokens_seen": 114973125,
      "step": 5341,
      "time_per_iteration": 2.653404474258423
    },
    {
      "auxiliary_loss_clip": 0.01110681,
      "auxiliary_loss_mlp": 0.01023054,
      "balance_loss_clip": 1.03812551,
      "balance_loss_mlp": 1.01520967,
      "epoch": 0.6423375217940239,
      "flos": 27672870637440.0,
      "grad_norm": 1.62401096740937,
      "language_loss": 0.82967788,
      "learning_rate": 1.1982614722199044e-06,
      "loss": 0.85101521,
      "num_input_tokens_seen": 114994300,
      "step": 5342,
      "time_per_iteration": 2.591768503189087
    },
    {
      "auxiliary_loss_clip": 0.01147619,
      "auxiliary_loss_mlp": 0.01025007,
      "balance_loss_clip": 1.04608083,
      "balance_loss_mlp": 1.01809049,
      "epoch": 0.642457764684663,
      "flos": 18369242259840.0,
      "grad_norm": 2.039110277937119,
      "language_loss": 0.77809697,
      "learning_rate": 1.1975478896489276e-06,
      "loss": 0.79982316,
      "num_input_tokens_seen": 115012135,
      "step": 5343,
      "time_per_iteration": 2.470501184463501
    },
    {
      "auxiliary_loss_clip": 0.01168439,
      "auxiliary_loss_mlp": 0.01023038,
      "balance_loss_clip": 1.04824841,
      "balance_loss_mlp": 1.01601088,
      "epoch": 0.6425780075753021,
      "flos": 19750509809280.0,
      "grad_norm": 1.9409994790536031,
      "language_loss": 0.76477146,
      "learning_rate": 1.1968344288092981e-06,
      "loss": 0.78668618,
      "num_input_tokens_seen": 115028715,
      "step": 5344,
      "time_per_iteration": 2.40632700920105
    },
    {
      "auxiliary_loss_clip": 0.01157403,
      "auxiliary_loss_mlp": 0.00762356,
      "balance_loss_clip": 1.0491153,
      "balance_loss_mlp": 1.00062203,
      "epoch": 0.6426982504659412,
      "flos": 20558536208640.0,
      "grad_norm": 2.5298245518084315,
      "language_loss": 0.64473689,
      "learning_rate": 1.1961210898092468e-06,
      "loss": 0.66393447,
      "num_input_tokens_seen": 115047665,
      "step": 5345,
      "time_per_iteration": 2.4848885536193848
    },
    {
      "auxiliary_loss_clip": 0.01149248,
      "auxiliary_loss_mlp": 0.01027569,
      "balance_loss_clip": 1.04882598,
      "balance_loss_mlp": 1.0194633,
      "epoch": 0.6428184933565803,
      "flos": 17851984456320.0,
      "grad_norm": 4.701172722725984,
      "language_loss": 0.796799,
      "learning_rate": 1.1954078727569874e-06,
      "loss": 0.81856716,
      "num_input_tokens_seen": 115064965,
      "step": 5346,
      "time_per_iteration": 2.4554507732391357
    },
    {
      "auxiliary_loss_clip": 0.01132288,
      "auxiliary_loss_mlp": 0.0076226,
      "balance_loss_clip": 1.04337668,
      "balance_loss_mlp": 1.00056815,
      "epoch": 0.6429387362472194,
      "flos": 22456953820800.0,
      "grad_norm": 1.6386806484125886,
      "language_loss": 0.77965021,
      "learning_rate": 1.1946947777607141e-06,
      "loss": 0.79859573,
      "num_input_tokens_seen": 115086100,
      "step": 5347,
      "time_per_iteration": 2.5322957038879395
    },
    {
      "auxiliary_loss_clip": 0.01109016,
      "auxiliary_loss_mlp": 0.01024265,
      "balance_loss_clip": 1.04210496,
      "balance_loss_mlp": 1.01631975,
      "epoch": 0.6430589791378585,
      "flos": 24752579005440.0,
      "grad_norm": 1.8584480875619018,
      "language_loss": 0.79986107,
      "learning_rate": 1.1939818049286024e-06,
      "loss": 0.82119393,
      "num_input_tokens_seen": 115104260,
      "step": 5348,
      "time_per_iteration": 2.569805383682251
    },
    {
      "auxiliary_loss_clip": 0.0108661,
      "auxiliary_loss_mlp": 0.0102786,
      "balance_loss_clip": 1.03920484,
      "balance_loss_mlp": 1.0205524,
      "epoch": 0.6431792220284975,
      "flos": 24901249397760.0,
      "grad_norm": 1.519905985688268,
      "language_loss": 0.75588173,
      "learning_rate": 1.1932689543688101e-06,
      "loss": 0.77702647,
      "num_input_tokens_seen": 115125365,
      "step": 5349,
      "time_per_iteration": 3.385685920715332
    },
    {
      "auxiliary_loss_clip": 0.01143903,
      "auxiliary_loss_mlp": 0.01026624,
      "balance_loss_clip": 1.04815221,
      "balance_loss_mlp": 1.01902127,
      "epoch": 0.6432994649191367,
      "flos": 21032305620480.0,
      "grad_norm": 2.2725091381337816,
      "language_loss": 0.7260707,
      "learning_rate": 1.1925562261894756e-06,
      "loss": 0.74777591,
      "num_input_tokens_seen": 115144445,
      "step": 5350,
      "time_per_iteration": 2.4838509559631348
    },
    {
      "auxiliary_loss_clip": 0.01139656,
      "auxiliary_loss_mlp": 0.01027563,
      "balance_loss_clip": 1.04515898,
      "balance_loss_mlp": 1.02014184,
      "epoch": 0.6434197078097758,
      "flos": 30884433655680.0,
      "grad_norm": 3.0386379653774607,
      "language_loss": 0.77880096,
      "learning_rate": 1.1918436204987207e-06,
      "loss": 0.80047309,
      "num_input_tokens_seen": 115166305,
      "step": 5351,
      "time_per_iteration": 2.5686447620391846
    },
    {
      "auxiliary_loss_clip": 0.01155893,
      "auxiliary_loss_mlp": 0.01024305,
      "balance_loss_clip": 1.05091619,
      "balance_loss_mlp": 1.01693153,
      "epoch": 0.6435399507004148,
      "flos": 15012492468480.0,
      "grad_norm": 1.9502419607140733,
      "language_loss": 0.81479049,
      "learning_rate": 1.191131137404645e-06,
      "loss": 0.83659244,
      "num_input_tokens_seen": 115183045,
      "step": 5352,
      "time_per_iteration": 3.273716926574707
    },
    {
      "auxiliary_loss_clip": 0.01116893,
      "auxiliary_loss_mlp": 0.010283,
      "balance_loss_clip": 1.04344296,
      "balance_loss_mlp": 1.02114797,
      "epoch": 0.643660193591054,
      "flos": 19901981462400.0,
      "grad_norm": 2.0284984924712854,
      "language_loss": 0.76935744,
      "learning_rate": 1.190418777015333e-06,
      "loss": 0.79080939,
      "num_input_tokens_seen": 115201955,
      "step": 5353,
      "time_per_iteration": 2.5105373859405518
    },
    {
      "auxiliary_loss_clip": 0.01143691,
      "auxiliary_loss_mlp": 0.0102059,
      "balance_loss_clip": 1.04684758,
      "balance_loss_mlp": 1.01346767,
      "epoch": 0.643780436481693,
      "flos": 24133622820480.0,
      "grad_norm": 1.4747283207564812,
      "language_loss": 0.7382971,
      "learning_rate": 1.1897065394388487e-06,
      "loss": 0.75993991,
      "num_input_tokens_seen": 115222395,
      "step": 5354,
      "time_per_iteration": 3.409649610519409
    },
    {
      "auxiliary_loss_clip": 0.01143752,
      "auxiliary_loss_mlp": 0.01029314,
      "balance_loss_clip": 1.05037212,
      "balance_loss_mlp": 1.02169657,
      "epoch": 0.6439006793723321,
      "flos": 23148808657920.0,
      "grad_norm": 1.6125624525076288,
      "language_loss": 0.76621556,
      "learning_rate": 1.1889944247832385e-06,
      "loss": 0.78794616,
      "num_input_tokens_seen": 115242635,
      "step": 5355,
      "time_per_iteration": 3.2758243083953857
    },
    {
      "auxiliary_loss_clip": 0.01157667,
      "auxiliary_loss_mlp": 0.01028165,
      "balance_loss_clip": 1.04512191,
      "balance_loss_mlp": 1.02062535,
      "epoch": 0.6440209222629713,
      "flos": 23617909301760.0,
      "grad_norm": 1.9098594187834383,
      "language_loss": 0.70562905,
      "learning_rate": 1.1882824331565283e-06,
      "loss": 0.72748739,
      "num_input_tokens_seen": 115262095,
      "step": 5356,
      "time_per_iteration": 2.521042823791504
    },
    {
      "auxiliary_loss_clip": 0.01123631,
      "auxiliary_loss_mlp": 0.01027701,
      "balance_loss_clip": 1.04210019,
      "balance_loss_mlp": 1.02037013,
      "epoch": 0.6441411651536103,
      "flos": 16544872535040.0,
      "grad_norm": 2.014874191587914,
      "language_loss": 0.89210314,
      "learning_rate": 1.1875705646667287e-06,
      "loss": 0.91361648,
      "num_input_tokens_seen": 115279985,
      "step": 5357,
      "time_per_iteration": 2.487450122833252
    },
    {
      "auxiliary_loss_clip": 0.01153017,
      "auxiliary_loss_mlp": 0.01026817,
      "balance_loss_clip": 1.04431057,
      "balance_loss_mlp": 1.01861548,
      "epoch": 0.6442614080442494,
      "flos": 25410965345280.0,
      "grad_norm": 1.9162354228599496,
      "language_loss": 0.75268722,
      "learning_rate": 1.1868588194218282e-06,
      "loss": 0.77448559,
      "num_input_tokens_seen": 115300365,
      "step": 5358,
      "time_per_iteration": 2.4772651195526123
    },
    {
      "auxiliary_loss_clip": 0.01148005,
      "auxiliary_loss_mlp": 0.01030101,
      "balance_loss_clip": 1.04510999,
      "balance_loss_mlp": 1.02232909,
      "epoch": 0.6443816509348885,
      "flos": 28294017552000.0,
      "grad_norm": 1.6328647738082025,
      "language_loss": 0.73876476,
      "learning_rate": 1.1861471975297979e-06,
      "loss": 0.76054585,
      "num_input_tokens_seen": 115322060,
      "step": 5359,
      "time_per_iteration": 2.5494964122772217
    },
    {
      "auxiliary_loss_clip": 0.01126722,
      "auxiliary_loss_mlp": 0.01022555,
      "balance_loss_clip": 1.04684877,
      "balance_loss_mlp": 1.01459765,
      "epoch": 0.6445018938255276,
      "flos": 36690075964800.0,
      "grad_norm": 1.6199011382781163,
      "language_loss": 0.70958543,
      "learning_rate": 1.185435699098591e-06,
      "loss": 0.73107815,
      "num_input_tokens_seen": 115348255,
      "step": 5360,
      "time_per_iteration": 2.69227933883667
    },
    {
      "auxiliary_loss_clip": 0.01145888,
      "auxiliary_loss_mlp": 0.01023233,
      "balance_loss_clip": 1.04590225,
      "balance_loss_mlp": 1.0156008,
      "epoch": 0.6446221367161666,
      "flos": 14501411804160.0,
      "grad_norm": 7.536487055703115,
      "language_loss": 0.78403491,
      "learning_rate": 1.1847243242361403e-06,
      "loss": 0.80572605,
      "num_input_tokens_seen": 115366845,
      "step": 5361,
      "time_per_iteration": 2.450035810470581
    },
    {
      "auxiliary_loss_clip": 0.0114605,
      "auxiliary_loss_mlp": 0.01034156,
      "balance_loss_clip": 1.04786444,
      "balance_loss_mlp": 1.02666068,
      "epoch": 0.6447423796068057,
      "flos": 24609367480320.0,
      "grad_norm": 1.9725811768542592,
      "language_loss": 0.7779808,
      "learning_rate": 1.1840130730503624e-06,
      "loss": 0.79978287,
      "num_input_tokens_seen": 115388125,
      "step": 5362,
      "time_per_iteration": 2.5733940601348877
    },
    {
      "auxiliary_loss_clip": 0.01171787,
      "auxiliary_loss_mlp": 0.01026473,
      "balance_loss_clip": 1.0492866,
      "balance_loss_mlp": 1.01912951,
      "epoch": 0.6448626224974449,
      "flos": 25047298097280.0,
      "grad_norm": 1.7002227044297158,
      "language_loss": 0.74608403,
      "learning_rate": 1.1833019456491518e-06,
      "loss": 0.76806664,
      "num_input_tokens_seen": 115409655,
      "step": 5363,
      "time_per_iteration": 2.517425298690796
    },
    {
      "auxiliary_loss_clip": 0.01159534,
      "auxiliary_loss_mlp": 0.01025886,
      "balance_loss_clip": 1.04968572,
      "balance_loss_mlp": 1.01842976,
      "epoch": 0.6449828653880839,
      "flos": 22530355263360.0,
      "grad_norm": 2.1289093804855015,
      "language_loss": 0.79254144,
      "learning_rate": 1.1825909421403871e-06,
      "loss": 0.81439567,
      "num_input_tokens_seen": 115428750,
      "step": 5364,
      "time_per_iteration": 2.473006010055542
    },
    {
      "auxiliary_loss_clip": 0.01156489,
      "auxiliary_loss_mlp": 0.01027053,
      "balance_loss_clip": 1.04714763,
      "balance_loss_mlp": 1.02008557,
      "epoch": 0.645103108278723,
      "flos": 25695736369920.0,
      "grad_norm": 1.6775703708739282,
      "language_loss": 0.76189846,
      "learning_rate": 1.181880062631926e-06,
      "loss": 0.78373384,
      "num_input_tokens_seen": 115448085,
      "step": 5365,
      "time_per_iteration": 2.4765913486480713
    },
    {
      "auxiliary_loss_clip": 0.01138044,
      "auxiliary_loss_mlp": 0.01031328,
      "balance_loss_clip": 1.04679561,
      "balance_loss_mlp": 1.02295315,
      "epoch": 0.6452233511693621,
      "flos": 27450331925760.0,
      "grad_norm": 2.003137936354261,
      "language_loss": 0.84850895,
      "learning_rate": 1.1811693072316093e-06,
      "loss": 0.87020272,
      "num_input_tokens_seen": 115465765,
      "step": 5366,
      "time_per_iteration": 2.517852783203125
    },
    {
      "auxiliary_loss_clip": 0.01169989,
      "auxiliary_loss_mlp": 0.00762463,
      "balance_loss_clip": 1.0467217,
      "balance_loss_mlp": 1.00054502,
      "epoch": 0.6453435940600012,
      "flos": 19208618254080.0,
      "grad_norm": 2.174860207167397,
      "language_loss": 0.84323335,
      "learning_rate": 1.1804586760472574e-06,
      "loss": 0.86255783,
      "num_input_tokens_seen": 115482230,
      "step": 5367,
      "time_per_iteration": 2.4059057235717773
    },
    {
      "auxiliary_loss_clip": 0.01126319,
      "auxiliary_loss_mlp": 0.01023378,
      "balance_loss_clip": 1.0435679,
      "balance_loss_mlp": 1.01556945,
      "epoch": 0.6454638369506402,
      "flos": 25737680476800.0,
      "grad_norm": 2.2075010916483717,
      "language_loss": 0.80187517,
      "learning_rate": 1.1797481691866736e-06,
      "loss": 0.82337213,
      "num_input_tokens_seen": 115499455,
      "step": 5368,
      "time_per_iteration": 2.530470609664917
    },
    {
      "auxiliary_loss_clip": 0.01136434,
      "auxiliary_loss_mlp": 0.01032601,
      "balance_loss_clip": 1.04761469,
      "balance_loss_mlp": 1.02525771,
      "epoch": 0.6455840798412794,
      "flos": 20989176364800.0,
      "grad_norm": 2.000739103417135,
      "language_loss": 0.83167076,
      "learning_rate": 1.1790377867576393e-06,
      "loss": 0.85336113,
      "num_input_tokens_seen": 115517205,
      "step": 5369,
      "time_per_iteration": 2.4924209117889404
    },
    {
      "auxiliary_loss_clip": 0.01146224,
      "auxiliary_loss_mlp": 0.01026275,
      "balance_loss_clip": 1.04642403,
      "balance_loss_mlp": 1.01892543,
      "epoch": 0.6457043227319185,
      "flos": 26067556005120.0,
      "grad_norm": 2.1072539658293605,
      "language_loss": 0.76444286,
      "learning_rate": 1.1783275288679203e-06,
      "loss": 0.78616786,
      "num_input_tokens_seen": 115534370,
      "step": 5370,
      "time_per_iteration": 2.5288474559783936
    },
    {
      "auxiliary_loss_clip": 0.01064654,
      "auxiliary_loss_mlp": 0.01002359,
      "balance_loss_clip": 1.01657021,
      "balance_loss_mlp": 1.0013876,
      "epoch": 0.6458245656225575,
      "flos": 60370831088640.0,
      "grad_norm": 0.8475207745816336,
      "language_loss": 0.57112044,
      "learning_rate": 1.177617395625262e-06,
      "loss": 0.59179056,
      "num_input_tokens_seen": 115592345,
      "step": 5371,
      "time_per_iteration": 3.0110514163970947
    },
    {
      "auxiliary_loss_clip": 0.01156936,
      "auxiliary_loss_mlp": 0.01028401,
      "balance_loss_clip": 1.04955232,
      "balance_loss_mlp": 1.02099466,
      "epoch": 0.6459448085131967,
      "flos": 23076771932160.0,
      "grad_norm": 2.109194996184222,
      "language_loss": 0.75144202,
      "learning_rate": 1.1769073871373908e-06,
      "loss": 0.77329534,
      "num_input_tokens_seen": 115612550,
      "step": 5372,
      "time_per_iteration": 2.4809679985046387
    },
    {
      "auxiliary_loss_clip": 0.01124834,
      "auxiliary_loss_mlp": 0.01022894,
      "balance_loss_clip": 1.04220545,
      "balance_loss_mlp": 1.01571178,
      "epoch": 0.6460650514038357,
      "flos": 22598190097920.0,
      "grad_norm": 1.6962795854723085,
      "language_loss": 0.83681768,
      "learning_rate": 1.176197503512015e-06,
      "loss": 0.85829496,
      "num_input_tokens_seen": 115632265,
      "step": 5373,
      "time_per_iteration": 2.5287370681762695
    },
    {
      "auxiliary_loss_clip": 0.01140674,
      "auxiliary_loss_mlp": 0.01026543,
      "balance_loss_clip": 1.0466876,
      "balance_loss_mlp": 1.01932752,
      "epoch": 0.6461852942944748,
      "flos": 20266726118400.0,
      "grad_norm": 2.207626833863588,
      "language_loss": 0.82171977,
      "learning_rate": 1.1754877448568223e-06,
      "loss": 0.8433919,
      "num_input_tokens_seen": 115651720,
      "step": 5374,
      "time_per_iteration": 2.4975805282592773
    },
    {
      "auxiliary_loss_clip": 0.01141588,
      "auxiliary_loss_mlp": 0.01025345,
      "balance_loss_clip": 1.04442525,
      "balance_loss_mlp": 1.0178293,
      "epoch": 0.646305537185114,
      "flos": 23367109564800.0,
      "grad_norm": 2.7149867036172752,
      "language_loss": 0.89946866,
      "learning_rate": 1.1747781112794837e-06,
      "loss": 0.92113799,
      "num_input_tokens_seen": 115668215,
      "step": 5375,
      "time_per_iteration": 3.359842538833618
    },
    {
      "auxiliary_loss_clip": 0.01127006,
      "auxiliary_loss_mlp": 0.0102665,
      "balance_loss_clip": 1.04610848,
      "balance_loss_mlp": 1.01922929,
      "epoch": 0.646425780075753,
      "flos": 24277480790400.0,
      "grad_norm": 1.814659732872634,
      "language_loss": 0.82686841,
      "learning_rate": 1.1740686028876487e-06,
      "loss": 0.848405,
      "num_input_tokens_seen": 115687080,
      "step": 5376,
      "time_per_iteration": 2.5411336421966553
    },
    {
      "auxiliary_loss_clip": 0.01152356,
      "auxiliary_loss_mlp": 0.01023774,
      "balance_loss_clip": 1.0474751,
      "balance_loss_mlp": 1.01667261,
      "epoch": 0.6465460229663921,
      "flos": 20813968800000.0,
      "grad_norm": 3.232957618848814,
      "language_loss": 0.74605536,
      "learning_rate": 1.1733592197889507e-06,
      "loss": 0.76781666,
      "num_input_tokens_seen": 115703990,
      "step": 5377,
      "time_per_iteration": 2.4445273876190186
    },
    {
      "auxiliary_loss_clip": 0.01149306,
      "auxiliary_loss_mlp": 0.0102324,
      "balance_loss_clip": 1.04729044,
      "balance_loss_mlp": 1.01636434,
      "epoch": 0.6466662658570312,
      "flos": 22853299466880.0,
      "grad_norm": 1.902984090622666,
      "language_loss": 0.72670501,
      "learning_rate": 1.1726499620910014e-06,
      "loss": 0.74843043,
      "num_input_tokens_seen": 115724270,
      "step": 5378,
      "time_per_iteration": 2.4676947593688965
    },
    {
      "auxiliary_loss_clip": 0.01155216,
      "auxiliary_loss_mlp": 0.0102417,
      "balance_loss_clip": 1.04725885,
      "balance_loss_mlp": 1.01644814,
      "epoch": 0.6467865087476703,
      "flos": 15304553953920.0,
      "grad_norm": 1.9012010274018538,
      "language_loss": 0.77490056,
      "learning_rate": 1.1719408299013955e-06,
      "loss": 0.7966944,
      "num_input_tokens_seen": 115742995,
      "step": 5379,
      "time_per_iteration": 3.349086046218872
    },
    {
      "auxiliary_loss_clip": 0.01170034,
      "auxiliary_loss_mlp": 0.01028877,
      "balance_loss_clip": 1.05068803,
      "balance_loss_mlp": 1.0214268,
      "epoch": 0.6469067516383094,
      "flos": 19573650218880.0,
      "grad_norm": 2.210741846198485,
      "language_loss": 0.75815099,
      "learning_rate": 1.1712318233277067e-06,
      "loss": 0.7801401,
      "num_input_tokens_seen": 115762015,
      "step": 5380,
      "time_per_iteration": 3.2264480590820312
    },
    {
      "auxiliary_loss_clip": 0.01063531,
      "auxiliary_loss_mlp": 0.01001287,
      "balance_loss_clip": 1.01692963,
      "balance_loss_mlp": 1.00033379,
      "epoch": 0.6470269945289485,
      "flos": 65098002522240.0,
      "grad_norm": 0.7497049174802937,
      "language_loss": 0.57815784,
      "learning_rate": 1.1705229424774916e-06,
      "loss": 0.59880602,
      "num_input_tokens_seen": 115816285,
      "step": 5381,
      "time_per_iteration": 3.631687641143799
    },
    {
      "auxiliary_loss_clip": 0.01138542,
      "auxiliary_loss_mlp": 0.01028806,
      "balance_loss_clip": 1.04396987,
      "balance_loss_mlp": 1.02137303,
      "epoch": 0.6471472374195876,
      "flos": 30696943639680.0,
      "grad_norm": 1.6849529909741638,
      "language_loss": 0.64126766,
      "learning_rate": 1.1698141874582867e-06,
      "loss": 0.66294122,
      "num_input_tokens_seen": 115837330,
      "step": 5382,
      "time_per_iteration": 2.56235671043396
    },
    {
      "auxiliary_loss_clip": 0.01168809,
      "auxiliary_loss_mlp": 0.01027412,
      "balance_loss_clip": 1.04950714,
      "balance_loss_mlp": 1.02075112,
      "epoch": 0.6472674803102266,
      "flos": 20521835487360.0,
      "grad_norm": 2.069375863564844,
      "language_loss": 0.72142899,
      "learning_rate": 1.169105558377609e-06,
      "loss": 0.74339116,
      "num_input_tokens_seen": 115857420,
      "step": 5383,
      "time_per_iteration": 2.416995048522949
    },
    {
      "auxiliary_loss_clip": 0.01118094,
      "auxiliary_loss_mlp": 0.00762362,
      "balance_loss_clip": 1.04979205,
      "balance_loss_mlp": 1.00058675,
      "epoch": 0.6473877232008658,
      "flos": 24715447320960.0,
      "grad_norm": 1.6796468554461612,
      "language_loss": 0.78325164,
      "learning_rate": 1.1683970553429587e-06,
      "loss": 0.80205619,
      "num_input_tokens_seen": 115878875,
      "step": 5384,
      "time_per_iteration": 2.5850071907043457
    },
    {
      "auxiliary_loss_clip": 0.01130428,
      "auxiliary_loss_mlp": 0.01026937,
      "balance_loss_clip": 1.04514277,
      "balance_loss_mlp": 1.01936769,
      "epoch": 0.6475079660915048,
      "flos": 15885552441600.0,
      "grad_norm": 3.2797018940203744,
      "language_loss": 0.8169831,
      "learning_rate": 1.1676886784618128e-06,
      "loss": 0.83855677,
      "num_input_tokens_seen": 115895540,
      "step": 5385,
      "time_per_iteration": 2.4666833877563477
    },
    {
      "auxiliary_loss_clip": 0.01157426,
      "auxiliary_loss_mlp": 0.01025667,
      "balance_loss_clip": 1.04839325,
      "balance_loss_mlp": 1.01796043,
      "epoch": 0.6476282089821439,
      "flos": 17381590922880.0,
      "grad_norm": 2.3627084171020756,
      "language_loss": 0.83817947,
      "learning_rate": 1.1669804278416332e-06,
      "loss": 0.86001039,
      "num_input_tokens_seen": 115910265,
      "step": 5386,
      "time_per_iteration": 2.4138023853302
    },
    {
      "auxiliary_loss_clip": 0.01147047,
      "auxiliary_loss_mlp": 0.01026758,
      "balance_loss_clip": 1.04809904,
      "balance_loss_mlp": 1.01897407,
      "epoch": 0.6477484518727831,
      "flos": 20194078861440.0,
      "grad_norm": 1.9173957365869634,
      "language_loss": 0.71284294,
      "learning_rate": 1.1662723035898602e-06,
      "loss": 0.73458099,
      "num_input_tokens_seen": 115930025,
      "step": 5387,
      "time_per_iteration": 2.4842822551727295
    },
    {
      "auxiliary_loss_clip": 0.01155594,
      "auxiliary_loss_mlp": 0.01023982,
      "balance_loss_clip": 1.04855776,
      "balance_loss_mlp": 1.01643562,
      "epoch": 0.6478686947634221,
      "flos": 25410426641280.0,
      "grad_norm": 1.5864928088235273,
      "language_loss": 0.8176682,
      "learning_rate": 1.165564305813915e-06,
      "loss": 0.83946395,
      "num_input_tokens_seen": 115949025,
      "step": 5388,
      "time_per_iteration": 2.4786365032196045
    },
    {
      "auxiliary_loss_clip": 0.01156247,
      "auxiliary_loss_mlp": 0.01024858,
      "balance_loss_clip": 1.04869533,
      "balance_loss_mlp": 1.01760387,
      "epoch": 0.6479889376540612,
      "flos": 20083581648000.0,
      "grad_norm": 1.7424551227277802,
      "language_loss": 0.81379735,
      "learning_rate": 1.1648564346212019e-06,
      "loss": 0.83560842,
      "num_input_tokens_seen": 115968145,
      "step": 5389,
      "time_per_iteration": 2.4599738121032715
    },
    {
      "auxiliary_loss_clip": 0.01152876,
      "auxiliary_loss_mlp": 0.01027676,
      "balance_loss_clip": 1.04916835,
      "balance_loss_mlp": 1.02048755,
      "epoch": 0.6481091805447003,
      "flos": 26758082039040.0,
      "grad_norm": 2.0962209175181616,
      "language_loss": 0.76356912,
      "learning_rate": 1.164148690119104e-06,
      "loss": 0.78537464,
      "num_input_tokens_seen": 115989425,
      "step": 5390,
      "time_per_iteration": 2.4863195419311523
    },
    {
      "auxiliary_loss_clip": 0.01167658,
      "auxiliary_loss_mlp": 0.01025513,
      "balance_loss_clip": 1.04841769,
      "balance_loss_mlp": 1.01823568,
      "epoch": 0.6482294234353394,
      "flos": 23952094462080.0,
      "grad_norm": 2.0303993663255695,
      "language_loss": 0.74334908,
      "learning_rate": 1.163441072414985e-06,
      "loss": 0.76528084,
      "num_input_tokens_seen": 116009630,
      "step": 5391,
      "time_per_iteration": 2.442458391189575
    },
    {
      "auxiliary_loss_clip": 0.01156777,
      "auxiliary_loss_mlp": 0.01024132,
      "balance_loss_clip": 1.04952657,
      "balance_loss_mlp": 1.01689279,
      "epoch": 0.6483496663259785,
      "flos": 26209833776640.0,
      "grad_norm": 1.984044679603604,
      "language_loss": 0.69866383,
      "learning_rate": 1.16273358161619e-06,
      "loss": 0.72047293,
      "num_input_tokens_seen": 116029965,
      "step": 5392,
      "time_per_iteration": 2.493894577026367
    },
    {
      "auxiliary_loss_clip": 0.0115253,
      "auxiliary_loss_mlp": 0.01026518,
      "balance_loss_clip": 1.04963517,
      "balance_loss_mlp": 1.01916242,
      "epoch": 0.6484699092166175,
      "flos": 20922239370240.0,
      "grad_norm": 1.934568824094898,
      "language_loss": 0.83545816,
      "learning_rate": 1.1620262178300446e-06,
      "loss": 0.8572486,
      "num_input_tokens_seen": 116048580,
      "step": 5393,
      "time_per_iteration": 2.4673571586608887
    },
    {
      "auxiliary_loss_clip": 0.0112838,
      "auxiliary_loss_mlp": 0.01024479,
      "balance_loss_clip": 1.04354692,
      "balance_loss_mlp": 1.01718903,
      "epoch": 0.6485901521072567,
      "flos": 33072865678080.0,
      "grad_norm": 2.366200624095203,
      "language_loss": 0.75639564,
      "learning_rate": 1.1613189811638563e-06,
      "loss": 0.77792418,
      "num_input_tokens_seen": 116070305,
      "step": 5394,
      "time_per_iteration": 2.6324853897094727
    },
    {
      "auxiliary_loss_clip": 0.01158177,
      "auxiliary_loss_mlp": 0.01027217,
      "balance_loss_clip": 1.04957438,
      "balance_loss_mlp": 1.02020764,
      "epoch": 0.6487103949978957,
      "flos": 22274060745600.0,
      "grad_norm": 1.5125506975357599,
      "language_loss": 0.77831137,
      "learning_rate": 1.1606118717249117e-06,
      "loss": 0.80016541,
      "num_input_tokens_seen": 116090405,
      "step": 5395,
      "time_per_iteration": 2.4912984371185303
    },
    {
      "auxiliary_loss_clip": 0.01173863,
      "auxiliary_loss_mlp": 0.01024893,
      "balance_loss_clip": 1.04964268,
      "balance_loss_mlp": 1.0173229,
      "epoch": 0.6488306378885348,
      "flos": 22930400010240.0,
      "grad_norm": 1.8538992817476412,
      "language_loss": 0.67611563,
      "learning_rate": 1.1599048896204787e-06,
      "loss": 0.69810319,
      "num_input_tokens_seen": 116110285,
      "step": 5396,
      "time_per_iteration": 2.4224696159362793
    },
    {
      "auxiliary_loss_clip": 0.0113375,
      "auxiliary_loss_mlp": 0.01026279,
      "balance_loss_clip": 1.0474118,
      "balance_loss_mlp": 1.01890576,
      "epoch": 0.648950880779174,
      "flos": 20376110010240.0,
      "grad_norm": 1.86818198474614,
      "language_loss": 0.80856854,
      "learning_rate": 1.1591980349578061e-06,
      "loss": 0.83016884,
      "num_input_tokens_seen": 116128955,
      "step": 5397,
      "time_per_iteration": 2.5147154331207275
    },
    {
      "auxiliary_loss_clip": 0.01038576,
      "auxiliary_loss_mlp": 0.01001414,
      "balance_loss_clip": 1.01368618,
      "balance_loss_mlp": 1.00051403,
      "epoch": 0.649071123669813,
      "flos": 59930889310080.0,
      "grad_norm": 0.7370653548124485,
      "language_loss": 0.54303443,
      "learning_rate": 1.158491307844123e-06,
      "loss": 0.5634343,
      "num_input_tokens_seen": 116188875,
      "step": 5398,
      "time_per_iteration": 3.062973976135254
    },
    {
      "auxiliary_loss_clip": 0.01143135,
      "auxiliary_loss_mlp": 0.01024168,
      "balance_loss_clip": 1.04818606,
      "balance_loss_mlp": 1.0168488,
      "epoch": 0.6491913665604521,
      "flos": 20446566537600.0,
      "grad_norm": 1.7038105458207862,
      "language_loss": 0.83832473,
      "learning_rate": 1.1577847083866387e-06,
      "loss": 0.85999775,
      "num_input_tokens_seen": 116207910,
      "step": 5399,
      "time_per_iteration": 2.464677333831787
    },
    {
      "auxiliary_loss_clip": 0.01132292,
      "auxiliary_loss_mlp": 0.01026611,
      "balance_loss_clip": 1.04479098,
      "balance_loss_mlp": 1.0187726,
      "epoch": 0.6493116094510912,
      "flos": 16946820702720.0,
      "grad_norm": 1.855177009785709,
      "language_loss": 0.71696377,
      "learning_rate": 1.1570782366925453e-06,
      "loss": 0.73855281,
      "num_input_tokens_seen": 116226425,
      "step": 5400,
      "time_per_iteration": 2.452134847640991
    },
    {
      "auxiliary_loss_clip": 0.01142262,
      "auxiliary_loss_mlp": 0.01023291,
      "balance_loss_clip": 1.04279935,
      "balance_loss_mlp": 1.01570308,
      "epoch": 0.6494318523417303,
      "flos": 18802935072000.0,
      "grad_norm": 1.8018560466709699,
      "language_loss": 0.75824016,
      "learning_rate": 1.1563718928690132e-06,
      "loss": 0.77989572,
      "num_input_tokens_seen": 116243860,
      "step": 5401,
      "time_per_iteration": 2.458969831466675
    },
    {
      "auxiliary_loss_clip": 0.01125747,
      "auxiliary_loss_mlp": 0.01027501,
      "balance_loss_clip": 1.04444623,
      "balance_loss_mlp": 1.0197736,
      "epoch": 0.6495520952323693,
      "flos": 18982847318400.0,
      "grad_norm": 2.1628364599989522,
      "language_loss": 0.716272,
      "learning_rate": 1.1556656770231942e-06,
      "loss": 0.73780447,
      "num_input_tokens_seen": 116260055,
      "step": 5402,
      "time_per_iteration": 3.3266797065734863
    },
    {
      "auxiliary_loss_clip": 0.01158118,
      "auxiliary_loss_mlp": 0.01027369,
      "balance_loss_clip": 1.04811358,
      "balance_loss_mlp": 1.02047014,
      "epoch": 0.6496723381230085,
      "flos": 22745388032640.0,
      "grad_norm": 1.6160907475737125,
      "language_loss": 0.76413393,
      "learning_rate": 1.1549595892622207e-06,
      "loss": 0.78598881,
      "num_input_tokens_seen": 116278825,
      "step": 5403,
      "time_per_iteration": 2.4608218669891357
    },
    {
      "auxiliary_loss_clip": 0.01023799,
      "auxiliary_loss_mlp": 0.01003853,
      "balance_loss_clip": 1.01659656,
      "balance_loss_mlp": 1.00304842,
      "epoch": 0.6497925810136476,
      "flos": 62145283887360.0,
      "grad_norm": 0.8366120045218466,
      "language_loss": 0.59027404,
      "learning_rate": 1.1542536296932047e-06,
      "loss": 0.61055058,
      "num_input_tokens_seen": 116342360,
      "step": 5404,
      "time_per_iteration": 3.0789506435394287
    },
    {
      "auxiliary_loss_clip": 0.01133697,
      "auxiliary_loss_mlp": 0.01026499,
      "balance_loss_clip": 1.04367888,
      "balance_loss_mlp": 1.01855385,
      "epoch": 0.6499128239042866,
      "flos": 20156731695360.0,
      "grad_norm": 1.5949901860122606,
      "language_loss": 0.69981307,
      "learning_rate": 1.1535477984232414e-06,
      "loss": 0.72141504,
      "num_input_tokens_seen": 116362235,
      "step": 5405,
      "time_per_iteration": 2.505502223968506
    },
    {
      "auxiliary_loss_clip": 0.01116274,
      "auxiliary_loss_mlp": 0.01027173,
      "balance_loss_clip": 1.03943741,
      "balance_loss_mlp": 1.01974607,
      "epoch": 0.6500330667949258,
      "flos": 24462420940800.0,
      "grad_norm": 1.741601822144197,
      "language_loss": 0.77370203,
      "learning_rate": 1.152842095559404e-06,
      "loss": 0.79513645,
      "num_input_tokens_seen": 116382895,
      "step": 5406,
      "time_per_iteration": 4.2207725048065186
    },
    {
      "auxiliary_loss_clip": 0.01146971,
      "auxiliary_loss_mlp": 0.01025716,
      "balance_loss_clip": 1.04587376,
      "balance_loss_mlp": 1.0186286,
      "epoch": 0.6501533096855648,
      "flos": 25477399549440.0,
      "grad_norm": 1.6914510232454572,
      "language_loss": 0.76452059,
      "learning_rate": 1.1521365212087474e-06,
      "loss": 0.78624749,
      "num_input_tokens_seen": 116402880,
      "step": 5407,
      "time_per_iteration": 2.523043632507324
    },
    {
      "auxiliary_loss_clip": 0.01155994,
      "auxiliary_loss_mlp": 0.01024231,
      "balance_loss_clip": 1.04617023,
      "balance_loss_mlp": 1.0165782,
      "epoch": 0.6502735525762039,
      "flos": 44819245347840.0,
      "grad_norm": 1.5923291623948574,
      "language_loss": 0.70774889,
      "learning_rate": 1.1514310754783062e-06,
      "loss": 0.7295512,
      "num_input_tokens_seen": 116425830,
      "step": 5408,
      "time_per_iteration": 3.4084830284118652
    },
    {
      "auxiliary_loss_clip": 0.01145397,
      "auxiliary_loss_mlp": 0.01023532,
      "balance_loss_clip": 1.04765832,
      "balance_loss_mlp": 1.0156939,
      "epoch": 0.6503937954668431,
      "flos": 28658546726400.0,
      "grad_norm": 1.8707926288294439,
      "language_loss": 0.73297626,
      "learning_rate": 1.1507257584750964e-06,
      "loss": 0.75466549,
      "num_input_tokens_seen": 116446010,
      "step": 5409,
      "time_per_iteration": 2.589836597442627
    },
    {
      "auxiliary_loss_clip": 0.01171445,
      "auxiliary_loss_mlp": 0.01025641,
      "balance_loss_clip": 1.04959369,
      "balance_loss_mlp": 1.01791,
      "epoch": 0.6505140383574821,
      "flos": 20922562592640.0,
      "grad_norm": 1.9796961251777545,
      "language_loss": 0.77842784,
      "learning_rate": 1.150020570306113e-06,
      "loss": 0.80039865,
      "num_input_tokens_seen": 116465150,
      "step": 5410,
      "time_per_iteration": 2.410038471221924
    },
    {
      "auxiliary_loss_clip": 0.01135092,
      "auxiliary_loss_mlp": 0.0102527,
      "balance_loss_clip": 1.04149389,
      "balance_loss_mlp": 1.01749802,
      "epoch": 0.6506342812481212,
      "flos": 20595236929920.0,
      "grad_norm": 2.6853216891952445,
      "language_loss": 0.74958539,
      "learning_rate": 1.1493155110783338e-06,
      "loss": 0.77118903,
      "num_input_tokens_seen": 116483675,
      "step": 5411,
      "time_per_iteration": 2.4797093868255615
    },
    {
      "auxiliary_loss_clip": 0.01155657,
      "auxiliary_loss_mlp": 0.01024005,
      "balance_loss_clip": 1.04776502,
      "balance_loss_mlp": 1.01623249,
      "epoch": 0.6507545241387603,
      "flos": 30226478279040.0,
      "grad_norm": 2.6621805251068626,
      "language_loss": 0.70573229,
      "learning_rate": 1.1486105808987155e-06,
      "loss": 0.72752893,
      "num_input_tokens_seen": 116505165,
      "step": 5412,
      "time_per_iteration": 2.545302629470825
    },
    {
      "auxiliary_loss_clip": 0.01158894,
      "auxiliary_loss_mlp": 0.0102059,
      "balance_loss_clip": 1.04960871,
      "balance_loss_mlp": 1.01298499,
      "epoch": 0.6508747670293994,
      "flos": 17128241320320.0,
      "grad_norm": 1.825299864888889,
      "language_loss": 0.81571674,
      "learning_rate": 1.1479057798741947e-06,
      "loss": 0.8375116,
      "num_input_tokens_seen": 116523220,
      "step": 5413,
      "time_per_iteration": 2.437554121017456
    },
    {
      "auxiliary_loss_clip": 0.01054643,
      "auxiliary_loss_mlp": 0.01009326,
      "balance_loss_clip": 1.02318025,
      "balance_loss_mlp": 1.00810421,
      "epoch": 0.6509950099200384,
      "flos": 68559826573440.0,
      "grad_norm": 0.7909187214930529,
      "language_loss": 0.53340364,
      "learning_rate": 1.14720110811169e-06,
      "loss": 0.55404335,
      "num_input_tokens_seen": 116580450,
      "step": 5414,
      "time_per_iteration": 3.0831449031829834
    },
    {
      "auxiliary_loss_clip": 0.01161779,
      "auxiliary_loss_mlp": 0.01027469,
      "balance_loss_clip": 1.049137,
      "balance_loss_mlp": 1.01983345,
      "epoch": 0.6511152528106776,
      "flos": 22347462188160.0,
      "grad_norm": 2.1337486389609532,
      "language_loss": 0.76643193,
      "learning_rate": 1.146496565718098e-06,
      "loss": 0.78832436,
      "num_input_tokens_seen": 116601020,
      "step": 5415,
      "time_per_iteration": 2.465686559677124
    },
    {
      "auxiliary_loss_clip": 0.01147364,
      "auxiliary_loss_mlp": 0.01026344,
      "balance_loss_clip": 1.05141306,
      "balance_loss_mlp": 1.01833868,
      "epoch": 0.6512354957013167,
      "flos": 20522158709760.0,
      "grad_norm": 1.9173268352475563,
      "language_loss": 0.75976503,
      "learning_rate": 1.1457921528002996e-06,
      "loss": 0.78150207,
      "num_input_tokens_seen": 116619455,
      "step": 5416,
      "time_per_iteration": 2.4744672775268555
    },
    {
      "auxiliary_loss_clip": 0.01171561,
      "auxiliary_loss_mlp": 0.00762552,
      "balance_loss_clip": 1.05003142,
      "balance_loss_mlp": 1.00070596,
      "epoch": 0.6513557385919557,
      "flos": 32337342881280.0,
      "grad_norm": 2.2128519736824797,
      "language_loss": 0.72480196,
      "learning_rate": 1.1450878694651522e-06,
      "loss": 0.74414313,
      "num_input_tokens_seen": 116640020,
      "step": 5417,
      "time_per_iteration": 2.5107059478759766
    },
    {
      "auxiliary_loss_clip": 0.01114932,
      "auxiliary_loss_mlp": 0.01023737,
      "balance_loss_clip": 1.04161572,
      "balance_loss_mlp": 1.0161196,
      "epoch": 0.6514759814825949,
      "flos": 12093206417280.0,
      "grad_norm": 7.295446121662826,
      "language_loss": 0.62675995,
      "learning_rate": 1.1443837158194954e-06,
      "loss": 0.64814663,
      "num_input_tokens_seen": 116655165,
      "step": 5418,
      "time_per_iteration": 2.500206708908081
    },
    {
      "auxiliary_loss_clip": 0.01133119,
      "auxiliary_loss_mlp": 0.01028949,
      "balance_loss_clip": 1.05281436,
      "balance_loss_mlp": 1.02090216,
      "epoch": 0.651596224373234,
      "flos": 22526907557760.0,
      "grad_norm": 1.880254572797962,
      "language_loss": 0.74593651,
      "learning_rate": 1.1436796919701484e-06,
      "loss": 0.76755726,
      "num_input_tokens_seen": 116673880,
      "step": 5419,
      "time_per_iteration": 2.518559694290161
    },
    {
      "auxiliary_loss_clip": 0.01145284,
      "auxiliary_loss_mlp": 0.01023396,
      "balance_loss_clip": 1.05036473,
      "balance_loss_mlp": 1.01603508,
      "epoch": 0.651716467263873,
      "flos": 27818955250560.0,
      "grad_norm": 1.9222445285583931,
      "language_loss": 0.61628288,
      "learning_rate": 1.1429757980239115e-06,
      "loss": 0.63796973,
      "num_input_tokens_seen": 116694305,
      "step": 5420,
      "time_per_iteration": 2.5443460941314697
    },
    {
      "auxiliary_loss_clip": 0.01171686,
      "auxiliary_loss_mlp": 0.01032989,
      "balance_loss_clip": 1.049119,
      "balance_loss_mlp": 1.02485895,
      "epoch": 0.6518367101545122,
      "flos": 24316300414080.0,
      "grad_norm": 2.2601579156734855,
      "language_loss": 0.81650531,
      "learning_rate": 1.1422720340875636e-06,
      "loss": 0.838552,
      "num_input_tokens_seen": 116713055,
      "step": 5421,
      "time_per_iteration": 2.431621789932251
    },
    {
      "auxiliary_loss_clip": 0.01162808,
      "auxiliary_loss_mlp": 0.01026526,
      "balance_loss_clip": 1.04787552,
      "balance_loss_mlp": 1.01940084,
      "epoch": 0.6519569530451512,
      "flos": 20011939971840.0,
      "grad_norm": 1.9156298586850238,
      "language_loss": 0.79172671,
      "learning_rate": 1.1415684002678671e-06,
      "loss": 0.81362003,
      "num_input_tokens_seen": 116731815,
      "step": 5422,
      "time_per_iteration": 2.43428373336792
    },
    {
      "auxiliary_loss_clip": 0.01146,
      "auxiliary_loss_mlp": 0.01030684,
      "balance_loss_clip": 1.04561782,
      "balance_loss_mlp": 1.0227654,
      "epoch": 0.6520771959357903,
      "flos": 21576064682880.0,
      "grad_norm": 2.4552864604426814,
      "language_loss": 0.77722645,
      "learning_rate": 1.1408648966715617e-06,
      "loss": 0.79899329,
      "num_input_tokens_seen": 116749335,
      "step": 5423,
      "time_per_iteration": 2.48447585105896
    },
    {
      "auxiliary_loss_clip": 0.01142831,
      "auxiliary_loss_mlp": 0.01028542,
      "balance_loss_clip": 1.04315734,
      "balance_loss_mlp": 1.02101958,
      "epoch": 0.6521974388264293,
      "flos": 22711021695360.0,
      "grad_norm": 1.7686189723053143,
      "language_loss": 0.72463131,
      "learning_rate": 1.1401615234053683e-06,
      "loss": 0.74634498,
      "num_input_tokens_seen": 116768155,
      "step": 5424,
      "time_per_iteration": 2.475473403930664
    },
    {
      "auxiliary_loss_clip": 0.01144117,
      "auxiliary_loss_mlp": 0.01029075,
      "balance_loss_clip": 1.04617238,
      "balance_loss_mlp": 1.02150536,
      "epoch": 0.6523176817170685,
      "flos": 23002939526400.0,
      "grad_norm": 2.013088659868647,
      "language_loss": 0.75732207,
      "learning_rate": 1.1394582805759885e-06,
      "loss": 0.77905393,
      "num_input_tokens_seen": 116787435,
      "step": 5425,
      "time_per_iteration": 2.492194652557373
    },
    {
      "auxiliary_loss_clip": 0.01158779,
      "auxiliary_loss_mlp": 0.01029036,
      "balance_loss_clip": 1.04983258,
      "balance_loss_mlp": 1.02158237,
      "epoch": 0.6524379246077076,
      "flos": 21688249835520.0,
      "grad_norm": 1.6860771855964514,
      "language_loss": 0.75490665,
      "learning_rate": 1.1387551682901022e-06,
      "loss": 0.77678478,
      "num_input_tokens_seen": 116808040,
      "step": 5426,
      "time_per_iteration": 2.4582362174987793
    },
    {
      "auxiliary_loss_clip": 0.01126798,
      "auxiliary_loss_mlp": 0.01025834,
      "balance_loss_clip": 1.04531419,
      "balance_loss_mlp": 1.01857066,
      "epoch": 0.6525581674983466,
      "flos": 19390936711680.0,
      "grad_norm": 3.468677449449453,
      "language_loss": 0.70685619,
      "learning_rate": 1.138052186654373e-06,
      "loss": 0.72838247,
      "num_input_tokens_seen": 116825510,
      "step": 5427,
      "time_per_iteration": 2.504934310913086
    },
    {
      "auxiliary_loss_clip": 0.01146192,
      "auxiliary_loss_mlp": 0.01023189,
      "balance_loss_clip": 1.04741454,
      "balance_loss_mlp": 1.01493907,
      "epoch": 0.6526784103889858,
      "flos": 17165444832000.0,
      "grad_norm": 2.2006353470320477,
      "language_loss": 0.88082296,
      "learning_rate": 1.1373493357754417e-06,
      "loss": 0.90251678,
      "num_input_tokens_seen": 116844415,
      "step": 5428,
      "time_per_iteration": 2.456697940826416
    },
    {
      "auxiliary_loss_clip": 0.01169934,
      "auxiliary_loss_mlp": 0.01021271,
      "balance_loss_clip": 1.04770613,
      "balance_loss_mlp": 1.01436329,
      "epoch": 0.6527986532796248,
      "flos": 18989168112000.0,
      "grad_norm": 1.7605706431425618,
      "language_loss": 0.77212018,
      "learning_rate": 1.1366466157599303e-06,
      "loss": 0.79403222,
      "num_input_tokens_seen": 116863690,
      "step": 5429,
      "time_per_iteration": 3.1312551498413086
    },
    {
      "auxiliary_loss_clip": 0.01111853,
      "auxiliary_loss_mlp": 0.00762844,
      "balance_loss_clip": 1.04262304,
      "balance_loss_mlp": 1.00061369,
      "epoch": 0.6529188961702639,
      "flos": 14238581011200.0,
      "grad_norm": 2.2214907247739797,
      "language_loss": 0.76363444,
      "learning_rate": 1.1359440267144412e-06,
      "loss": 0.78238142,
      "num_input_tokens_seen": 116881145,
      "step": 5430,
      "time_per_iteration": 2.5234646797180176
    },
    {
      "auxiliary_loss_clip": 0.01159768,
      "auxiliary_loss_mlp": 0.01022424,
      "balance_loss_clip": 1.04864776,
      "balance_loss_mlp": 1.0154984,
      "epoch": 0.653039139060903,
      "flos": 36682929158400.0,
      "grad_norm": 1.9871571570191675,
      "language_loss": 0.74024165,
      "learning_rate": 1.1352415687455556e-06,
      "loss": 0.76206356,
      "num_input_tokens_seen": 116902405,
      "step": 5431,
      "time_per_iteration": 2.5878005027770996
    },
    {
      "auxiliary_loss_clip": 0.01158132,
      "auxiliary_loss_mlp": 0.01029207,
      "balance_loss_clip": 1.04976785,
      "balance_loss_mlp": 1.02188492,
      "epoch": 0.6531593819515421,
      "flos": 25376275785600.0,
      "grad_norm": 2.3825868980395932,
      "language_loss": 0.63941169,
      "learning_rate": 1.1345392419598362e-06,
      "loss": 0.66128504,
      "num_input_tokens_seen": 116921285,
      "step": 5432,
      "time_per_iteration": 3.373856782913208
    },
    {
      "auxiliary_loss_clip": 0.01151061,
      "auxiliary_loss_mlp": 0.01020679,
      "balance_loss_clip": 1.04601049,
      "balance_loss_mlp": 1.01269817,
      "epoch": 0.6532796248421812,
      "flos": 21178533888000.0,
      "grad_norm": 1.6477537244735991,
      "language_loss": 0.71806282,
      "learning_rate": 1.1338370464638263e-06,
      "loss": 0.73978025,
      "num_input_tokens_seen": 116940685,
      "step": 5433,
      "time_per_iteration": 3.294677495956421
    },
    {
      "auxiliary_loss_clip": 0.01169666,
      "auxiliary_loss_mlp": 0.01020712,
      "balance_loss_clip": 1.04741144,
      "balance_loss_mlp": 1.01335645,
      "epoch": 0.6533998677328203,
      "flos": 17675950878720.0,
      "grad_norm": 2.336478210524243,
      "language_loss": 0.64025503,
      "learning_rate": 1.1331349823640474e-06,
      "loss": 0.66215879,
      "num_input_tokens_seen": 116958115,
      "step": 5434,
      "time_per_iteration": 2.3974545001983643
    },
    {
      "auxiliary_loss_clip": 0.01157395,
      "auxiliary_loss_mlp": 0.00761877,
      "balance_loss_clip": 1.04690075,
      "balance_loss_mlp": 1.00064099,
      "epoch": 0.6535201106234594,
      "flos": 28400384701440.0,
      "grad_norm": 2.3189662681660304,
      "language_loss": 0.78374422,
      "learning_rate": 1.132433049767003e-06,
      "loss": 0.80293697,
      "num_input_tokens_seen": 116976030,
      "step": 5435,
      "time_per_iteration": 3.2574713230133057
    },
    {
      "auxiliary_loss_clip": 0.01141604,
      "auxiliary_loss_mlp": 0.01025959,
      "balance_loss_clip": 1.04689407,
      "balance_loss_mlp": 1.01935434,
      "epoch": 0.6536403535140984,
      "flos": 23586667447680.0,
      "grad_norm": 1.5918954219305315,
      "language_loss": 0.80939412,
      "learning_rate": 1.1317312487791748e-06,
      "loss": 0.83106983,
      "num_input_tokens_seen": 116997680,
      "step": 5436,
      "time_per_iteration": 2.504192352294922
    },
    {
      "auxiliary_loss_clip": 0.01151214,
      "auxiliary_loss_mlp": 0.01027757,
      "balance_loss_clip": 1.04646468,
      "balance_loss_mlp": 1.02033031,
      "epoch": 0.6537605964047376,
      "flos": 21579476474880.0,
      "grad_norm": 3.93432374522333,
      "language_loss": 0.73504812,
      "learning_rate": 1.1310295795070253e-06,
      "loss": 0.75683784,
      "num_input_tokens_seen": 117017620,
      "step": 5437,
      "time_per_iteration": 2.4524824619293213
    },
    {
      "auxiliary_loss_clip": 0.01119438,
      "auxiliary_loss_mlp": 0.01028518,
      "balance_loss_clip": 1.04329741,
      "balance_loss_mlp": 1.02091265,
      "epoch": 0.6538808392953767,
      "flos": 26833997433600.0,
      "grad_norm": 1.8605797673593854,
      "language_loss": 0.80882508,
      "learning_rate": 1.1303280420569982e-06,
      "loss": 0.83030462,
      "num_input_tokens_seen": 117039505,
      "step": 5438,
      "time_per_iteration": 2.5812337398529053
    },
    {
      "auxiliary_loss_clip": 0.01152315,
      "auxiliary_loss_mlp": 0.01023282,
      "balance_loss_clip": 1.04673338,
      "balance_loss_mlp": 1.01579022,
      "epoch": 0.6540010821860157,
      "flos": 30738241301760.0,
      "grad_norm": 1.6635113604138412,
      "language_loss": 0.77096975,
      "learning_rate": 1.1296266365355158e-06,
      "loss": 0.79272574,
      "num_input_tokens_seen": 117062890,
      "step": 5439,
      "time_per_iteration": 2.53440260887146
    },
    {
      "auxiliary_loss_clip": 0.01131821,
      "auxiliary_loss_mlp": 0.01022346,
      "balance_loss_clip": 1.0462091,
      "balance_loss_mlp": 1.01437128,
      "epoch": 0.6541213250766549,
      "flos": 26907147480960.0,
      "grad_norm": 1.8552373161488265,
      "language_loss": 0.73964345,
      "learning_rate": 1.1289253630489806e-06,
      "loss": 0.76118511,
      "num_input_tokens_seen": 117083940,
      "step": 5440,
      "time_per_iteration": 2.552569627761841
    },
    {
      "auxiliary_loss_clip": 0.01161062,
      "auxiliary_loss_mlp": 0.01029219,
      "balance_loss_clip": 1.04713321,
      "balance_loss_mlp": 1.02129793,
      "epoch": 0.6542415679672939,
      "flos": 19172384409600.0,
      "grad_norm": 2.275298212771805,
      "language_loss": 0.72369605,
      "learning_rate": 1.1282242217037753e-06,
      "loss": 0.74559891,
      "num_input_tokens_seen": 117101440,
      "step": 5441,
      "time_per_iteration": 2.428621530532837
    },
    {
      "auxiliary_loss_clip": 0.01110524,
      "auxiliary_loss_mlp": 0.01024917,
      "balance_loss_clip": 1.03989649,
      "balance_loss_mlp": 1.01717472,
      "epoch": 0.654361810857933,
      "flos": 48173517100800.0,
      "grad_norm": 3.4533814683365014,
      "language_loss": 0.61571622,
      "learning_rate": 1.127523212606262e-06,
      "loss": 0.63707066,
      "num_input_tokens_seen": 117124265,
      "step": 5442,
      "time_per_iteration": 2.7673609256744385
    },
    {
      "auxiliary_loss_clip": 0.01156203,
      "auxiliary_loss_mlp": 0.01026603,
      "balance_loss_clip": 1.04887748,
      "balance_loss_mlp": 1.01908684,
      "epoch": 0.6544820537485722,
      "flos": 26943165843840.0,
      "grad_norm": 1.5330886653022804,
      "language_loss": 0.72986728,
      "learning_rate": 1.1268223358627835e-06,
      "loss": 0.75169533,
      "num_input_tokens_seen": 117146755,
      "step": 5443,
      "time_per_iteration": 2.502192974090576
    },
    {
      "auxiliary_loss_clip": 0.01170755,
      "auxiliary_loss_mlp": 0.01025362,
      "balance_loss_clip": 1.04827738,
      "balance_loss_mlp": 1.01768541,
      "epoch": 0.6546022966392112,
      "flos": 20886328748160.0,
      "grad_norm": 1.7444612670351287,
      "language_loss": 0.72069454,
      "learning_rate": 1.126121591579663e-06,
      "loss": 0.74265575,
      "num_input_tokens_seen": 117165960,
      "step": 5444,
      "time_per_iteration": 2.4083096981048584
    },
    {
      "auxiliary_loss_clip": 0.01153383,
      "auxiliary_loss_mlp": 0.01026073,
      "balance_loss_clip": 1.04902899,
      "balance_loss_mlp": 1.01859808,
      "epoch": 0.6547225395298503,
      "flos": 24936693143040.0,
      "grad_norm": 1.52777908215882,
      "language_loss": 0.68855822,
      "learning_rate": 1.1254209798632018e-06,
      "loss": 0.71035278,
      "num_input_tokens_seen": 117186980,
      "step": 5445,
      "time_per_iteration": 2.481923818588257
    },
    {
      "auxiliary_loss_clip": 0.01088637,
      "auxiliary_loss_mlp": 0.01022439,
      "balance_loss_clip": 1.0399189,
      "balance_loss_mlp": 1.01489317,
      "epoch": 0.6548427824204894,
      "flos": 22565942663040.0,
      "grad_norm": 1.5885633187903272,
      "language_loss": 0.84326208,
      "learning_rate": 1.124720500819683e-06,
      "loss": 0.86437285,
      "num_input_tokens_seen": 117205135,
      "step": 5446,
      "time_per_iteration": 2.578939199447632
    },
    {
      "auxiliary_loss_clip": 0.01175364,
      "auxiliary_loss_mlp": 0.01029573,
      "balance_loss_clip": 1.05280805,
      "balance_loss_mlp": 1.02147877,
      "epoch": 0.6549630253111285,
      "flos": 18442500048000.0,
      "grad_norm": 1.8464784197272823,
      "language_loss": 0.82653767,
      "learning_rate": 1.1240201545553682e-06,
      "loss": 0.84858704,
      "num_input_tokens_seen": 117222935,
      "step": 5447,
      "time_per_iteration": 2.41939640045166
    },
    {
      "auxiliary_loss_clip": 0.01127309,
      "auxiliary_loss_mlp": 0.01024715,
      "balance_loss_clip": 1.04546309,
      "balance_loss_mlp": 1.01740122,
      "epoch": 0.6550832682017675,
      "flos": 25187313312000.0,
      "grad_norm": 1.7592498599445936,
      "language_loss": 0.73017538,
      "learning_rate": 1.1233199411764987e-06,
      "loss": 0.75169563,
      "num_input_tokens_seen": 117242370,
      "step": 5448,
      "time_per_iteration": 2.546719789505005
    },
    {
      "auxiliary_loss_clip": 0.01116508,
      "auxiliary_loss_mlp": 0.01026089,
      "balance_loss_clip": 1.0417757,
      "balance_loss_mlp": 1.01867139,
      "epoch": 0.6552035110924067,
      "flos": 22748153379840.0,
      "grad_norm": 1.7906602316990663,
      "language_loss": 0.68703872,
      "learning_rate": 1.1226198607892978e-06,
      "loss": 0.70846474,
      "num_input_tokens_seen": 117262930,
      "step": 5449,
      "time_per_iteration": 2.534228801727295
    },
    {
      "auxiliary_loss_clip": 0.01120039,
      "auxiliary_loss_mlp": 0.01025884,
      "balance_loss_clip": 1.04696107,
      "balance_loss_mlp": 1.01824594,
      "epoch": 0.6553237539830458,
      "flos": 21799178012160.0,
      "grad_norm": 1.7359555512365068,
      "language_loss": 0.79595459,
      "learning_rate": 1.1219199134999664e-06,
      "loss": 0.81741381,
      "num_input_tokens_seen": 117281430,
      "step": 5450,
      "time_per_iteration": 2.5475001335144043
    },
    {
      "auxiliary_loss_clip": 0.01145622,
      "auxiliary_loss_mlp": 0.0102828,
      "balance_loss_clip": 1.04718912,
      "balance_loss_mlp": 1.01979852,
      "epoch": 0.6554439968736848,
      "flos": 20887226588160.0,
      "grad_norm": 2.2396766975448985,
      "language_loss": 0.78496528,
      "learning_rate": 1.1212200994146863e-06,
      "loss": 0.80670428,
      "num_input_tokens_seen": 117299185,
      "step": 5451,
      "time_per_iteration": 2.4714713096618652
    },
    {
      "auxiliary_loss_clip": 0.01125096,
      "auxiliary_loss_mlp": 0.01029435,
      "balance_loss_clip": 1.04054236,
      "balance_loss_mlp": 1.02187109,
      "epoch": 0.655564239764324,
      "flos": 16139045698560.0,
      "grad_norm": 2.603008830866016,
      "language_loss": 0.75873518,
      "learning_rate": 1.120520418639618e-06,
      "loss": 0.78028047,
      "num_input_tokens_seen": 117317720,
      "step": 5452,
      "time_per_iteration": 2.5121986865997314
    },
    {
      "auxiliary_loss_clip": 0.01157432,
      "auxiliary_loss_mlp": 0.01028738,
      "balance_loss_clip": 1.05039299,
      "balance_loss_mlp": 1.02167535,
      "epoch": 0.655684482654963,
      "flos": 29570354496000.0,
      "grad_norm": 1.9105713941414475,
      "language_loss": 0.83696365,
      "learning_rate": 1.119820871280903e-06,
      "loss": 0.85882533,
      "num_input_tokens_seen": 117338795,
      "step": 5453,
      "time_per_iteration": 2.527688503265381
    },
    {
      "auxiliary_loss_clip": 0.01155859,
      "auxiliary_loss_mlp": 0.01024572,
      "balance_loss_clip": 1.04797316,
      "balance_loss_mlp": 1.01706171,
      "epoch": 0.6558047255456021,
      "flos": 29789409588480.0,
      "grad_norm": 4.816395132912529,
      "language_loss": 0.73438352,
      "learning_rate": 1.1191214574446614e-06,
      "loss": 0.75618786,
      "num_input_tokens_seen": 117359040,
      "step": 5454,
      "time_per_iteration": 2.5190632343292236
    },
    {
      "auxiliary_loss_clip": 0.01137683,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.04529715,
      "balance_loss_mlp": 1.01910353,
      "epoch": 0.6559249684362413,
      "flos": 29059166090880.0,
      "grad_norm": 2.3987333741800696,
      "language_loss": 0.79803085,
      "learning_rate": 1.118422177236995e-06,
      "loss": 0.81967425,
      "num_input_tokens_seen": 117380865,
      "step": 5455,
      "time_per_iteration": 2.5466363430023193
    },
    {
      "auxiliary_loss_clip": 0.0114459,
      "auxiliary_loss_mlp": 0.01031501,
      "balance_loss_clip": 1.04652262,
      "balance_loss_mlp": 1.02335334,
      "epoch": 0.6560452113268803,
      "flos": 20225464369920.0,
      "grad_norm": 1.943398448947495,
      "language_loss": 0.85624105,
      "learning_rate": 1.1177230307639835e-06,
      "loss": 0.87800193,
      "num_input_tokens_seen": 117398405,
      "step": 5456,
      "time_per_iteration": 3.2690234184265137
    },
    {
      "auxiliary_loss_clip": 0.01124994,
      "auxiliary_loss_mlp": 0.01025436,
      "balance_loss_clip": 1.04343605,
      "balance_loss_mlp": 1.01791942,
      "epoch": 0.6561654542175194,
      "flos": 25045538330880.0,
      "grad_norm": 1.663899665356716,
      "language_loss": 0.7848624,
      "learning_rate": 1.1170240181316865e-06,
      "loss": 0.80636674,
      "num_input_tokens_seen": 117419850,
      "step": 5457,
      "time_per_iteration": 2.5490152835845947
    },
    {
      "auxiliary_loss_clip": 0.01124641,
      "auxiliary_loss_mlp": 0.01028031,
      "balance_loss_clip": 1.04216897,
      "balance_loss_mlp": 1.0201993,
      "epoch": 0.6562856971081584,
      "flos": 22856711258880.0,
      "grad_norm": 2.726734622191998,
      "language_loss": 0.79474413,
      "learning_rate": 1.1163251394461442e-06,
      "loss": 0.81627089,
      "num_input_tokens_seen": 117438330,
      "step": 5458,
      "time_per_iteration": 2.506521701812744
    },
    {
      "auxiliary_loss_clip": 0.01154774,
      "auxiliary_loss_mlp": 0.01026847,
      "balance_loss_clip": 1.04819155,
      "balance_loss_mlp": 1.01946175,
      "epoch": 0.6564059399987976,
      "flos": 18872565586560.0,
      "grad_norm": 1.91155008434475,
      "language_loss": 0.82672256,
      "learning_rate": 1.1156263948133746e-06,
      "loss": 0.84853876,
      "num_input_tokens_seen": 117454985,
      "step": 5459,
      "time_per_iteration": 3.284507989883423
    },
    {
      "auxiliary_loss_clip": 0.01108252,
      "auxiliary_loss_mlp": 0.00762642,
      "balance_loss_clip": 1.04414308,
      "balance_loss_mlp": 1.00072885,
      "epoch": 0.6565261828894366,
      "flos": 25484187219840.0,
      "grad_norm": 2.238904793705515,
      "language_loss": 0.77235126,
      "learning_rate": 1.1149277843393787e-06,
      "loss": 0.79106021,
      "num_input_tokens_seen": 117476145,
      "step": 5460,
      "time_per_iteration": 3.3520314693450928
    },
    {
      "auxiliary_loss_clip": 0.01095583,
      "auxiliary_loss_mlp": 0.00762707,
      "balance_loss_clip": 1.0371139,
      "balance_loss_mlp": 1.00071025,
      "epoch": 0.6566464257800757,
      "flos": 19683500987520.0,
      "grad_norm": 2.47829113441089,
      "language_loss": 0.63581449,
      "learning_rate": 1.1142293081301342e-06,
      "loss": 0.65439737,
      "num_input_tokens_seen": 117494025,
      "step": 5461,
      "time_per_iteration": 2.589726686477661
    },
    {
      "auxiliary_loss_clip": 0.01138806,
      "auxiliary_loss_mlp": 0.01020857,
      "balance_loss_clip": 1.04576254,
      "balance_loss_mlp": 1.01400793,
      "epoch": 0.6567666686707149,
      "flos": 23514127931520.0,
      "grad_norm": 1.7557402581691772,
      "language_loss": 0.68078631,
      "learning_rate": 1.1135309662915995e-06,
      "loss": 0.70238292,
      "num_input_tokens_seen": 117514190,
      "step": 5462,
      "time_per_iteration": 3.3392550945281982
    },
    {
      "auxiliary_loss_clip": 0.01121384,
      "auxiliary_loss_mlp": 0.01025404,
      "balance_loss_clip": 1.04316413,
      "balance_loss_mlp": 1.01804876,
      "epoch": 0.6568869115613539,
      "flos": 32781342896640.0,
      "grad_norm": 2.0053983555877184,
      "language_loss": 0.60525131,
      "learning_rate": 1.112832758929712e-06,
      "loss": 0.62671912,
      "num_input_tokens_seen": 117536800,
      "step": 5463,
      "time_per_iteration": 2.649257183074951
    },
    {
      "auxiliary_loss_clip": 0.01155649,
      "auxiliary_loss_mlp": 0.01031179,
      "balance_loss_clip": 1.04928041,
      "balance_loss_mlp": 1.02368033,
      "epoch": 0.657007154451993,
      "flos": 18442428220800.0,
      "grad_norm": 1.8745029126214443,
      "language_loss": 0.74964637,
      "learning_rate": 1.11213468615039e-06,
      "loss": 0.77151465,
      "num_input_tokens_seen": 117556230,
      "step": 5464,
      "time_per_iteration": 2.450623035430908
    },
    {
      "auxiliary_loss_clip": 0.01100681,
      "auxiliary_loss_mlp": 0.01026109,
      "balance_loss_clip": 1.04264569,
      "balance_loss_mlp": 1.01894748,
      "epoch": 0.6571273973426321,
      "flos": 25156717902720.0,
      "grad_norm": 1.4900733433035265,
      "language_loss": 0.75171518,
      "learning_rate": 1.1114367480595292e-06,
      "loss": 0.77298307,
      "num_input_tokens_seen": 117577310,
      "step": 5465,
      "time_per_iteration": 2.615804433822632
    },
    {
      "auxiliary_loss_clip": 0.01101774,
      "auxiliary_loss_mlp": 0.01030364,
      "balance_loss_clip": 1.04615283,
      "balance_loss_mlp": 1.02239525,
      "epoch": 0.6572476402332712,
      "flos": 17529830352000.0,
      "grad_norm": 1.7735486969208059,
      "language_loss": 0.81244195,
      "learning_rate": 1.1107389447630086e-06,
      "loss": 0.8337633,
      "num_input_tokens_seen": 117596010,
      "step": 5466,
      "time_per_iteration": 2.557216167449951
    },
    {
      "auxiliary_loss_clip": 0.01137687,
      "auxiliary_loss_mlp": 0.00761851,
      "balance_loss_clip": 1.04379416,
      "balance_loss_mlp": 1.00078666,
      "epoch": 0.6573678831239103,
      "flos": 17014260487680.0,
      "grad_norm": 1.9922111222417023,
      "language_loss": 0.78405643,
      "learning_rate": 1.1100412763666818e-06,
      "loss": 0.80305183,
      "num_input_tokens_seen": 117611270,
      "step": 5467,
      "time_per_iteration": 2.4717767238616943
    },
    {
      "auxiliary_loss_clip": 0.01144888,
      "auxiliary_loss_mlp": 0.01023207,
      "balance_loss_clip": 1.04768229,
      "balance_loss_mlp": 1.01533329,
      "epoch": 0.6574881260145494,
      "flos": 23910078528000.0,
      "grad_norm": 1.4934403424944827,
      "language_loss": 0.80037481,
      "learning_rate": 1.1093437429763865e-06,
      "loss": 0.82205576,
      "num_input_tokens_seen": 117631535,
      "step": 5468,
      "time_per_iteration": 2.519512414932251
    },
    {
      "auxiliary_loss_clip": 0.01157556,
      "auxiliary_loss_mlp": 0.01019966,
      "balance_loss_clip": 1.0499711,
      "balance_loss_mlp": 1.01305163,
      "epoch": 0.6576083689051885,
      "flos": 11218458504960.0,
      "grad_norm": 2.173774977131815,
      "language_loss": 0.73243797,
      "learning_rate": 1.1086463446979361e-06,
      "loss": 0.75421321,
      "num_input_tokens_seen": 117649885,
      "step": 5469,
      "time_per_iteration": 2.473111629486084
    },
    {
      "auxiliary_loss_clip": 0.01161031,
      "auxiliary_loss_mlp": 0.01024323,
      "balance_loss_clip": 1.05172396,
      "balance_loss_mlp": 1.01703048,
      "epoch": 0.6577286117958275,
      "flos": 22455553190400.0,
      "grad_norm": 1.756797979653362,
      "language_loss": 0.77532804,
      "learning_rate": 1.1079490816371277e-06,
      "loss": 0.79718161,
      "num_input_tokens_seen": 117669650,
      "step": 5470,
      "time_per_iteration": 2.466010808944702
    },
    {
      "auxiliary_loss_clip": 0.01158691,
      "auxiliary_loss_mlp": 0.00762291,
      "balance_loss_clip": 1.04814124,
      "balance_loss_mlp": 1.00074315,
      "epoch": 0.6578488546864667,
      "flos": 21872184405120.0,
      "grad_norm": 2.3502520897543393,
      "language_loss": 0.75057077,
      "learning_rate": 1.1072519538997352e-06,
      "loss": 0.76978064,
      "num_input_tokens_seen": 117688790,
      "step": 5471,
      "time_per_iteration": 2.4627366065979004
    },
    {
      "auxiliary_loss_clip": 0.01144903,
      "auxiliary_loss_mlp": 0.01023788,
      "balance_loss_clip": 1.04452133,
      "balance_loss_mlp": 1.01660573,
      "epoch": 0.6579690975771058,
      "flos": 23543753673600.0,
      "grad_norm": 1.744828416464258,
      "language_loss": 0.82241738,
      "learning_rate": 1.1065549615915095e-06,
      "loss": 0.84410429,
      "num_input_tokens_seen": 117708620,
      "step": 5472,
      "time_per_iteration": 2.516986608505249
    },
    {
      "auxiliary_loss_clip": 0.01161106,
      "auxiliary_loss_mlp": 0.0102833,
      "balance_loss_clip": 1.0532831,
      "balance_loss_mlp": 1.0206176,
      "epoch": 0.6580893404677448,
      "flos": 32743995730560.0,
      "grad_norm": 2.1604951934973347,
      "language_loss": 0.78304708,
      "learning_rate": 1.105858104818187e-06,
      "loss": 0.80494142,
      "num_input_tokens_seen": 117729775,
      "step": 5473,
      "time_per_iteration": 2.5387651920318604
    },
    {
      "auxiliary_loss_clip": 0.01161532,
      "auxiliary_loss_mlp": 0.01025005,
      "balance_loss_clip": 1.04977083,
      "balance_loss_mlp": 1.01672578,
      "epoch": 0.658209583358384,
      "flos": 15888138220800.0,
      "grad_norm": 6.193159378752744,
      "language_loss": 0.74840915,
      "learning_rate": 1.105161383685478e-06,
      "loss": 0.77027452,
      "num_input_tokens_seen": 117746160,
      "step": 5474,
      "time_per_iteration": 2.4260239601135254
    },
    {
      "auxiliary_loss_clip": 0.01042022,
      "auxiliary_loss_mlp": 0.01001845,
      "balance_loss_clip": 1.01838005,
      "balance_loss_mlp": 1.00092697,
      "epoch": 0.658329826249023,
      "flos": 62695902447360.0,
      "grad_norm": 0.7297184140266282,
      "language_loss": 0.56311542,
      "learning_rate": 1.1044647982990771e-06,
      "loss": 0.58355409,
      "num_input_tokens_seen": 117808045,
      "step": 5475,
      "time_per_iteration": 3.0598082542419434
    },
    {
      "auxiliary_loss_clip": 0.01145704,
      "auxiliary_loss_mlp": 0.01026571,
      "balance_loss_clip": 1.04793811,
      "balance_loss_mlp": 1.01879227,
      "epoch": 0.6584500691396621,
      "flos": 31722624501120.0,
      "grad_norm": 2.511164672358488,
      "language_loss": 0.64645272,
      "learning_rate": 1.1037683487646536e-06,
      "loss": 0.66817558,
      "num_input_tokens_seen": 117828330,
      "step": 5476,
      "time_per_iteration": 2.5600411891937256
    },
    {
      "auxiliary_loss_clip": 0.01142695,
      "auxiliary_loss_mlp": 0.00762484,
      "balance_loss_clip": 1.04982102,
      "balance_loss_mlp": 1.00066936,
      "epoch": 0.6585703120303013,
      "flos": 18406086635520.0,
      "grad_norm": 2.104784881122586,
      "language_loss": 0.77099991,
      "learning_rate": 1.1030720351878583e-06,
      "loss": 0.79005164,
      "num_input_tokens_seen": 117846450,
      "step": 5477,
      "time_per_iteration": 2.44301700592041
    },
    {
      "auxiliary_loss_clip": 0.01054978,
      "auxiliary_loss_mlp": 0.01001845,
      "balance_loss_clip": 1.01835775,
      "balance_loss_mlp": 1.00098693,
      "epoch": 0.6586905549209403,
      "flos": 58309880434560.0,
      "grad_norm": 0.8136869411179377,
      "language_loss": 0.57718468,
      "learning_rate": 1.102375857674323e-06,
      "loss": 0.59775293,
      "num_input_tokens_seen": 117908365,
      "step": 5478,
      "time_per_iteration": 3.032545804977417
    },
    {
      "auxiliary_loss_clip": 0.01143578,
      "auxiliary_loss_mlp": 0.01024578,
      "balance_loss_clip": 1.04626369,
      "balance_loss_mlp": 1.01717186,
      "epoch": 0.6588107978115794,
      "flos": 22782627457920.0,
      "grad_norm": 1.7751988157288778,
      "language_loss": 0.9038468,
      "learning_rate": 1.1016798163296561e-06,
      "loss": 0.92552835,
      "num_input_tokens_seen": 117927565,
      "step": 5479,
      "time_per_iteration": 2.505192756652832
    },
    {
      "auxiliary_loss_clip": 0.01160699,
      "auxiliary_loss_mlp": 0.01021794,
      "balance_loss_clip": 1.0491817,
      "balance_loss_mlp": 1.01402748,
      "epoch": 0.6589310407022185,
      "flos": 20667525050880.0,
      "grad_norm": 1.8999939411733766,
      "language_loss": 0.66394627,
      "learning_rate": 1.1009839112594471e-06,
      "loss": 0.68577117,
      "num_input_tokens_seen": 117945590,
      "step": 5480,
      "time_per_iteration": 2.445777177810669
    },
    {
      "auxiliary_loss_clip": 0.01160084,
      "auxiliary_loss_mlp": 0.01029979,
      "balance_loss_clip": 1.04912531,
      "balance_loss_mlp": 1.02225971,
      "epoch": 0.6590512835928576,
      "flos": 25630595055360.0,
      "grad_norm": 2.1081587316969688,
      "language_loss": 0.71923256,
      "learning_rate": 1.1002881425692638e-06,
      "loss": 0.74113327,
      "num_input_tokens_seen": 117966020,
      "step": 5481,
      "time_per_iteration": 2.4979910850524902
    },
    {
      "auxiliary_loss_clip": 0.01151807,
      "auxiliary_loss_mlp": 0.0102442,
      "balance_loss_clip": 1.04621863,
      "balance_loss_mlp": 1.01660919,
      "epoch": 0.6591715264834966,
      "flos": 23726108044800.0,
      "grad_norm": 1.6499540453889339,
      "language_loss": 0.75149906,
      "learning_rate": 1.0995925103646532e-06,
      "loss": 0.77326131,
      "num_input_tokens_seen": 117984620,
      "step": 5482,
      "time_per_iteration": 3.2330353260040283
    },
    {
      "auxiliary_loss_clip": 0.01126224,
      "auxiliary_loss_mlp": 0.0102355,
      "balance_loss_clip": 1.0471046,
      "balance_loss_mlp": 1.01577759,
      "epoch": 0.6592917693741358,
      "flos": 35773850822400.0,
      "grad_norm": 1.7927925427408702,
      "language_loss": 0.66550684,
      "learning_rate": 1.0988970147511437e-06,
      "loss": 0.68700463,
      "num_input_tokens_seen": 118006500,
      "step": 5483,
      "time_per_iteration": 2.6555371284484863
    },
    {
      "auxiliary_loss_clip": 0.01144911,
      "auxiliary_loss_mlp": 0.01027313,
      "balance_loss_clip": 1.04939961,
      "balance_loss_mlp": 1.01949286,
      "epoch": 0.6594120122647749,
      "flos": 21396834794880.0,
      "grad_norm": 5.828404758710859,
      "language_loss": 0.80168962,
      "learning_rate": 1.0982016558342405e-06,
      "loss": 0.82341182,
      "num_input_tokens_seen": 118025470,
      "step": 5484,
      "time_per_iteration": 2.4823062419891357
    },
    {
      "auxiliary_loss_clip": 0.01173003,
      "auxiliary_loss_mlp": 0.01022017,
      "balance_loss_clip": 1.05158246,
      "balance_loss_mlp": 1.01477814,
      "epoch": 0.6595322551554139,
      "flos": 19351829779200.0,
      "grad_norm": 2.092667029019394,
      "language_loss": 0.71033549,
      "learning_rate": 1.0975064337194291e-06,
      "loss": 0.73228574,
      "num_input_tokens_seen": 118043515,
      "step": 5485,
      "time_per_iteration": 2.4008774757385254
    },
    {
      "auxiliary_loss_clip": 0.01127299,
      "auxiliary_loss_mlp": 0.01037349,
      "balance_loss_clip": 1.0471586,
      "balance_loss_mlp": 1.0298183,
      "epoch": 0.6596524980460531,
      "flos": 16837113588480.0,
      "grad_norm": 2.6069982098550684,
      "language_loss": 0.7033574,
      "learning_rate": 1.0968113485121743e-06,
      "loss": 0.72500384,
      "num_input_tokens_seen": 118063105,
      "step": 5486,
      "time_per_iteration": 3.3639791011810303
    },
    {
      "auxiliary_loss_clip": 0.01157822,
      "auxiliary_loss_mlp": 0.0076278,
      "balance_loss_clip": 1.04614782,
      "balance_loss_mlp": 1.00071931,
      "epoch": 0.6597727409366921,
      "flos": 21798567480960.0,
      "grad_norm": 1.727159501699965,
      "language_loss": 0.8022033,
      "learning_rate": 1.0961164003179185e-06,
      "loss": 0.82140934,
      "num_input_tokens_seen": 118081615,
      "step": 5487,
      "time_per_iteration": 3.3240482807159424
    },
    {
      "auxiliary_loss_clip": 0.01128229,
      "auxiliary_loss_mlp": 0.01026839,
      "balance_loss_clip": 1.04485822,
      "balance_loss_mlp": 1.01906073,
      "epoch": 0.6598929838273312,
      "flos": 23730704985600.0,
      "grad_norm": 1.7684741516972529,
      "language_loss": 0.84352738,
      "learning_rate": 1.0954215892420884e-06,
      "loss": 0.86507809,
      "num_input_tokens_seen": 118102315,
      "step": 5488,
      "time_per_iteration": 2.5419845581054688
    },
    {
      "auxiliary_loss_clip": 0.01134156,
      "auxiliary_loss_mlp": 0.01032148,
      "balance_loss_clip": 1.04790759,
      "balance_loss_mlp": 1.02376807,
      "epoch": 0.6600132267179702,
      "flos": 19974520978560.0,
      "grad_norm": 1.7655967211589083,
      "language_loss": 0.70552593,
      "learning_rate": 1.094726915390082e-06,
      "loss": 0.727189,
      "num_input_tokens_seen": 118120650,
      "step": 5489,
      "time_per_iteration": 3.272987127304077
    },
    {
      "auxiliary_loss_clip": 0.01159659,
      "auxiliary_loss_mlp": 0.01026516,
      "balance_loss_clip": 1.04994094,
      "balance_loss_mlp": 1.01894069,
      "epoch": 0.6601334696086094,
      "flos": 22342649765760.0,
      "grad_norm": 1.7514712984569456,
      "language_loss": 0.69936025,
      "learning_rate": 1.0940323788672836e-06,
      "loss": 0.72122204,
      "num_input_tokens_seen": 118139825,
      "step": 5490,
      "time_per_iteration": 2.4691755771636963
    },
    {
      "auxiliary_loss_clip": 0.01154527,
      "auxiliary_loss_mlp": 0.01023903,
      "balance_loss_clip": 1.04891181,
      "balance_loss_mlp": 1.0162468,
      "epoch": 0.6602537124992485,
      "flos": 25703098657920.0,
      "grad_norm": 1.5914370212075002,
      "language_loss": 0.73699164,
      "learning_rate": 1.0933379797790522e-06,
      "loss": 0.75877589,
      "num_input_tokens_seen": 118159240,
      "step": 5491,
      "time_per_iteration": 2.4835219383239746
    },
    {
      "auxiliary_loss_clip": 0.0117339,
      "auxiliary_loss_mlp": 0.01027736,
      "balance_loss_clip": 1.0516032,
      "balance_loss_mlp": 1.01980305,
      "epoch": 0.6603739553898875,
      "flos": 25848572739840.0,
      "grad_norm": 2.181969012948015,
      "language_loss": 0.7085371,
      "learning_rate": 1.0926437182307293e-06,
      "loss": 0.73054838,
      "num_input_tokens_seen": 118178050,
      "step": 5492,
      "time_per_iteration": 2.4552559852600098
    },
    {
      "auxiliary_loss_clip": 0.01147338,
      "auxiliary_loss_mlp": 0.01025934,
      "balance_loss_clip": 1.0467391,
      "balance_loss_mlp": 1.0182209,
      "epoch": 0.6604941982805267,
      "flos": 24570296461440.0,
      "grad_norm": 1.7549829047717922,
      "language_loss": 0.77862883,
      "learning_rate": 1.0919495943276338e-06,
      "loss": 0.80036157,
      "num_input_tokens_seen": 118199070,
      "step": 5493,
      "time_per_iteration": 2.5302581787109375
    },
    {
      "auxiliary_loss_clip": 0.01131722,
      "auxiliary_loss_mlp": 0.01025673,
      "balance_loss_clip": 1.04236674,
      "balance_loss_mlp": 1.01735234,
      "epoch": 0.6606144411711657,
      "flos": 13261775581440.0,
      "grad_norm": 3.1017645903864124,
      "language_loss": 0.7631433,
      "learning_rate": 1.0912556081750611e-06,
      "loss": 0.7847172,
      "num_input_tokens_seen": 118217000,
      "step": 5494,
      "time_per_iteration": 2.5310869216918945
    },
    {
      "auxiliary_loss_clip": 0.01142509,
      "auxiliary_loss_mlp": 0.01026035,
      "balance_loss_clip": 1.04902244,
      "balance_loss_mlp": 1.01875758,
      "epoch": 0.6607346840618048,
      "flos": 25155281358720.0,
      "grad_norm": 1.9716868875742357,
      "language_loss": 0.76578814,
      "learning_rate": 1.0905617598782909e-06,
      "loss": 0.78747356,
      "num_input_tokens_seen": 118237205,
      "step": 5495,
      "time_per_iteration": 2.502443313598633
    },
    {
      "auxiliary_loss_clip": 0.01108803,
      "auxiliary_loss_mlp": 0.01026434,
      "balance_loss_clip": 1.04394794,
      "balance_loss_mlp": 1.01927519,
      "epoch": 0.660854926952444,
      "flos": 17638029095040.0,
      "grad_norm": 3.1336762055953695,
      "language_loss": 0.80862933,
      "learning_rate": 1.0898680495425775e-06,
      "loss": 0.82998168,
      "num_input_tokens_seen": 118255495,
      "step": 5496,
      "time_per_iteration": 2.5191776752471924
    },
    {
      "auxiliary_loss_clip": 0.01147984,
      "auxiliary_loss_mlp": 0.01027016,
      "balance_loss_clip": 1.04920626,
      "balance_loss_mlp": 1.01960707,
      "epoch": 0.660975169843083,
      "flos": 16836000266880.0,
      "grad_norm": 1.627412866153545,
      "language_loss": 0.80131406,
      "learning_rate": 1.0891744772731594e-06,
      "loss": 0.82306409,
      "num_input_tokens_seen": 118273310,
      "step": 5497,
      "time_per_iteration": 2.4517807960510254
    },
    {
      "auxiliary_loss_clip": 0.01159899,
      "auxiliary_loss_mlp": 0.01030258,
      "balance_loss_clip": 1.0488503,
      "balance_loss_mlp": 1.02308798,
      "epoch": 0.6610954127337221,
      "flos": 26870410846080.0,
      "grad_norm": 1.5313410030880796,
      "language_loss": 0.66256523,
      "learning_rate": 1.088481043175248e-06,
      "loss": 0.68446678,
      "num_input_tokens_seen": 118293880,
      "step": 5498,
      "time_per_iteration": 2.50740385055542
    },
    {
      "auxiliary_loss_clip": 0.01130624,
      "auxiliary_loss_mlp": 0.01023767,
      "balance_loss_clip": 1.04173291,
      "balance_loss_mlp": 1.0161705,
      "epoch": 0.6612156556243612,
      "flos": 26465697331200.0,
      "grad_norm": 1.5800179556884986,
      "language_loss": 0.75429577,
      "learning_rate": 1.0877877473540368e-06,
      "loss": 0.77583969,
      "num_input_tokens_seen": 118314465,
      "step": 5499,
      "time_per_iteration": 2.5167829990386963
    },
    {
      "auxiliary_loss_clip": 0.01173366,
      "auxiliary_loss_mlp": 0.01021604,
      "balance_loss_clip": 1.04949057,
      "balance_loss_mlp": 1.01420736,
      "epoch": 0.6613358985150003,
      "flos": 19791915212160.0,
      "grad_norm": 1.695776739112023,
      "language_loss": 0.72491169,
      "learning_rate": 1.0870945899147002e-06,
      "loss": 0.74686146,
      "num_input_tokens_seen": 118331110,
      "step": 5500,
      "time_per_iteration": 2.3996667861938477
    },
    {
      "auxiliary_loss_clip": 0.01154867,
      "auxiliary_loss_mlp": 0.01027601,
      "balance_loss_clip": 1.04924369,
      "balance_loss_mlp": 1.02043664,
      "epoch": 0.6614561414056394,
      "flos": 26831627136000.0,
      "grad_norm": 1.782129191087351,
      "language_loss": 0.76254296,
      "learning_rate": 1.0864015709623879e-06,
      "loss": 0.78436768,
      "num_input_tokens_seen": 118351980,
      "step": 5501,
      "time_per_iteration": 2.4809956550598145
    },
    {
      "auxiliary_loss_clip": 0.01160495,
      "auxiliary_loss_mlp": 0.01025799,
      "balance_loss_clip": 1.04785395,
      "balance_loss_mlp": 1.01853371,
      "epoch": 0.6615763842962785,
      "flos": 22894597128960.0,
      "grad_norm": 2.2878075119431918,
      "language_loss": 0.80370712,
      "learning_rate": 1.0857086906022313e-06,
      "loss": 0.82557011,
      "num_input_tokens_seen": 118370315,
      "step": 5502,
      "time_per_iteration": 2.4603171348571777
    },
    {
      "auxiliary_loss_clip": 0.01092524,
      "auxiliary_loss_mlp": 0.01024673,
      "balance_loss_clip": 1.04411626,
      "balance_loss_mlp": 1.01672781,
      "epoch": 0.6616966271869176,
      "flos": 24790321221120.0,
      "grad_norm": 1.9567973686304647,
      "language_loss": 0.73015428,
      "learning_rate": 1.0850159489393388e-06,
      "loss": 0.75132626,
      "num_input_tokens_seen": 118389575,
      "step": 5503,
      "time_per_iteration": 2.591839551925659
    },
    {
      "auxiliary_loss_clip": 0.01120105,
      "auxiliary_loss_mlp": 0.01025006,
      "balance_loss_clip": 1.04040778,
      "balance_loss_mlp": 1.01731658,
      "epoch": 0.6618168700775566,
      "flos": 17202109639680.0,
      "grad_norm": 1.7726425937653567,
      "language_loss": 0.8221457,
      "learning_rate": 1.0843233460787992e-06,
      "loss": 0.84359682,
      "num_input_tokens_seen": 118406790,
      "step": 5504,
      "time_per_iteration": 2.5015242099761963
    },
    {
      "auxiliary_loss_clip": 0.01121837,
      "auxiliary_loss_mlp": 0.01025203,
      "balance_loss_clip": 1.04845929,
      "balance_loss_mlp": 1.01745415,
      "epoch": 0.6619371129681958,
      "flos": 25447091448960.0,
      "grad_norm": 1.817277690297216,
      "language_loss": 0.77930063,
      "learning_rate": 1.0836308821256805e-06,
      "loss": 0.800771,
      "num_input_tokens_seen": 118427590,
      "step": 5505,
      "time_per_iteration": 2.550518751144409
    },
    {
      "auxiliary_loss_clip": 0.01156609,
      "auxiliary_loss_mlp": 0.01026321,
      "balance_loss_clip": 1.04958797,
      "balance_loss_mlp": 1.01912665,
      "epoch": 0.6620573558588349,
      "flos": 18040444139520.0,
      "grad_norm": 2.0885954603458066,
      "language_loss": 0.78144073,
      "learning_rate": 1.0829385571850282e-06,
      "loss": 0.80327004,
      "num_input_tokens_seen": 118444570,
      "step": 5506,
      "time_per_iteration": 2.432097911834717
    },
    {
      "auxiliary_loss_clip": 0.01175803,
      "auxiliary_loss_mlp": 0.01022833,
      "balance_loss_clip": 1.05067146,
      "balance_loss_mlp": 1.014727,
      "epoch": 0.6621775987494739,
      "flos": 17785586165760.0,
      "grad_norm": 2.6374579240350045,
      "language_loss": 0.83635974,
      "learning_rate": 1.0822463713618679e-06,
      "loss": 0.85834622,
      "num_input_tokens_seen": 118461425,
      "step": 5507,
      "time_per_iteration": 2.390392541885376
    },
    {
      "auxiliary_loss_clip": 0.01129776,
      "auxiliary_loss_mlp": 0.01027962,
      "balance_loss_clip": 1.0454073,
      "balance_loss_mlp": 1.02062166,
      "epoch": 0.6622978416401131,
      "flos": 17492590926720.0,
      "grad_norm": 2.0372885234616733,
      "language_loss": 0.84817189,
      "learning_rate": 1.0815543247612034e-06,
      "loss": 0.86974925,
      "num_input_tokens_seen": 118478495,
      "step": 5508,
      "time_per_iteration": 2.498204231262207
    },
    {
      "auxiliary_loss_clip": 0.01139714,
      "auxiliary_loss_mlp": 0.01020393,
      "balance_loss_clip": 1.04204845,
      "balance_loss_mlp": 1.01306212,
      "epoch": 0.6624180845307521,
      "flos": 21648352803840.0,
      "grad_norm": 1.6316674313800148,
      "language_loss": 0.82571679,
      "learning_rate": 1.0808624174880168e-06,
      "loss": 0.84731787,
      "num_input_tokens_seen": 118499145,
      "step": 5509,
      "time_per_iteration": 3.2900233268737793
    },
    {
      "auxiliary_loss_clip": 0.01170201,
      "auxiliary_loss_mlp": 0.01022683,
      "balance_loss_clip": 1.05057073,
      "balance_loss_mlp": 1.01573944,
      "epoch": 0.6625383274213912,
      "flos": 23805902108160.0,
      "grad_norm": 1.681469688867829,
      "language_loss": 0.79641908,
      "learning_rate": 1.080170649647272e-06,
      "loss": 0.81834799,
      "num_input_tokens_seen": 118518950,
      "step": 5510,
      "time_per_iteration": 2.4409148693084717
    },
    {
      "auxiliary_loss_clip": 0.01169568,
      "auxiliary_loss_mlp": 0.01022581,
      "balance_loss_clip": 1.04933012,
      "balance_loss_mlp": 1.01496696,
      "epoch": 0.6626585703120303,
      "flos": 33262941473280.0,
      "grad_norm": 1.5850588106199062,
      "language_loss": 0.67071486,
      "learning_rate": 1.0794790213439068e-06,
      "loss": 0.69263631,
      "num_input_tokens_seen": 118545850,
      "step": 5511,
      "time_per_iteration": 2.583749294281006
    },
    {
      "auxiliary_loss_clip": 0.01117278,
      "auxiliary_loss_mlp": 0.01029025,
      "balance_loss_clip": 1.04536104,
      "balance_loss_mlp": 1.0209192,
      "epoch": 0.6627788132026694,
      "flos": 22085780630400.0,
      "grad_norm": 2.1677491242586373,
      "language_loss": 0.78728771,
      "learning_rate": 1.078787532682843e-06,
      "loss": 0.80875069,
      "num_input_tokens_seen": 118563325,
      "step": 5512,
      "time_per_iteration": 3.5486109256744385
    },
    {
      "auxiliary_loss_clip": 0.01153587,
      "auxiliary_loss_mlp": 0.01028332,
      "balance_loss_clip": 1.04797149,
      "balance_loss_mlp": 1.02104831,
      "epoch": 0.6628990560933085,
      "flos": 36173608260480.0,
      "grad_norm": 2.457118403339589,
      "language_loss": 0.7613343,
      "learning_rate": 1.0780961837689773e-06,
      "loss": 0.78315347,
      "num_input_tokens_seen": 118582835,
      "step": 5513,
      "time_per_iteration": 3.4394383430480957
    },
    {
      "auxiliary_loss_clip": 0.01137165,
      "auxiliary_loss_mlp": 0.010237,
      "balance_loss_clip": 1.0473299,
      "balance_loss_mlp": 1.0163449,
      "epoch": 0.6630192989839476,
      "flos": 18513567106560.0,
      "grad_norm": 1.8659128961843627,
      "language_loss": 0.69941652,
      "learning_rate": 1.0774049747071883e-06,
      "loss": 0.72102517,
      "num_input_tokens_seen": 118600715,
      "step": 5514,
      "time_per_iteration": 2.46176815032959
    },
    {
      "auxiliary_loss_clip": 0.0111058,
      "auxiliary_loss_mlp": 0.01029971,
      "balance_loss_clip": 1.04435849,
      "balance_loss_mlp": 1.02205515,
      "epoch": 0.6631395418745867,
      "flos": 35809510049280.0,
      "grad_norm": 1.786747177903583,
      "language_loss": 0.68000865,
      "learning_rate": 1.076713905602332e-06,
      "loss": 0.70141411,
      "num_input_tokens_seen": 118621290,
      "step": 5515,
      "time_per_iteration": 2.6638760566711426
    },
    {
      "auxiliary_loss_clip": 0.01160182,
      "auxiliary_loss_mlp": 0.01022697,
      "balance_loss_clip": 1.04950476,
      "balance_loss_mlp": 1.01544881,
      "epoch": 0.6632597847652257,
      "flos": 20047742853120.0,
      "grad_norm": 1.825025368132173,
      "language_loss": 0.81068814,
      "learning_rate": 1.07602297655924e-06,
      "loss": 0.83251691,
      "num_input_tokens_seen": 118639610,
      "step": 5516,
      "time_per_iteration": 3.2048749923706055
    },
    {
      "auxiliary_loss_clip": 0.01171914,
      "auxiliary_loss_mlp": 0.0102587,
      "balance_loss_clip": 1.05165792,
      "balance_loss_mlp": 1.01868165,
      "epoch": 0.6633800276558649,
      "flos": 21214480423680.0,
      "grad_norm": 1.8426631964134386,
      "language_loss": 0.81222248,
      "learning_rate": 1.0753321876827292e-06,
      "loss": 0.83420026,
      "num_input_tokens_seen": 118658895,
      "step": 5517,
      "time_per_iteration": 2.432898998260498
    },
    {
      "auxiliary_loss_clip": 0.01169314,
      "auxiliary_loss_mlp": 0.01023304,
      "balance_loss_clip": 1.0471499,
      "balance_loss_mlp": 1.01585925,
      "epoch": 0.663500270546504,
      "flos": 23987753688960.0,
      "grad_norm": 1.7610862714714586,
      "language_loss": 0.74105859,
      "learning_rate": 1.0746415390775893e-06,
      "loss": 0.76298475,
      "num_input_tokens_seen": 118677025,
      "step": 5518,
      "time_per_iteration": 2.430189609527588
    },
    {
      "auxiliary_loss_clip": 0.01170991,
      "auxiliary_loss_mlp": 0.01026131,
      "balance_loss_clip": 1.05153191,
      "balance_loss_mlp": 1.0189929,
      "epoch": 0.663620513437143,
      "flos": 17932389050880.0,
      "grad_norm": 2.193754933738702,
      "language_loss": 0.76538527,
      "learning_rate": 1.0739510308485939e-06,
      "loss": 0.7873565,
      "num_input_tokens_seen": 118694240,
      "step": 5519,
      "time_per_iteration": 2.3850111961364746
    },
    {
      "auxiliary_loss_clip": 0.01046035,
      "auxiliary_loss_mlp": 0.01002203,
      "balance_loss_clip": 1.01747537,
      "balance_loss_mlp": 1.00130308,
      "epoch": 0.6637407563277821,
      "flos": 57840241086720.0,
      "grad_norm": 0.805975172859145,
      "language_loss": 0.62494946,
      "learning_rate": 1.07326066310049e-06,
      "loss": 0.64543176,
      "num_input_tokens_seen": 118758365,
      "step": 5520,
      "time_per_iteration": 3.115778684616089
    },
    {
      "auxiliary_loss_clip": 0.01124287,
      "auxiliary_loss_mlp": 0.01025816,
      "balance_loss_clip": 1.04352069,
      "balance_loss_mlp": 1.01741815,
      "epoch": 0.6638609992184212,
      "flos": 27306007079040.0,
      "grad_norm": 2.9635734425357914,
      "language_loss": 0.7924149,
      "learning_rate": 1.0725704359380059e-06,
      "loss": 0.81391597,
      "num_input_tokens_seen": 118778220,
      "step": 5521,
      "time_per_iteration": 2.5515520572662354
    },
    {
      "auxiliary_loss_clip": 0.01160887,
      "auxiliary_loss_mlp": 0.01022939,
      "balance_loss_clip": 1.04841661,
      "balance_loss_mlp": 1.01603103,
      "epoch": 0.6639812421090603,
      "flos": 18624854419200.0,
      "grad_norm": 1.9282956814426322,
      "language_loss": 0.72169173,
      "learning_rate": 1.0718803494658497e-06,
      "loss": 0.74352998,
      "num_input_tokens_seen": 118797110,
      "step": 5522,
      "time_per_iteration": 2.4254212379455566
    },
    {
      "auxiliary_loss_clip": 0.01078893,
      "auxiliary_loss_mlp": 0.01031956,
      "balance_loss_clip": 1.04172552,
      "balance_loss_mlp": 1.02364695,
      "epoch": 0.6641014849996993,
      "flos": 15924479806080.0,
      "grad_norm": 2.014979946928305,
      "language_loss": 0.83849549,
      "learning_rate": 1.071190403788707e-06,
      "loss": 0.859604,
      "num_input_tokens_seen": 118812415,
      "step": 5523,
      "time_per_iteration": 2.5648577213287354
    },
    {
      "auxiliary_loss_clip": 0.01135727,
      "auxiliary_loss_mlp": 0.01026546,
      "balance_loss_clip": 1.04852486,
      "balance_loss_mlp": 1.01884484,
      "epoch": 0.6642217278903385,
      "flos": 26505486622080.0,
      "grad_norm": 1.732655601491865,
      "language_loss": 0.75329524,
      "learning_rate": 1.0705005990112415e-06,
      "loss": 0.77491796,
      "num_input_tokens_seen": 118832195,
      "step": 5524,
      "time_per_iteration": 2.5725104808807373
    },
    {
      "auxiliary_loss_clip": 0.01104152,
      "auxiliary_loss_mlp": 0.01030067,
      "balance_loss_clip": 1.04475117,
      "balance_loss_mlp": 1.02262878,
      "epoch": 0.6643419707809776,
      "flos": 15377308951680.0,
      "grad_norm": 3.315039872400583,
      "language_loss": 0.7430492,
      "learning_rate": 1.0698109352380957e-06,
      "loss": 0.76439142,
      "num_input_tokens_seen": 118849795,
      "step": 5525,
      "time_per_iteration": 2.5100502967834473
    },
    {
      "auxiliary_loss_clip": 0.01169231,
      "auxiliary_loss_mlp": 0.01024484,
      "balance_loss_clip": 1.04901254,
      "balance_loss_mlp": 1.01753092,
      "epoch": 0.6644622136716166,
      "flos": 25117610970240.0,
      "grad_norm": 2.391484785397507,
      "language_loss": 0.7815215,
      "learning_rate": 1.0691214125738909e-06,
      "loss": 0.80345863,
      "num_input_tokens_seen": 118870000,
      "step": 5526,
      "time_per_iteration": 2.450190305709839
    },
    {
      "auxiliary_loss_clip": 0.01070438,
      "auxiliary_loss_mlp": 0.01001672,
      "balance_loss_clip": 1.01508987,
      "balance_loss_mlp": 1.00081372,
      "epoch": 0.6645824565622558,
      "flos": 66201717680640.0,
      "grad_norm": 0.7918490834424107,
      "language_loss": 0.57526654,
      "learning_rate": 1.0684320311232287e-06,
      "loss": 0.59598768,
      "num_input_tokens_seen": 118932905,
      "step": 5527,
      "time_per_iteration": 3.0717484951019287
    },
    {
      "auxiliary_loss_clip": 0.01141057,
      "auxiliary_loss_mlp": 0.01025754,
      "balance_loss_clip": 1.04597092,
      "balance_loss_mlp": 1.01792765,
      "epoch": 0.6647026994528948,
      "flos": 25082131311360.0,
      "grad_norm": 1.8932580601366746,
      "language_loss": 0.81569481,
      "learning_rate": 1.0677427909906865e-06,
      "loss": 0.83736289,
      "num_input_tokens_seen": 118953355,
      "step": 5528,
      "time_per_iteration": 2.6524579524993896
    },
    {
      "auxiliary_loss_clip": 0.01175936,
      "auxiliary_loss_mlp": 0.01030558,
      "balance_loss_clip": 1.05256224,
      "balance_loss_mlp": 1.02252936,
      "epoch": 0.6648229423435339,
      "flos": 18222187979520.0,
      "grad_norm": 1.9618559863124314,
      "language_loss": 0.71879601,
      "learning_rate": 1.0670536922808216e-06,
      "loss": 0.74086094,
      "num_input_tokens_seen": 118973480,
      "step": 5529,
      "time_per_iteration": 2.5966434478759766
    },
    {
      "auxiliary_loss_clip": 0.01142242,
      "auxiliary_loss_mlp": 0.01025324,
      "balance_loss_clip": 1.04695535,
      "balance_loss_mlp": 1.01833844,
      "epoch": 0.6649431852341731,
      "flos": 18296882311680.0,
      "grad_norm": 4.6810062473447145,
      "language_loss": 0.71760809,
      "learning_rate": 1.06636473509817e-06,
      "loss": 0.73928374,
      "num_input_tokens_seen": 118989860,
      "step": 5530,
      "time_per_iteration": 2.50557804107666
    },
    {
      "auxiliary_loss_clip": 0.01137328,
      "auxiliary_loss_mlp": 0.00762909,
      "balance_loss_clip": 1.04518366,
      "balance_loss_mlp": 1.00060987,
      "epoch": 0.6650634281248121,
      "flos": 17019575700480.0,
      "grad_norm": 2.0045402465359885,
      "language_loss": 0.80801779,
      "learning_rate": 1.0656759195472447e-06,
      "loss": 0.82702017,
      "num_input_tokens_seen": 119007150,
      "step": 5531,
      "time_per_iteration": 2.468345880508423
    },
    {
      "auxiliary_loss_clip": 0.01048867,
      "auxiliary_loss_mlp": 0.01001513,
      "balance_loss_clip": 1.01583886,
      "balance_loss_mlp": 1.00061309,
      "epoch": 0.6651836710154512,
      "flos": 69294810666240.0,
      "grad_norm": 0.7688547156161302,
      "language_loss": 0.59770262,
      "learning_rate": 1.0649872457325414e-06,
      "loss": 0.61820644,
      "num_input_tokens_seen": 119068435,
      "step": 5532,
      "time_per_iteration": 3.0359108448028564
    },
    {
      "auxiliary_loss_clip": 0.01060317,
      "auxiliary_loss_mlp": 0.01001244,
      "balance_loss_clip": 1.01387858,
      "balance_loss_mlp": 1.00034988,
      "epoch": 0.6653039139060903,
      "flos": 66883444882560.0,
      "grad_norm": 0.8540739284539888,
      "language_loss": 0.55148345,
      "learning_rate": 1.0642987137585278e-06,
      "loss": 0.57209909,
      "num_input_tokens_seen": 119127960,
      "step": 5533,
      "time_per_iteration": 2.9857611656188965
    },
    {
      "auxiliary_loss_clip": 0.01141108,
      "auxiliary_loss_mlp": 0.01025743,
      "balance_loss_clip": 1.04616153,
      "balance_loss_mlp": 1.01859641,
      "epoch": 0.6654241567967294,
      "flos": 21470056669440.0,
      "grad_norm": 1.677822983000996,
      "language_loss": 0.82282197,
      "learning_rate": 1.0636103237296561e-06,
      "loss": 0.84449053,
      "num_input_tokens_seen": 119146885,
      "step": 5534,
      "time_per_iteration": 2.504901647567749
    },
    {
      "auxiliary_loss_clip": 0.01155336,
      "auxiliary_loss_mlp": 0.01027502,
      "balance_loss_clip": 1.05058169,
      "balance_loss_mlp": 1.02081406,
      "epoch": 0.6655443996873684,
      "flos": 25119514391040.0,
      "grad_norm": 1.7147288675738555,
      "language_loss": 0.84097803,
      "learning_rate": 1.062922075750353e-06,
      "loss": 0.86280644,
      "num_input_tokens_seen": 119166900,
      "step": 5535,
      "time_per_iteration": 3.223299741744995
    },
    {
      "auxiliary_loss_clip": 0.0113081,
      "auxiliary_loss_mlp": 0.01024059,
      "balance_loss_clip": 1.04607654,
      "balance_loss_mlp": 1.01684368,
      "epoch": 0.6656646425780076,
      "flos": 17457326749440.0,
      "grad_norm": 1.9615097527664374,
      "language_loss": 0.72253799,
      "learning_rate": 1.0622339699250267e-06,
      "loss": 0.74408662,
      "num_input_tokens_seen": 119184820,
      "step": 5536,
      "time_per_iteration": 2.5011141300201416
    },
    {
      "auxiliary_loss_clip": 0.01127225,
      "auxiliary_loss_mlp": 0.0102323,
      "balance_loss_clip": 1.04398274,
      "balance_loss_mlp": 1.01616645,
      "epoch": 0.6657848854686467,
      "flos": 23434190213760.0,
      "grad_norm": 1.742560073184748,
      "language_loss": 0.79410326,
      "learning_rate": 1.0615460063580624e-06,
      "loss": 0.81560779,
      "num_input_tokens_seen": 119203295,
      "step": 5537,
      "time_per_iteration": 2.5618491172790527
    },
    {
      "auxiliary_loss_clip": 0.01145145,
      "auxiliary_loss_mlp": 0.01024693,
      "balance_loss_clip": 1.04666209,
      "balance_loss_mlp": 1.01788604,
      "epoch": 0.6659051283592857,
      "flos": 11509909459200.0,
      "grad_norm": 1.7497521672685548,
      "language_loss": 0.73423094,
      "learning_rate": 1.060858185153821e-06,
      "loss": 0.75592935,
      "num_input_tokens_seen": 119221395,
      "step": 5538,
      "time_per_iteration": 2.4606072902679443
    },
    {
      "auxiliary_loss_clip": 0.01148398,
      "auxiliary_loss_mlp": 0.01024114,
      "balance_loss_clip": 1.04701769,
      "balance_loss_mlp": 1.01643658,
      "epoch": 0.6660253712499249,
      "flos": 20594554571520.0,
      "grad_norm": 2.1705685113249564,
      "language_loss": 0.76340932,
      "learning_rate": 1.0601705064166474e-06,
      "loss": 0.78513443,
      "num_input_tokens_seen": 119239790,
      "step": 5539,
      "time_per_iteration": 3.314908981323242
    },
    {
      "auxiliary_loss_clip": 0.01140862,
      "auxiliary_loss_mlp": 0.01027304,
      "balance_loss_clip": 1.04999018,
      "balance_loss_mlp": 1.01981151,
      "epoch": 0.666145614140564,
      "flos": 21251504367360.0,
      "grad_norm": 1.98334632172986,
      "language_loss": 0.7346797,
      "learning_rate": 1.0594829702508596e-06,
      "loss": 0.75636137,
      "num_input_tokens_seen": 119257505,
      "step": 5540,
      "time_per_iteration": 3.3076515197753906
    },
    {
      "auxiliary_loss_clip": 0.01129677,
      "auxiliary_loss_mlp": 0.01022363,
      "balance_loss_clip": 1.04428756,
      "balance_loss_mlp": 1.0152936,
      "epoch": 0.666265857031203,
      "flos": 33726188200320.0,
      "grad_norm": 3.951639745684329,
      "language_loss": 0.54971969,
      "learning_rate": 1.0587955767607592e-06,
      "loss": 0.57124007,
      "num_input_tokens_seen": 119279365,
      "step": 5541,
      "time_per_iteration": 2.6498446464538574
    },
    {
      "auxiliary_loss_clip": 0.01171984,
      "auxiliary_loss_mlp": 0.01027053,
      "balance_loss_clip": 1.05083442,
      "balance_loss_mlp": 1.01951885,
      "epoch": 0.6663860999218422,
      "flos": 17456644391040.0,
      "grad_norm": 2.1541790337222495,
      "language_loss": 0.76832557,
      "learning_rate": 1.0581083260506206e-06,
      "loss": 0.79031593,
      "num_input_tokens_seen": 119296150,
      "step": 5542,
      "time_per_iteration": 2.4024274349212646
    },
    {
      "auxiliary_loss_clip": 0.01140099,
      "auxiliary_loss_mlp": 0.01025213,
      "balance_loss_clip": 1.04537153,
      "balance_loss_mlp": 1.01805437,
      "epoch": 0.6665063428124812,
      "flos": 17676740977920.0,
      "grad_norm": 2.126645932810646,
      "language_loss": 0.76756954,
      "learning_rate": 1.0574212182246993e-06,
      "loss": 0.7892226,
      "num_input_tokens_seen": 119314845,
      "step": 5543,
      "time_per_iteration": 3.2622148990631104
    },
    {
      "auxiliary_loss_clip": 0.01146619,
      "auxiliary_loss_mlp": 0.01024303,
      "balance_loss_clip": 1.04625523,
      "balance_loss_mlp": 1.01603603,
      "epoch": 0.6666265857031203,
      "flos": 27673265687040.0,
      "grad_norm": 2.236547372071661,
      "language_loss": 0.76049602,
      "learning_rate": 1.0567342533872303e-06,
      "loss": 0.78220528,
      "num_input_tokens_seen": 119334875,
      "step": 5544,
      "time_per_iteration": 2.5519330501556396
    },
    {
      "auxiliary_loss_clip": 0.01144798,
      "auxiliary_loss_mlp": 0.01025208,
      "balance_loss_clip": 1.04801226,
      "balance_loss_mlp": 1.01758456,
      "epoch": 0.6667468285937594,
      "flos": 25046831220480.0,
      "grad_norm": 1.7298219727837763,
      "language_loss": 0.81157649,
      "learning_rate": 1.0560474316424255e-06,
      "loss": 0.83327657,
      "num_input_tokens_seen": 119354635,
      "step": 5545,
      "time_per_iteration": 2.52504563331604
    },
    {
      "auxiliary_loss_clip": 0.01142637,
      "auxiliary_loss_mlp": 0.01029472,
      "balance_loss_clip": 1.04534674,
      "balance_loss_mlp": 1.0210973,
      "epoch": 0.6668670714843985,
      "flos": 22780472641920.0,
      "grad_norm": 2.8473794769995115,
      "language_loss": 0.73465812,
      "learning_rate": 1.0553607530944746e-06,
      "loss": 0.75637919,
      "num_input_tokens_seen": 119372690,
      "step": 5546,
      "time_per_iteration": 2.498440980911255
    },
    {
      "auxiliary_loss_clip": 0.0112748,
      "auxiliary_loss_mlp": 0.01027001,
      "balance_loss_clip": 1.04397225,
      "balance_loss_mlp": 1.01971757,
      "epoch": 0.6669873143750376,
      "flos": 22163886754560.0,
      "grad_norm": 2.2452083971023606,
      "language_loss": 0.89550394,
      "learning_rate": 1.0546742178475463e-06,
      "loss": 0.91704875,
      "num_input_tokens_seen": 119391685,
      "step": 5547,
      "time_per_iteration": 2.5326223373413086
    },
    {
      "auxiliary_loss_clip": 0.0111963,
      "auxiliary_loss_mlp": 0.0102169,
      "balance_loss_clip": 1.04582059,
      "balance_loss_mlp": 1.0146569,
      "epoch": 0.6671075572656767,
      "flos": 20514832335360.0,
      "grad_norm": 1.714228263651133,
      "language_loss": 0.86919343,
      "learning_rate": 1.0539878260057868e-06,
      "loss": 0.89060658,
      "num_input_tokens_seen": 119410725,
      "step": 5548,
      "time_per_iteration": 2.643723487854004
    },
    {
      "auxiliary_loss_clip": 0.01161279,
      "auxiliary_loss_mlp": 0.01023832,
      "balance_loss_clip": 1.05214167,
      "balance_loss_mlp": 1.0153923,
      "epoch": 0.6672278001563158,
      "flos": 17931203902080.0,
      "grad_norm": 2.4429975224497356,
      "language_loss": 0.68255234,
      "learning_rate": 1.0533015776733226e-06,
      "loss": 0.70440346,
      "num_input_tokens_seen": 119426875,
      "step": 5549,
      "time_per_iteration": 2.5756430625915527
    },
    {
      "auxiliary_loss_clip": 0.01141113,
      "auxiliary_loss_mlp": 0.01024153,
      "balance_loss_clip": 1.04726541,
      "balance_loss_mlp": 1.0162375,
      "epoch": 0.6673480430469548,
      "flos": 22342146975360.0,
      "grad_norm": 2.388490199378825,
      "language_loss": 0.78299356,
      "learning_rate": 1.0526154729542566e-06,
      "loss": 0.80464613,
      "num_input_tokens_seen": 119446935,
      "step": 5550,
      "time_per_iteration": 2.6153078079223633
    },
    {
      "auxiliary_loss_clip": 0.01128532,
      "auxiliary_loss_mlp": 0.0102835,
      "balance_loss_clip": 1.04624462,
      "balance_loss_mlp": 1.02025509,
      "epoch": 0.6674682859375939,
      "flos": 20703830722560.0,
      "grad_norm": 2.2910147188210037,
      "language_loss": 0.79933047,
      "learning_rate": 1.0519295119526699e-06,
      "loss": 0.82089937,
      "num_input_tokens_seen": 119463240,
      "step": 5551,
      "time_per_iteration": 2.6724295616149902
    },
    {
      "auxiliary_loss_clip": 0.01145757,
      "auxiliary_loss_mlp": 0.01021979,
      "balance_loss_clip": 1.04754233,
      "balance_loss_mlp": 1.01433754,
      "epoch": 0.667588528828233,
      "flos": 26206673379840.0,
      "grad_norm": 1.5638411910386156,
      "language_loss": 0.83154821,
      "learning_rate": 1.0512436947726227e-06,
      "loss": 0.85322547,
      "num_input_tokens_seen": 119484655,
      "step": 5552,
      "time_per_iteration": 2.6712429523468018
    },
    {
      "auxiliary_loss_clip": 0.01127753,
      "auxiliary_loss_mlp": 0.01019472,
      "balance_loss_clip": 1.0423404,
      "balance_loss_mlp": 1.01151538,
      "epoch": 0.6677087717188721,
      "flos": 23071025756160.0,
      "grad_norm": 2.159858252587647,
      "language_loss": 0.65541106,
      "learning_rate": 1.0505580215181517e-06,
      "loss": 0.67688334,
      "num_input_tokens_seen": 119502895,
      "step": 5553,
      "time_per_iteration": 2.5411529541015625
    },
    {
      "auxiliary_loss_clip": 0.01028389,
      "auxiliary_loss_mlp": 0.01000971,
      "balance_loss_clip": 1.01356769,
      "balance_loss_mlp": 1.00015473,
      "epoch": 0.6678290146095112,
      "flos": 70941315219840.0,
      "grad_norm": 0.9283409737524261,
      "language_loss": 0.56688583,
      "learning_rate": 1.0498724922932753e-06,
      "loss": 0.58717942,
      "num_input_tokens_seen": 119561010,
      "step": 5554,
      "time_per_iteration": 3.0315301418304443
    },
    {
      "auxiliary_loss_clip": 0.01176606,
      "auxiliary_loss_mlp": 0.01025569,
      "balance_loss_clip": 1.05287588,
      "balance_loss_mlp": 1.0175705,
      "epoch": 0.6679492575001503,
      "flos": 18661088263680.0,
      "grad_norm": 2.0944287112077267,
      "language_loss": 0.86728042,
      "learning_rate": 1.0491871072019851e-06,
      "loss": 0.88930219,
      "num_input_tokens_seen": 119578900,
      "step": 5555,
      "time_per_iteration": 2.3964390754699707
    },
    {
      "auxiliary_loss_clip": 0.01133213,
      "auxiliary_loss_mlp": 0.01027566,
      "balance_loss_clip": 1.04447389,
      "balance_loss_mlp": 1.02036595,
      "epoch": 0.6680695003907894,
      "flos": 29711985822720.0,
      "grad_norm": 1.7539777547302515,
      "language_loss": 0.63716698,
      "learning_rate": 1.0485018663482555e-06,
      "loss": 0.65877473,
      "num_input_tokens_seen": 119598920,
      "step": 5556,
      "time_per_iteration": 2.5800023078918457
    },
    {
      "auxiliary_loss_clip": 0.01153377,
      "auxiliary_loss_mlp": 0.01021948,
      "balance_loss_clip": 1.04778492,
      "balance_loss_mlp": 1.01407397,
      "epoch": 0.6681897432814284,
      "flos": 28218964083840.0,
      "grad_norm": 2.540842597012571,
      "language_loss": 0.70446754,
      "learning_rate": 1.0478167698360354e-06,
      "loss": 0.72622073,
      "num_input_tokens_seen": 119618220,
      "step": 5557,
      "time_per_iteration": 2.4940547943115234
    },
    {
      "auxiliary_loss_clip": 0.0115044,
      "auxiliary_loss_mlp": 0.01024418,
      "balance_loss_clip": 1.04641485,
      "balance_loss_mlp": 1.01664209,
      "epoch": 0.6683099861720676,
      "flos": 25046543911680.0,
      "grad_norm": 2.6775474687521372,
      "language_loss": 0.70299375,
      "learning_rate": 1.0471318177692556e-06,
      "loss": 0.72474235,
      "num_input_tokens_seen": 119638520,
      "step": 5558,
      "time_per_iteration": 2.478252410888672
    },
    {
      "auxiliary_loss_clip": 0.0111925,
      "auxiliary_loss_mlp": 0.01027877,
      "balance_loss_clip": 1.04527569,
      "balance_loss_mlp": 1.02043796,
      "epoch": 0.6684302290627067,
      "flos": 22996977868800.0,
      "grad_norm": 3.2440330190469435,
      "language_loss": 0.76142788,
      "learning_rate": 1.046447010251821e-06,
      "loss": 0.78289914,
      "num_input_tokens_seen": 119655850,
      "step": 5559,
      "time_per_iteration": 2.5569262504577637
    },
    {
      "auxiliary_loss_clip": 0.01142883,
      "auxiliary_loss_mlp": 0.01025282,
      "balance_loss_clip": 1.04899716,
      "balance_loss_mlp": 1.01816535,
      "epoch": 0.6685504719533457,
      "flos": 26573824247040.0,
      "grad_norm": 2.138365777084513,
      "language_loss": 0.75626755,
      "learning_rate": 1.0457623473876157e-06,
      "loss": 0.77794921,
      "num_input_tokens_seen": 119675355,
      "step": 5560,
      "time_per_iteration": 2.578597068786621
    },
    {
      "auxiliary_loss_clip": 0.01169231,
      "auxiliary_loss_mlp": 0.01024956,
      "balance_loss_clip": 1.04872036,
      "balance_loss_mlp": 1.01811969,
      "epoch": 0.6686707148439849,
      "flos": 28986087870720.0,
      "grad_norm": 1.868925726003123,
      "language_loss": 0.71220756,
      "learning_rate": 1.0450778292805046e-06,
      "loss": 0.73414946,
      "num_input_tokens_seen": 119695340,
      "step": 5561,
      "time_per_iteration": 2.4805686473846436
    },
    {
      "auxiliary_loss_clip": 0.01159164,
      "auxiliary_loss_mlp": 0.01026473,
      "balance_loss_clip": 1.04660451,
      "balance_loss_mlp": 1.01919794,
      "epoch": 0.6687909577346239,
      "flos": 23623152687360.0,
      "grad_norm": 2.0247041898382103,
      "language_loss": 0.78722423,
      "learning_rate": 1.0443934560343267e-06,
      "loss": 0.8090806,
      "num_input_tokens_seen": 119716750,
      "step": 5562,
      "time_per_iteration": 3.313943862915039
    },
    {
      "auxiliary_loss_clip": 0.01115409,
      "auxiliary_loss_mlp": 0.01024399,
      "balance_loss_clip": 1.04246783,
      "balance_loss_mlp": 1.01709461,
      "epoch": 0.668911200625263,
      "flos": 23148593176320.0,
      "grad_norm": 2.0556870240121534,
      "language_loss": 0.78214395,
      "learning_rate": 1.0437092277529034e-06,
      "loss": 0.80354202,
      "num_input_tokens_seen": 119736005,
      "step": 5563,
      "time_per_iteration": 2.525938034057617
    },
    {
      "auxiliary_loss_clip": 0.01124505,
      "auxiliary_loss_mlp": 0.01025881,
      "balance_loss_clip": 1.04078555,
      "balance_loss_mlp": 1.01880646,
      "epoch": 0.6690314435159022,
      "flos": 18551919853440.0,
      "grad_norm": 2.0346624199803913,
      "language_loss": 0.73141664,
      "learning_rate": 1.0430251445400292e-06,
      "loss": 0.75292051,
      "num_input_tokens_seen": 119754050,
      "step": 5564,
      "time_per_iteration": 2.4879183769226074
    },
    {
      "auxiliary_loss_clip": 0.0108693,
      "auxiliary_loss_mlp": 0.01025199,
      "balance_loss_clip": 1.04578424,
      "balance_loss_mlp": 1.01773643,
      "epoch": 0.6691516864065412,
      "flos": 31759540704000.0,
      "grad_norm": 2.29490684102977,
      "language_loss": 0.62464035,
      "learning_rate": 1.0423412064994787e-06,
      "loss": 0.64576161,
      "num_input_tokens_seen": 119774820,
      "step": 5565,
      "time_per_iteration": 2.6734249591827393
    },
    {
      "auxiliary_loss_clip": 0.01130492,
      "auxiliary_loss_mlp": 0.01023299,
      "balance_loss_clip": 1.0444777,
      "balance_loss_mlp": 1.01587188,
      "epoch": 0.6692719292971803,
      "flos": 34933864296960.0,
      "grad_norm": 2.1299899439090906,
      "language_loss": 0.73788476,
      "learning_rate": 1.0416574137350064e-06,
      "loss": 0.75942266,
      "num_input_tokens_seen": 119795525,
      "step": 5566,
      "time_per_iteration": 3.489614963531494
    },
    {
      "auxiliary_loss_clip": 0.01149282,
      "auxiliary_loss_mlp": 0.01027406,
      "balance_loss_clip": 1.04733348,
      "balance_loss_mlp": 1.01975322,
      "epoch": 0.6693921721878194,
      "flos": 20449188230400.0,
      "grad_norm": 2.0446985089643475,
      "language_loss": 0.81325543,
      "learning_rate": 1.0409737663503428e-06,
      "loss": 0.83502233,
      "num_input_tokens_seen": 119813905,
      "step": 5567,
      "time_per_iteration": 3.256037473678589
    },
    {
      "auxiliary_loss_clip": 0.0115332,
      "auxiliary_loss_mlp": 0.0102903,
      "balance_loss_clip": 1.04498446,
      "balance_loss_mlp": 1.02097762,
      "epoch": 0.6695124150784585,
      "flos": 16614538963200.0,
      "grad_norm": 2.436174842291626,
      "language_loss": 0.82967198,
      "learning_rate": 1.040290264449196e-06,
      "loss": 0.8514955,
      "num_input_tokens_seen": 119832010,
      "step": 5568,
      "time_per_iteration": 2.4484620094299316
    },
    {
      "auxiliary_loss_clip": 0.01152272,
      "auxiliary_loss_mlp": 0.01028968,
      "balance_loss_clip": 1.0483197,
      "balance_loss_mlp": 1.02178299,
      "epoch": 0.6696326579690975,
      "flos": 26652145852800.0,
      "grad_norm": 3.8486687727426134,
      "language_loss": 0.63670647,
      "learning_rate": 1.0396069081352532e-06,
      "loss": 0.65851885,
      "num_input_tokens_seen": 119851165,
      "step": 5569,
      "time_per_iteration": 3.2229678630828857
    },
    {
      "auxiliary_loss_clip": 0.01068641,
      "auxiliary_loss_mlp": 0.01002517,
      "balance_loss_clip": 1.0130806,
      "balance_loss_mlp": 1.00153947,
      "epoch": 0.6697529008597367,
      "flos": 66964603662720.0,
      "grad_norm": 0.7728842107677465,
      "language_loss": 0.56076854,
      "learning_rate": 1.0389236975121782e-06,
      "loss": 0.58148009,
      "num_input_tokens_seen": 119906015,
      "step": 5570,
      "time_per_iteration": 2.9311363697052
    },
    {
      "auxiliary_loss_clip": 0.01172728,
      "auxiliary_loss_mlp": 0.01021099,
      "balance_loss_clip": 1.04959297,
      "balance_loss_mlp": 1.01331449,
      "epoch": 0.6698731437503758,
      "flos": 20886939279360.0,
      "grad_norm": 2.2919833931668183,
      "language_loss": 0.71307224,
      "learning_rate": 1.0382406326836147e-06,
      "loss": 0.7350105,
      "num_input_tokens_seen": 119925160,
      "step": 5571,
      "time_per_iteration": 2.4139564037323
    },
    {
      "auxiliary_loss_clip": 0.01163883,
      "auxiliary_loss_mlp": 0.01025774,
      "balance_loss_clip": 1.05001009,
      "balance_loss_mlp": 1.01755416,
      "epoch": 0.6699933866410148,
      "flos": 20409470766720.0,
      "grad_norm": 1.8400274426332732,
      "language_loss": 0.7611568,
      "learning_rate": 1.0375577137531828e-06,
      "loss": 0.7830534,
      "num_input_tokens_seen": 119943720,
      "step": 5572,
      "time_per_iteration": 2.4486207962036133
    },
    {
      "auxiliary_loss_clip": 0.01146559,
      "auxiliary_loss_mlp": 0.01025664,
      "balance_loss_clip": 1.0486697,
      "balance_loss_mlp": 1.01762354,
      "epoch": 0.670113629531654,
      "flos": 29023075900800.0,
      "grad_norm": 1.5204870415256866,
      "language_loss": 0.72181427,
      "learning_rate": 1.0368749408244802e-06,
      "loss": 0.74353653,
      "num_input_tokens_seen": 119966640,
      "step": 5573,
      "time_per_iteration": 2.5619585514068604
    },
    {
      "auxiliary_loss_clip": 0.0115276,
      "auxiliary_loss_mlp": 0.01028009,
      "balance_loss_clip": 1.04887307,
      "balance_loss_mlp": 1.0206275,
      "epoch": 0.670233872422293,
      "flos": 19791699730560.0,
      "grad_norm": 2.0509017342801608,
      "language_loss": 0.78862309,
      "learning_rate": 1.0361923140010836e-06,
      "loss": 0.81043077,
      "num_input_tokens_seen": 119985125,
      "step": 5574,
      "time_per_iteration": 2.4595985412597656
    },
    {
      "auxiliary_loss_clip": 0.01158599,
      "auxiliary_loss_mlp": 0.01021459,
      "balance_loss_clip": 1.04609215,
      "balance_loss_mlp": 1.01359069,
      "epoch": 0.6703541153129321,
      "flos": 24243689070720.0,
      "grad_norm": 2.4843856840743532,
      "language_loss": 0.63171947,
      "learning_rate": 1.0355098333865455e-06,
      "loss": 0.65351999,
      "num_input_tokens_seen": 120004355,
      "step": 5575,
      "time_per_iteration": 2.474578380584717
    },
    {
      "auxiliary_loss_clip": 0.01156168,
      "auxiliary_loss_mlp": 0.01028428,
      "balance_loss_clip": 1.05293179,
      "balance_loss_mlp": 1.02114725,
      "epoch": 0.6704743582035713,
      "flos": 26688523351680.0,
      "grad_norm": 1.5922564903622123,
      "language_loss": 0.69083208,
      "learning_rate": 1.0348274990844006e-06,
      "loss": 0.71267807,
      "num_input_tokens_seen": 120027115,
      "step": 5576,
      "time_per_iteration": 2.507566213607788
    },
    {
      "auxiliary_loss_clip": 0.0115701,
      "auxiliary_loss_mlp": 0.01026959,
      "balance_loss_clip": 1.04970455,
      "balance_loss_mlp": 1.01944304,
      "epoch": 0.6705946010942103,
      "flos": 23514379326720.0,
      "grad_norm": 1.7193449444515767,
      "language_loss": 0.73016864,
      "learning_rate": 1.034145311198155e-06,
      "loss": 0.75200832,
      "num_input_tokens_seen": 120047130,
      "step": 5577,
      "time_per_iteration": 2.457211494445801
    },
    {
      "auxiliary_loss_clip": 0.01166544,
      "auxiliary_loss_mlp": 0.01020109,
      "balance_loss_clip": 1.04788625,
      "balance_loss_mlp": 1.01320028,
      "epoch": 0.6707148439848494,
      "flos": 24061011477120.0,
      "grad_norm": 1.6692131200243168,
      "language_loss": 0.6354475,
      "learning_rate": 1.0334632698312989e-06,
      "loss": 0.65731406,
      "num_input_tokens_seen": 120067925,
      "step": 5578,
      "time_per_iteration": 2.4303925037384033
    },
    {
      "auxiliary_loss_clip": 0.01135283,
      "auxiliary_loss_mlp": 0.01029213,
      "balance_loss_clip": 1.04557943,
      "balance_loss_mlp": 1.02127075,
      "epoch": 0.6708350868754885,
      "flos": 22528667324160.0,
      "grad_norm": 2.0549815873472648,
      "language_loss": 0.75402439,
      "learning_rate": 1.032781375087295e-06,
      "loss": 0.77566946,
      "num_input_tokens_seen": 120087825,
      "step": 5579,
      "time_per_iteration": 2.4749789237976074
    },
    {
      "auxiliary_loss_clip": 0.01147297,
      "auxiliary_loss_mlp": 0.01024672,
      "balance_loss_clip": 1.05082631,
      "balance_loss_mlp": 1.01757944,
      "epoch": 0.6709553297661276,
      "flos": 25227749047680.0,
      "grad_norm": 1.795518310229102,
      "language_loss": 0.67651784,
      "learning_rate": 1.0320996270695891e-06,
      "loss": 0.69823748,
      "num_input_tokens_seen": 120108895,
      "step": 5580,
      "time_per_iteration": 2.531343936920166
    },
    {
      "auxiliary_loss_clip": 0.01127814,
      "auxiliary_loss_mlp": 0.01024539,
      "balance_loss_clip": 1.04389405,
      "balance_loss_mlp": 1.0167253,
      "epoch": 0.6710755726567667,
      "flos": 20448757267200.0,
      "grad_norm": 1.7611695924655835,
      "language_loss": 0.73460591,
      "learning_rate": 1.0314180258815998e-06,
      "loss": 0.75612944,
      "num_input_tokens_seen": 120127535,
      "step": 5581,
      "time_per_iteration": 2.5152859687805176
    },
    {
      "auxiliary_loss_clip": 0.01118559,
      "auxiliary_loss_mlp": 0.01022809,
      "balance_loss_clip": 1.04242873,
      "balance_loss_mlp": 1.01573372,
      "epoch": 0.6711958155474057,
      "flos": 25995411538560.0,
      "grad_norm": 1.5454506611844128,
      "language_loss": 0.7423296,
      "learning_rate": 1.0307365716267247e-06,
      "loss": 0.76374328,
      "num_input_tokens_seen": 120147980,
      "step": 5582,
      "time_per_iteration": 2.5693109035491943
    },
    {
      "auxiliary_loss_clip": 0.01156179,
      "auxiliary_loss_mlp": 0.01024722,
      "balance_loss_clip": 1.04835892,
      "balance_loss_mlp": 1.01736093,
      "epoch": 0.6713160584380449,
      "flos": 19937712516480.0,
      "grad_norm": 1.9203590595304838,
      "language_loss": 0.78199959,
      "learning_rate": 1.0300552644083423e-06,
      "loss": 0.80380857,
      "num_input_tokens_seen": 120166905,
      "step": 5583,
      "time_per_iteration": 2.4478225708007812
    },
    {
      "auxiliary_loss_clip": 0.0113581,
      "auxiliary_loss_mlp": 0.01022827,
      "balance_loss_clip": 1.04927218,
      "balance_loss_mlp": 1.01506698,
      "epoch": 0.6714363013286839,
      "flos": 18223373128320.0,
      "grad_norm": 2.2636071456574576,
      "language_loss": 0.72501814,
      "learning_rate": 1.0293741043298036e-06,
      "loss": 0.7466045,
      "num_input_tokens_seen": 120185255,
      "step": 5584,
      "time_per_iteration": 2.4972448348999023
    },
    {
      "auxiliary_loss_clip": 0.0113461,
      "auxiliary_loss_mlp": 0.01029248,
      "balance_loss_clip": 1.05077291,
      "balance_loss_mlp": 1.02131486,
      "epoch": 0.671556544219323,
      "flos": 25812374808960.0,
      "grad_norm": 4.039987393102375,
      "language_loss": 0.71544707,
      "learning_rate": 1.0286930914944436e-06,
      "loss": 0.7370857,
      "num_input_tokens_seen": 120205070,
      "step": 5585,
      "time_per_iteration": 2.544395923614502
    },
    {
      "auxiliary_loss_clip": 0.01170176,
      "auxiliary_loss_mlp": 0.01025143,
      "balance_loss_clip": 1.04692054,
      "balance_loss_mlp": 1.01794577,
      "epoch": 0.6716767871099621,
      "flos": 15850431918720.0,
      "grad_norm": 2.4592464930007387,
      "language_loss": 0.77125227,
      "learning_rate": 1.0280122260055684e-06,
      "loss": 0.7932055,
      "num_input_tokens_seen": 120220780,
      "step": 5586,
      "time_per_iteration": 2.398461103439331
    },
    {
      "auxiliary_loss_clip": 0.0117304,
      "auxiliary_loss_mlp": 0.01029076,
      "balance_loss_clip": 1.0507741,
      "balance_loss_mlp": 1.02094603,
      "epoch": 0.6717970300006012,
      "flos": 19756112330880.0,
      "grad_norm": 1.930092579841454,
      "language_loss": 0.820786,
      "learning_rate": 1.0273315079664652e-06,
      "loss": 0.84280711,
      "num_input_tokens_seen": 120238735,
      "step": 5587,
      "time_per_iteration": 2.4364752769470215
    },
    {
      "auxiliary_loss_clip": 0.01158978,
      "auxiliary_loss_mlp": 0.01022519,
      "balance_loss_clip": 1.04962456,
      "balance_loss_mlp": 1.01512241,
      "epoch": 0.6719172728912403,
      "flos": 25485049146240.0,
      "grad_norm": 2.1537941645484624,
      "language_loss": 0.74342352,
      "learning_rate": 1.0266509374803992e-06,
      "loss": 0.76523852,
      "num_input_tokens_seen": 120259895,
      "step": 5588,
      "time_per_iteration": 2.4993505477905273
    },
    {
      "auxiliary_loss_clip": 0.01170433,
      "auxiliary_loss_mlp": 0.00762336,
      "balance_loss_clip": 1.0483737,
      "balance_loss_mlp": 1.00059915,
      "epoch": 0.6720375157818794,
      "flos": 15880344969600.0,
      "grad_norm": 4.557218838866834,
      "language_loss": 0.84566152,
      "learning_rate": 1.0259705146506123e-06,
      "loss": 0.86498922,
      "num_input_tokens_seen": 120274790,
      "step": 5589,
      "time_per_iteration": 3.1294379234313965
    },
    {
      "auxiliary_loss_clip": 0.0115964,
      "auxiliary_loss_mlp": 0.01027452,
      "balance_loss_clip": 1.04859781,
      "balance_loss_mlp": 1.02023971,
      "epoch": 0.6721577586725185,
      "flos": 32010843231360.0,
      "grad_norm": 2.0961986089760716,
      "language_loss": 0.77502471,
      "learning_rate": 1.025290239580324e-06,
      "loss": 0.79689562,
      "num_input_tokens_seen": 120295460,
      "step": 5590,
      "time_per_iteration": 2.5450472831726074
    },
    {
      "auxiliary_loss_clip": 0.01114423,
      "auxiliary_loss_mlp": 0.01028646,
      "balance_loss_clip": 1.04314291,
      "balance_loss_mlp": 1.02130306,
      "epoch": 0.6722780015631575,
      "flos": 20737873837440.0,
      "grad_norm": 1.6192387201139165,
      "language_loss": 0.75764674,
      "learning_rate": 1.0246101123727313e-06,
      "loss": 0.77907741,
      "num_input_tokens_seen": 120314440,
      "step": 5591,
      "time_per_iteration": 2.5433592796325684
    },
    {
      "auxiliary_loss_clip": 0.01155405,
      "auxiliary_loss_mlp": 0.01030723,
      "balance_loss_clip": 1.04615891,
      "balance_loss_mlp": 1.023839,
      "epoch": 0.6723982444537967,
      "flos": 16909617191040.0,
      "grad_norm": 1.8983292224355133,
      "language_loss": 0.78469092,
      "learning_rate": 1.0239301331310085e-06,
      "loss": 0.80655217,
      "num_input_tokens_seen": 120332060,
      "step": 5592,
      "time_per_iteration": 3.2444515228271484
    },
    {
      "auxiliary_loss_clip": 0.01153206,
      "auxiliary_loss_mlp": 0.01025997,
      "balance_loss_clip": 1.04712093,
      "balance_loss_mlp": 1.01868939,
      "epoch": 0.6725184873444358,
      "flos": 20667812359680.0,
      "grad_norm": 1.6893222952765115,
      "language_loss": 0.88701761,
      "learning_rate": 1.0232503019583088e-06,
      "loss": 0.90880966,
      "num_input_tokens_seen": 120351670,
      "step": 5593,
      "time_per_iteration": 2.4609792232513428
    },
    {
      "auxiliary_loss_clip": 0.01151919,
      "auxiliary_loss_mlp": 0.01027316,
      "balance_loss_clip": 1.04745698,
      "balance_loss_mlp": 1.01968396,
      "epoch": 0.6726387302350748,
      "flos": 23727616416000.0,
      "grad_norm": 1.8391984003790942,
      "language_loss": 0.69723374,
      "learning_rate": 1.0225706189577619e-06,
      "loss": 0.71902609,
      "num_input_tokens_seen": 120370195,
      "step": 5594,
      "time_per_iteration": 3.295260190963745
    },
    {
      "auxiliary_loss_clip": 0.01158551,
      "auxiliary_loss_mlp": 0.01025529,
      "balance_loss_clip": 1.04849505,
      "balance_loss_mlp": 1.01766157,
      "epoch": 0.672758973125714,
      "flos": 15188274650880.0,
      "grad_norm": 2.4426036981054753,
      "language_loss": 0.7461617,
      "learning_rate": 1.021891084232475e-06,
      "loss": 0.76800251,
      "num_input_tokens_seen": 120388130,
      "step": 5595,
      "time_per_iteration": 2.416186571121216
    },
    {
      "auxiliary_loss_clip": 0.01154593,
      "auxiliary_loss_mlp": 0.01027123,
      "balance_loss_clip": 1.04541349,
      "balance_loss_mlp": 1.01931763,
      "epoch": 0.672879216016353,
      "flos": 18077252601600.0,
      "grad_norm": 3.0578557889030766,
      "language_loss": 0.80070502,
      "learning_rate": 1.0212116978855325e-06,
      "loss": 0.82252216,
      "num_input_tokens_seen": 120406145,
      "step": 5596,
      "time_per_iteration": 3.1221911907196045
    },
    {
      "auxiliary_loss_clip": 0.01126155,
      "auxiliary_loss_mlp": 0.01021038,
      "balance_loss_clip": 1.04460335,
      "balance_loss_mlp": 1.01410341,
      "epoch": 0.6729994589069921,
      "flos": 23476349802240.0,
      "grad_norm": 1.710851739438485,
      "language_loss": 0.78964907,
      "learning_rate": 1.020532460019997e-06,
      "loss": 0.81112099,
      "num_input_tokens_seen": 120425395,
      "step": 5597,
      "time_per_iteration": 2.524912118911743
    },
    {
      "auxiliary_loss_clip": 0.01091593,
      "auxiliary_loss_mlp": 0.01026481,
      "balance_loss_clip": 1.04280186,
      "balance_loss_mlp": 1.01897693,
      "epoch": 0.6731197017976313,
      "flos": 26322018929280.0,
      "grad_norm": 1.9175062700451448,
      "language_loss": 0.71083069,
      "learning_rate": 1.0198533707389096e-06,
      "loss": 0.73201144,
      "num_input_tokens_seen": 120446270,
      "step": 5598,
      "time_per_iteration": 2.6498665809631348
    },
    {
      "auxiliary_loss_clip": 0.01153994,
      "auxiliary_loss_mlp": 0.00762519,
      "balance_loss_clip": 1.04830921,
      "balance_loss_mlp": 1.00066662,
      "epoch": 0.6732399446882703,
      "flos": 21616428591360.0,
      "grad_norm": 4.479655399412542,
      "language_loss": 0.73157728,
      "learning_rate": 1.0191744301452853e-06,
      "loss": 0.75074244,
      "num_input_tokens_seen": 120465570,
      "step": 5599,
      "time_per_iteration": 2.437769889831543
    },
    {
      "auxiliary_loss_clip": 0.0116932,
      "auxiliary_loss_mlp": 0.01026874,
      "balance_loss_clip": 1.04865634,
      "balance_loss_mlp": 1.01938212,
      "epoch": 0.6733601875789094,
      "flos": 25880173729920.0,
      "grad_norm": 1.6607557564656918,
      "language_loss": 0.70416641,
      "learning_rate": 1.0184956383421208e-06,
      "loss": 0.72612834,
      "num_input_tokens_seen": 120484220,
      "step": 5600,
      "time_per_iteration": 2.4456331729888916
    },
    {
      "auxiliary_loss_clip": 0.01157701,
      "auxiliary_loss_mlp": 0.01025374,
      "balance_loss_clip": 1.04857826,
      "balance_loss_mlp": 1.01795614,
      "epoch": 0.6734804304695485,
      "flos": 22929573997440.0,
      "grad_norm": 1.9796694561357622,
      "language_loss": 0.65360671,
      "learning_rate": 1.017816995432387e-06,
      "loss": 0.67543745,
      "num_input_tokens_seen": 120503320,
      "step": 5601,
      "time_per_iteration": 2.464118719100952
    },
    {
      "auxiliary_loss_clip": 0.01142149,
      "auxiliary_loss_mlp": 0.01023438,
      "balance_loss_clip": 1.04594636,
      "balance_loss_mlp": 1.0158149,
      "epoch": 0.6736006733601876,
      "flos": 18697968552960.0,
      "grad_norm": 2.2448856128377646,
      "language_loss": 0.74459243,
      "learning_rate": 1.0171385015190353e-06,
      "loss": 0.76624835,
      "num_input_tokens_seen": 120523180,
      "step": 5602,
      "time_per_iteration": 2.466320276260376
    },
    {
      "auxiliary_loss_clip": 0.01140067,
      "auxiliary_loss_mlp": 0.00762745,
      "balance_loss_clip": 1.05057693,
      "balance_loss_mlp": 1.00071073,
      "epoch": 0.6737209162508266,
      "flos": 19427745173760.0,
      "grad_norm": 2.108869883709841,
      "language_loss": 0.7301839,
      "learning_rate": 1.0164601567049908e-06,
      "loss": 0.74921191,
      "num_input_tokens_seen": 120541710,
      "step": 5603,
      "time_per_iteration": 2.4771969318389893
    },
    {
      "auxiliary_loss_clip": 0.01142128,
      "auxiliary_loss_mlp": 0.01026764,
      "balance_loss_clip": 1.04705882,
      "balance_loss_mlp": 1.0188005,
      "epoch": 0.6738411591414658,
      "flos": 20158060498560.0,
      "grad_norm": 1.5791783503968118,
      "language_loss": 0.8043083,
      "learning_rate": 1.015781961093158e-06,
      "loss": 0.82599723,
      "num_input_tokens_seen": 120561030,
      "step": 5604,
      "time_per_iteration": 2.4761765003204346
    },
    {
      "auxiliary_loss_clip": 0.01143991,
      "auxiliary_loss_mlp": 0.01026062,
      "balance_loss_clip": 1.04299426,
      "balance_loss_mlp": 1.01859403,
      "epoch": 0.6739614020321049,
      "flos": 21653847584640.0,
      "grad_norm": 1.9417755597284516,
      "language_loss": 0.77187514,
      "learning_rate": 1.0151039147864197e-06,
      "loss": 0.79357564,
      "num_input_tokens_seen": 120581005,
      "step": 5605,
      "time_per_iteration": 2.4847517013549805
    },
    {
      "auxiliary_loss_clip": 0.010836,
      "auxiliary_loss_mlp": 0.01023915,
      "balance_loss_clip": 1.04823756,
      "balance_loss_mlp": 1.01589823,
      "epoch": 0.6740816449227439,
      "flos": 19171702051200.0,
      "grad_norm": 3.8051607042570157,
      "language_loss": 0.65878797,
      "learning_rate": 1.0144260178876336e-06,
      "loss": 0.6798631,
      "num_input_tokens_seen": 120600350,
      "step": 5606,
      "time_per_iteration": 2.6127288341522217
    },
    {
      "auxiliary_loss_clip": 0.01147329,
      "auxiliary_loss_mlp": 0.01020958,
      "balance_loss_clip": 1.04611468,
      "balance_loss_mlp": 1.01394498,
      "epoch": 0.6742018878133831,
      "flos": 21097015971840.0,
      "grad_norm": 2.5932104089263657,
      "language_loss": 0.67258775,
      "learning_rate": 1.0137482704996388e-06,
      "loss": 0.69427061,
      "num_input_tokens_seen": 120614700,
      "step": 5607,
      "time_per_iteration": 2.466639518737793
    },
    {
      "auxiliary_loss_clip": 0.01129851,
      "auxiliary_loss_mlp": 0.0102748,
      "balance_loss_clip": 1.04481328,
      "balance_loss_mlp": 1.01980269,
      "epoch": 0.6743221307040221,
      "flos": 23549966726400.0,
      "grad_norm": 4.481755641195638,
      "language_loss": 0.79232001,
      "learning_rate": 1.0130706727252461e-06,
      "loss": 0.81389338,
      "num_input_tokens_seen": 120631755,
      "step": 5608,
      "time_per_iteration": 2.5173239707946777
    },
    {
      "auxiliary_loss_clip": 0.0113129,
      "auxiliary_loss_mlp": 0.0102803,
      "balance_loss_clip": 1.04560113,
      "balance_loss_mlp": 1.02046633,
      "epoch": 0.6744423735946612,
      "flos": 16249542912000.0,
      "grad_norm": 2.7969143063846333,
      "language_loss": 0.68080229,
      "learning_rate": 1.0123932246672468e-06,
      "loss": 0.70239544,
      "num_input_tokens_seen": 120645900,
      "step": 5609,
      "time_per_iteration": 2.4734508991241455
    },
    {
      "auxiliary_loss_clip": 0.01026974,
      "auxiliary_loss_mlp": 0.00753028,
      "balance_loss_clip": 1.01471448,
      "balance_loss_mlp": 0.9998517,
      "epoch": 0.6745626164853004,
      "flos": 57843257829120.0,
      "grad_norm": 0.7804044379563917,
      "language_loss": 0.55836391,
      "learning_rate": 1.0117159264284114e-06,
      "loss": 0.57616389,
      "num_input_tokens_seen": 120709070,
      "step": 5610,
      "time_per_iteration": 3.112058162689209
    },
    {
      "auxiliary_loss_clip": 0.011472,
      "auxiliary_loss_mlp": 0.0102519,
      "balance_loss_clip": 1.04830813,
      "balance_loss_mlp": 1.0180583,
      "epoch": 0.6746828593759394,
      "flos": 20485027025280.0,
      "grad_norm": 1.8025215069265428,
      "language_loss": 0.77049118,
      "learning_rate": 1.0110387781114837e-06,
      "loss": 0.79221511,
      "num_input_tokens_seen": 120727685,
      "step": 5611,
      "time_per_iteration": 2.481444835662842
    },
    {
      "auxiliary_loss_clip": 0.01167564,
      "auxiliary_loss_mlp": 0.01026111,
      "balance_loss_clip": 1.0479244,
      "balance_loss_mlp": 1.01867855,
      "epoch": 0.6748031022665785,
      "flos": 19208223204480.0,
      "grad_norm": 2.1587504679037273,
      "language_loss": 0.77094889,
      "learning_rate": 1.0103617798191872e-06,
      "loss": 0.7928856,
      "num_input_tokens_seen": 120747160,
      "step": 5612,
      "time_per_iteration": 2.4058477878570557
    },
    {
      "auxiliary_loss_clip": 0.01141566,
      "auxiliary_loss_mlp": 0.01023379,
      "balance_loss_clip": 1.04869998,
      "balance_loss_mlp": 1.01558232,
      "epoch": 0.6749233451572175,
      "flos": 15195026407680.0,
      "grad_norm": 2.550253404705486,
      "language_loss": 0.82835215,
      "learning_rate": 1.0096849316542217e-06,
      "loss": 0.85000163,
      "num_input_tokens_seen": 120763710,
      "step": 5613,
      "time_per_iteration": 2.4668567180633545
    },
    {
      "auxiliary_loss_clip": 0.01073191,
      "auxiliary_loss_mlp": 0.01020718,
      "balance_loss_clip": 1.03747666,
      "balance_loss_mlp": 1.01296961,
      "epoch": 0.6750435880478567,
      "flos": 26499489050880.0,
      "grad_norm": 2.187206510248746,
      "language_loss": 0.74766231,
      "learning_rate": 1.0090082337192643e-06,
      "loss": 0.76860142,
      "num_input_tokens_seen": 120783355,
      "step": 5614,
      "time_per_iteration": 2.652745246887207
    },
    {
      "auxiliary_loss_clip": 0.01091801,
      "auxiliary_loss_mlp": 0.0102714,
      "balance_loss_clip": 1.03603482,
      "balance_loss_mlp": 1.01978159,
      "epoch": 0.6751638309384957,
      "flos": 23404313076480.0,
      "grad_norm": 1.9790381324417856,
      "language_loss": 0.78248245,
      "learning_rate": 1.0083316861169705e-06,
      "loss": 0.80367184,
      "num_input_tokens_seen": 120802090,
      "step": 5615,
      "time_per_iteration": 2.5952847003936768
    },
    {
      "auxiliary_loss_clip": 0.01131985,
      "auxiliary_loss_mlp": 0.01023283,
      "balance_loss_clip": 1.04292858,
      "balance_loss_mlp": 1.01494479,
      "epoch": 0.6752840738291348,
      "flos": 23441408847360.0,
      "grad_norm": 2.5462130476113334,
      "language_loss": 0.71312702,
      "learning_rate": 1.0076552889499713e-06,
      "loss": 0.73467976,
      "num_input_tokens_seen": 120822855,
      "step": 5616,
      "time_per_iteration": 3.2727737426757812
    },
    {
      "auxiliary_loss_clip": 0.01156204,
      "auxiliary_loss_mlp": 0.01025133,
      "balance_loss_clip": 1.0497458,
      "balance_loss_mlp": 1.01835907,
      "epoch": 0.675404316719774,
      "flos": 30335826257280.0,
      "grad_norm": 2.3543251133200545,
      "language_loss": 0.73736441,
      "learning_rate": 1.006979042320876e-06,
      "loss": 0.75917774,
      "num_input_tokens_seen": 120843070,
      "step": 5617,
      "time_per_iteration": 2.512648344039917
    },
    {
      "auxiliary_loss_clip": 0.01137699,
      "auxiliary_loss_mlp": 0.0102067,
      "balance_loss_clip": 1.04344237,
      "balance_loss_mlp": 1.01275742,
      "epoch": 0.675524559610413,
      "flos": 23622613983360.0,
      "grad_norm": 2.268019991433727,
      "language_loss": 0.62844241,
      "learning_rate": 1.0063029463322702e-06,
      "loss": 0.65002608,
      "num_input_tokens_seen": 120863345,
      "step": 5618,
      "time_per_iteration": 2.510601282119751
    },
    {
      "auxiliary_loss_clip": 0.01107286,
      "auxiliary_loss_mlp": 0.00762807,
      "balance_loss_clip": 1.04074073,
      "balance_loss_mlp": 1.00069857,
      "epoch": 0.6756448025010521,
      "flos": 21248631279360.0,
      "grad_norm": 2.3227387283972925,
      "language_loss": 0.752038,
      "learning_rate": 1.0056270010867164e-06,
      "loss": 0.7707389,
      "num_input_tokens_seen": 120880915,
      "step": 5619,
      "time_per_iteration": 3.3766841888427734
    },
    {
      "auxiliary_loss_clip": 0.01140258,
      "auxiliary_loss_mlp": 0.01027995,
      "balance_loss_clip": 1.04161584,
      "balance_loss_mlp": 1.02000773,
      "epoch": 0.6757650453916912,
      "flos": 21646521210240.0,
      "grad_norm": 2.4100007711594023,
      "language_loss": 0.77961367,
      "learning_rate": 1.004951206686758e-06,
      "loss": 0.80129623,
      "num_input_tokens_seen": 120899190,
      "step": 5620,
      "time_per_iteration": 3.291287660598755
    },
    {
      "auxiliary_loss_clip": 0.01150151,
      "auxiliary_loss_mlp": 0.01031102,
      "balance_loss_clip": 1.04589152,
      "balance_loss_mlp": 1.02357376,
      "epoch": 0.6758852882823303,
      "flos": 21795658479360.0,
      "grad_norm": 2.526588143775918,
      "language_loss": 0.71496648,
      "learning_rate": 1.0042755632349087e-06,
      "loss": 0.73677897,
      "num_input_tokens_seen": 120916080,
      "step": 5621,
      "time_per_iteration": 2.4356589317321777
    },
    {
      "auxiliary_loss_clip": 0.0112673,
      "auxiliary_loss_mlp": 0.01027917,
      "balance_loss_clip": 1.04364705,
      "balance_loss_mlp": 1.02014494,
      "epoch": 0.6760055311729694,
      "flos": 27088783580160.0,
      "grad_norm": 2.2547904810140884,
      "language_loss": 0.62818456,
      "learning_rate": 1.0036000708336653e-06,
      "loss": 0.64973104,
      "num_input_tokens_seen": 120935210,
      "step": 5622,
      "time_per_iteration": 3.3229434490203857
    },
    {
      "auxiliary_loss_clip": 0.01145278,
      "auxiliary_loss_mlp": 0.01028595,
      "balance_loss_clip": 1.04803848,
      "balance_loss_mlp": 1.02096009,
      "epoch": 0.6761257740636085,
      "flos": 17999792922240.0,
      "grad_norm": 2.546558619472363,
      "language_loss": 0.79283339,
      "learning_rate": 1.0029247295854984e-06,
      "loss": 0.8145721,
      "num_input_tokens_seen": 120951830,
      "step": 5623,
      "time_per_iteration": 2.4617576599121094
    },
    {
      "auxiliary_loss_clip": 0.01134581,
      "auxiliary_loss_mlp": 0.01025958,
      "balance_loss_clip": 1.04796314,
      "balance_loss_mlp": 1.0190202,
      "epoch": 0.6762460169542476,
      "flos": 15121912273920.0,
      "grad_norm": 2.0280833066145294,
      "language_loss": 0.71425265,
      "learning_rate": 1.0022495395928588e-06,
      "loss": 0.73585802,
      "num_input_tokens_seen": 120970310,
      "step": 5624,
      "time_per_iteration": 2.5078158378601074
    },
    {
      "auxiliary_loss_clip": 0.01070963,
      "auxiliary_loss_mlp": 0.01002205,
      "balance_loss_clip": 1.01545882,
      "balance_loss_mlp": 1.00110233,
      "epoch": 0.6763662598448866,
      "flos": 67886970030720.0,
      "grad_norm": 0.7908352360251991,
      "language_loss": 0.62347364,
      "learning_rate": 1.0015745009581697e-06,
      "loss": 0.64420533,
      "num_input_tokens_seen": 121031915,
      "step": 5625,
      "time_per_iteration": 3.0608749389648438
    },
    {
      "auxiliary_loss_clip": 0.01153176,
      "auxiliary_loss_mlp": 0.01023455,
      "balance_loss_clip": 1.04903865,
      "balance_loss_mlp": 1.01634443,
      "epoch": 0.6764865027355258,
      "flos": 20631829910400.0,
      "grad_norm": 1.8648441227093218,
      "language_loss": 0.66912121,
      "learning_rate": 1.0008996137838343e-06,
      "loss": 0.69088751,
      "num_input_tokens_seen": 121050890,
      "step": 5626,
      "time_per_iteration": 2.4541313648223877
    },
    {
      "auxiliary_loss_clip": 0.01174715,
      "auxiliary_loss_mlp": 0.01026978,
      "balance_loss_clip": 1.05032587,
      "balance_loss_mlp": 1.01912785,
      "epoch": 0.6766067456261649,
      "flos": 21215809226880.0,
      "grad_norm": 1.9934190685119748,
      "language_loss": 0.79698277,
      "learning_rate": 1.000224878172234e-06,
      "loss": 0.81899965,
      "num_input_tokens_seen": 121070015,
      "step": 5627,
      "time_per_iteration": 2.4292218685150146
    },
    {
      "auxiliary_loss_clip": 0.01158191,
      "auxiliary_loss_mlp": 0.01024436,
      "balance_loss_clip": 1.04711056,
      "balance_loss_mlp": 1.01689017,
      "epoch": 0.6767269885168039,
      "flos": 19938251220480.0,
      "grad_norm": 5.117672667238519,
      "language_loss": 0.72491854,
      "learning_rate": 9.99550294225724e-07,
      "loss": 0.74674487,
      "num_input_tokens_seen": 121089170,
      "step": 5628,
      "time_per_iteration": 2.4629600048065186
    },
    {
      "auxiliary_loss_clip": 0.01117135,
      "auxiliary_loss_mlp": 0.01028004,
      "balance_loss_clip": 1.04237509,
      "balance_loss_mlp": 1.02004123,
      "epoch": 0.6768472314074431,
      "flos": 20814076540800.0,
      "grad_norm": 2.005189705533757,
      "language_loss": 0.72329581,
      "learning_rate": 9.988758620466402e-07,
      "loss": 0.74474722,
      "num_input_tokens_seen": 121108040,
      "step": 5629,
      "time_per_iteration": 2.5583128929138184
    },
    {
      "auxiliary_loss_clip": 0.01103109,
      "auxiliary_loss_mlp": 0.01024231,
      "balance_loss_clip": 1.04010439,
      "balance_loss_mlp": 1.01713777,
      "epoch": 0.6769674742980821,
      "flos": 23186012169600.0,
      "grad_norm": 1.5534478381106351,
      "language_loss": 0.76141167,
      "learning_rate": 9.982015817372917e-07,
      "loss": 0.78268504,
      "num_input_tokens_seen": 121128480,
      "step": 5630,
      "time_per_iteration": 2.6117234230041504
    },
    {
      "auxiliary_loss_clip": 0.0110992,
      "auxiliary_loss_mlp": 0.01024894,
      "balance_loss_clip": 1.03948462,
      "balance_loss_mlp": 1.01715136,
      "epoch": 0.6770877171887212,
      "flos": 24242934885120.0,
      "grad_norm": 2.455341481587658,
      "language_loss": 0.81844622,
      "learning_rate": 9.975274533999657e-07,
      "loss": 0.83979434,
      "num_input_tokens_seen": 121148010,
      "step": 5631,
      "time_per_iteration": 2.5707125663757324
    },
    {
      "auxiliary_loss_clip": 0.01171228,
      "auxiliary_loss_mlp": 0.01028535,
      "balance_loss_clip": 1.04837656,
      "balance_loss_mlp": 1.02057219,
      "epoch": 0.6772079600793603,
      "flos": 18141567903360.0,
      "grad_norm": 2.783368541098781,
      "language_loss": 0.83714271,
      "learning_rate": 9.96853477136929e-07,
      "loss": 0.8591404,
      "num_input_tokens_seen": 121162755,
      "step": 5632,
      "time_per_iteration": 2.3982958793640137
    },
    {
      "auxiliary_loss_clip": 0.01119539,
      "auxiliary_loss_mlp": 0.01023011,
      "balance_loss_clip": 1.04142535,
      "balance_loss_mlp": 1.01550364,
      "epoch": 0.6773282029699994,
      "flos": 22452069571200.0,
      "grad_norm": 2.588647034896108,
      "language_loss": 0.75288522,
      "learning_rate": 9.96179653050422e-07,
      "loss": 0.77431077,
      "num_input_tokens_seen": 121182915,
      "step": 5633,
      "time_per_iteration": 2.650327682495117
    },
    {
      "auxiliary_loss_clip": 0.0112272,
      "auxiliary_loss_mlp": 0.01024975,
      "balance_loss_clip": 1.04434836,
      "balance_loss_mlp": 1.0171752,
      "epoch": 0.6774484458606385,
      "flos": 18693730748160.0,
      "grad_norm": 2.1216665205693612,
      "language_loss": 0.73863685,
      "learning_rate": 9.955059812426635e-07,
      "loss": 0.76011372,
      "num_input_tokens_seen": 121200445,
      "step": 5634,
      "time_per_iteration": 2.4857876300811768
    },
    {
      "auxiliary_loss_clip": 0.01172598,
      "auxiliary_loss_mlp": 0.01025694,
      "balance_loss_clip": 1.05231726,
      "balance_loss_mlp": 1.01793623,
      "epoch": 0.6775686887512776,
      "flos": 25994046821760.0,
      "grad_norm": 3.080649696603547,
      "language_loss": 0.82902557,
      "learning_rate": 9.948324618158493e-07,
      "loss": 0.85100847,
      "num_input_tokens_seen": 121220785,
      "step": 5635,
      "time_per_iteration": 2.453493118286133
    },
    {
      "auxiliary_loss_clip": 0.01156853,
      "auxiliary_loss_mlp": 0.01027071,
      "balance_loss_clip": 1.04499531,
      "balance_loss_mlp": 1.01935267,
      "epoch": 0.6776889316419167,
      "flos": 13587987922560.0,
      "grad_norm": 2.1913885316081774,
      "language_loss": 0.77460003,
      "learning_rate": 9.941590948721502e-07,
      "loss": 0.79643929,
      "num_input_tokens_seen": 121237985,
      "step": 5636,
      "time_per_iteration": 2.4147961139678955
    },
    {
      "auxiliary_loss_clip": 0.01136668,
      "auxiliary_loss_mlp": 0.01023797,
      "balance_loss_clip": 1.04621148,
      "balance_loss_mlp": 1.01712763,
      "epoch": 0.6778091745325557,
      "flos": 27601121220480.0,
      "grad_norm": 1.6560030482033634,
      "language_loss": 0.76536179,
      "learning_rate": 9.934858805137188e-07,
      "loss": 0.78696638,
      "num_input_tokens_seen": 121258635,
      "step": 5637,
      "time_per_iteration": 2.535013437271118
    },
    {
      "auxiliary_loss_clip": 0.01149996,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 1.04593265,
      "balance_loss_mlp": 1.01905704,
      "epoch": 0.6779294174231949,
      "flos": 18734058743040.0,
      "grad_norm": 1.6317033481784171,
      "language_loss": 0.80404937,
      "learning_rate": 9.92812818842677e-07,
      "loss": 0.82581282,
      "num_input_tokens_seen": 121277810,
      "step": 5638,
      "time_per_iteration": 2.4280025959014893
    },
    {
      "auxiliary_loss_clip": 0.01154159,
      "auxiliary_loss_mlp": 0.01024871,
      "balance_loss_clip": 1.04875386,
      "balance_loss_mlp": 1.01741755,
      "epoch": 0.678049660313834,
      "flos": 45873797765760.0,
      "grad_norm": 1.728243752163979,
      "language_loss": 0.64215094,
      "learning_rate": 9.921399099611306e-07,
      "loss": 0.66394126,
      "num_input_tokens_seen": 121298975,
      "step": 5639,
      "time_per_iteration": 2.7496955394744873
    },
    {
      "auxiliary_loss_clip": 0.01131616,
      "auxiliary_loss_mlp": 0.00762256,
      "balance_loss_clip": 1.04502487,
      "balance_loss_mlp": 1.00069189,
      "epoch": 0.678169903204473,
      "flos": 19974556892160.0,
      "grad_norm": 1.6785396115507052,
      "language_loss": 0.69072688,
      "learning_rate": 9.914671539711588e-07,
      "loss": 0.70966566,
      "num_input_tokens_seen": 121318495,
      "step": 5640,
      "time_per_iteration": 2.531480073928833
    },
    {
      "auxiliary_loss_clip": 0.01082379,
      "auxiliary_loss_mlp": 0.01023493,
      "balance_loss_clip": 1.04075789,
      "balance_loss_mlp": 1.01645339,
      "epoch": 0.6782901460951122,
      "flos": 21395613732480.0,
      "grad_norm": 2.152253398748627,
      "language_loss": 0.78218687,
      "learning_rate": 9.90794550974817e-07,
      "loss": 0.80324554,
      "num_input_tokens_seen": 121338890,
      "step": 5641,
      "time_per_iteration": 2.6066811084747314
    },
    {
      "auxiliary_loss_clip": 0.01122211,
      "auxiliary_loss_mlp": 0.01030058,
      "balance_loss_clip": 1.04400957,
      "balance_loss_mlp": 1.0220263,
      "epoch": 0.6784103889857512,
      "flos": 21434002392960.0,
      "grad_norm": 3.9983651578338155,
      "language_loss": 0.81072259,
      "learning_rate": 9.901221010741407e-07,
      "loss": 0.83224529,
      "num_input_tokens_seen": 121358210,
      "step": 5642,
      "time_per_iteration": 3.3362085819244385
    },
    {
      "auxiliary_loss_clip": 0.01159912,
      "auxiliary_loss_mlp": 0.01026965,
      "balance_loss_clip": 1.04770613,
      "balance_loss_mlp": 1.01973498,
      "epoch": 0.6785306318763903,
      "flos": 32671923091200.0,
      "grad_norm": 1.8751191017556763,
      "language_loss": 0.74953198,
      "learning_rate": 9.894498043711375e-07,
      "loss": 0.77140081,
      "num_input_tokens_seen": 121379955,
      "step": 5643,
      "time_per_iteration": 2.5505597591400146
    },
    {
      "auxiliary_loss_clip": 0.01139171,
      "auxiliary_loss_mlp": 0.01023712,
      "balance_loss_clip": 1.04471803,
      "balance_loss_mlp": 1.01592207,
      "epoch": 0.6786508747670293,
      "flos": 25632139340160.0,
      "grad_norm": 2.6386515211156576,
      "language_loss": 0.69506228,
      "learning_rate": 9.887776609677962e-07,
      "loss": 0.71669114,
      "num_input_tokens_seen": 121401325,
      "step": 5644,
      "time_per_iteration": 2.537651300430298
    },
    {
      "auxiliary_loss_clip": 0.01116392,
      "auxiliary_loss_mlp": 0.01023004,
      "balance_loss_clip": 1.03983223,
      "balance_loss_mlp": 1.01574421,
      "epoch": 0.6787711176576685,
      "flos": 19171881619200.0,
      "grad_norm": 1.6030564398697167,
      "language_loss": 0.72312951,
      "learning_rate": 9.88105670966079e-07,
      "loss": 0.74452347,
      "num_input_tokens_seen": 121419785,
      "step": 5645,
      "time_per_iteration": 3.385706663131714
    },
    {
      "auxiliary_loss_clip": 0.01100664,
      "auxiliary_loss_mlp": 0.01023297,
      "balance_loss_clip": 1.04292321,
      "balance_loss_mlp": 1.01652002,
      "epoch": 0.6788913605483076,
      "flos": 13985159581440.0,
      "grad_norm": 1.8973781655752906,
      "language_loss": 0.78938323,
      "learning_rate": 9.874338344679283e-07,
      "loss": 0.81062281,
      "num_input_tokens_seen": 121435630,
      "step": 5646,
      "time_per_iteration": 3.346597671508789
    },
    {
      "auxiliary_loss_clip": 0.01166025,
      "auxiliary_loss_mlp": 0.01025239,
      "balance_loss_clip": 1.0481391,
      "balance_loss_mlp": 1.01830745,
      "epoch": 0.6790116034389466,
      "flos": 22017586659840.0,
      "grad_norm": 1.6880262874068503,
      "language_loss": 0.73913878,
      "learning_rate": 9.86762151575259e-07,
      "loss": 0.76105142,
      "num_input_tokens_seen": 121455625,
      "step": 5647,
      "time_per_iteration": 2.4226245880126953
    },
    {
      "auxiliary_loss_clip": 0.0111374,
      "auxiliary_loss_mlp": 0.00761737,
      "balance_loss_clip": 1.04470301,
      "balance_loss_mlp": 1.00074291,
      "epoch": 0.6791318463295858,
      "flos": 20922454851840.0,
      "grad_norm": 1.489565348735229,
      "language_loss": 0.80265582,
      "learning_rate": 9.860906223899651e-07,
      "loss": 0.8214106,
      "num_input_tokens_seen": 121475020,
      "step": 5648,
      "time_per_iteration": 2.552877902984619
    },
    {
      "auxiliary_loss_clip": 0.01146673,
      "auxiliary_loss_mlp": 0.01027643,
      "balance_loss_clip": 1.04672384,
      "balance_loss_mlp": 1.02049041,
      "epoch": 0.6792520892202248,
      "flos": 28512749422080.0,
      "grad_norm": 1.736217289933732,
      "language_loss": 0.75607276,
      "learning_rate": 9.854192470139184e-07,
      "loss": 0.77781594,
      "num_input_tokens_seen": 121496500,
      "step": 5649,
      "time_per_iteration": 3.318429708480835
    },
    {
      "auxiliary_loss_clip": 0.01139207,
      "auxiliary_loss_mlp": 0.01027389,
      "balance_loss_clip": 1.04700947,
      "balance_loss_mlp": 1.02064776,
      "epoch": 0.6793723321108639,
      "flos": 20011904058240.0,
      "grad_norm": 2.1702929071846317,
      "language_loss": 0.71754479,
      "learning_rate": 9.847480255489645e-07,
      "loss": 0.73921072,
      "num_input_tokens_seen": 121515525,
      "step": 5650,
      "time_per_iteration": 2.4825146198272705
    },
    {
      "auxiliary_loss_clip": 0.01144409,
      "auxiliary_loss_mlp": 0.01023607,
      "balance_loss_clip": 1.04576802,
      "balance_loss_mlp": 1.01654983,
      "epoch": 0.6794925750015031,
      "flos": 26649488246400.0,
      "grad_norm": 2.8997197262430947,
      "language_loss": 0.69172394,
      "learning_rate": 9.840769580969295e-07,
      "loss": 0.71340412,
      "num_input_tokens_seen": 121535965,
      "step": 5651,
      "time_per_iteration": 2.545766830444336
    },
    {
      "auxiliary_loss_clip": 0.01147496,
      "auxiliary_loss_mlp": 0.01022858,
      "balance_loss_clip": 1.04616284,
      "balance_loss_mlp": 1.01576543,
      "epoch": 0.6796128178921421,
      "flos": 21580374314880.0,
      "grad_norm": 1.863022121432462,
      "language_loss": 0.79777551,
      "learning_rate": 9.834060447596114e-07,
      "loss": 0.81947905,
      "num_input_tokens_seen": 121555235,
      "step": 5652,
      "time_per_iteration": 2.531484603881836
    },
    {
      "auxiliary_loss_clip": 0.01156653,
      "auxiliary_loss_mlp": 0.01022512,
      "balance_loss_clip": 1.04559255,
      "balance_loss_mlp": 1.01466799,
      "epoch": 0.6797330607827812,
      "flos": 22492002516480.0,
      "grad_norm": 2.1144039437172575,
      "language_loss": 0.77824962,
      "learning_rate": 9.827352856387868e-07,
      "loss": 0.80004132,
      "num_input_tokens_seen": 121574945,
      "step": 5653,
      "time_per_iteration": 2.479072332382202
    },
    {
      "auxiliary_loss_clip": 0.01021827,
      "auxiliary_loss_mlp": 0.01002487,
      "balance_loss_clip": 1.01295829,
      "balance_loss_mlp": 1.00139046,
      "epoch": 0.6798533036734203,
      "flos": 66306648286080.0,
      "grad_norm": 0.7751288183603556,
      "language_loss": 0.64240062,
      "learning_rate": 9.820646808362118e-07,
      "loss": 0.66264367,
      "num_input_tokens_seen": 121641200,
      "step": 5654,
      "time_per_iteration": 3.2031667232513428
    },
    {
      "auxiliary_loss_clip": 0.01136489,
      "auxiliary_loss_mlp": 0.01026423,
      "balance_loss_clip": 1.04510152,
      "balance_loss_mlp": 1.01930022,
      "epoch": 0.6799735465640594,
      "flos": 16180163792640.0,
      "grad_norm": 2.072371955800868,
      "language_loss": 0.72879732,
      "learning_rate": 9.813942304536154e-07,
      "loss": 0.75042641,
      "num_input_tokens_seen": 121659170,
      "step": 5655,
      "time_per_iteration": 2.459920644760132
    },
    {
      "auxiliary_loss_clip": 0.01141359,
      "auxiliary_loss_mlp": 0.01028855,
      "balance_loss_clip": 1.04512191,
      "balance_loss_mlp": 1.02179813,
      "epoch": 0.6800937894546984,
      "flos": 22125749489280.0,
      "grad_norm": 1.7380389302859622,
      "language_loss": 0.63639247,
      "learning_rate": 9.807239345927043e-07,
      "loss": 0.65809464,
      "num_input_tokens_seen": 121679180,
      "step": 5656,
      "time_per_iteration": 2.489053249359131
    },
    {
      "auxiliary_loss_clip": 0.01142877,
      "auxiliary_loss_mlp": 0.0102627,
      "balance_loss_clip": 1.042503,
      "balance_loss_mlp": 1.01863456,
      "epoch": 0.6802140323453376,
      "flos": 31612953300480.0,
      "grad_norm": 2.528153415639979,
      "language_loss": 0.72309339,
      "learning_rate": 9.80053793355162e-07,
      "loss": 0.74478483,
      "num_input_tokens_seen": 121697875,
      "step": 5657,
      "time_per_iteration": 2.5537240505218506
    },
    {
      "auxiliary_loss_clip": 0.01109629,
      "auxiliary_loss_mlp": 0.0102752,
      "balance_loss_clip": 1.04368401,
      "balance_loss_mlp": 1.01950884,
      "epoch": 0.6803342752359767,
      "flos": 17712938908800.0,
      "grad_norm": 2.1152385311104633,
      "language_loss": 0.7483629,
      "learning_rate": 9.793838068426472e-07,
      "loss": 0.76973438,
      "num_input_tokens_seen": 121715570,
      "step": 5658,
      "time_per_iteration": 2.50657057762146
    },
    {
      "auxiliary_loss_clip": 0.01168706,
      "auxiliary_loss_mlp": 0.01025539,
      "balance_loss_clip": 1.04902971,
      "balance_loss_mlp": 1.01799059,
      "epoch": 0.6804545181266157,
      "flos": 11326800902400.0,
      "grad_norm": 2.6904177023535283,
      "language_loss": 0.60966849,
      "learning_rate": 9.78713975156799e-07,
      "loss": 0.63161087,
      "num_input_tokens_seen": 121731435,
      "step": 5659,
      "time_per_iteration": 2.4328441619873047
    },
    {
      "auxiliary_loss_clip": 0.01130212,
      "auxiliary_loss_mlp": 0.01024989,
      "balance_loss_clip": 1.04981244,
      "balance_loss_mlp": 1.01722813,
      "epoch": 0.6805747610172549,
      "flos": 29350976181120.0,
      "grad_norm": 1.6504029865941823,
      "language_loss": 0.71490681,
      "learning_rate": 9.780442983992273e-07,
      "loss": 0.73645884,
      "num_input_tokens_seen": 121749950,
      "step": 5660,
      "time_per_iteration": 2.57289719581604
    },
    {
      "auxiliary_loss_clip": 0.01135305,
      "auxiliary_loss_mlp": 0.01025417,
      "balance_loss_clip": 1.04475737,
      "balance_loss_mlp": 1.01796675,
      "epoch": 0.680695003907894,
      "flos": 37631868612480.0,
      "grad_norm": 1.707315684940208,
      "language_loss": 0.71813023,
      "learning_rate": 9.773747766715238e-07,
      "loss": 0.73973745,
      "num_input_tokens_seen": 121770770,
      "step": 5661,
      "time_per_iteration": 2.6178576946258545
    },
    {
      "auxiliary_loss_clip": 0.01141844,
      "auxiliary_loss_mlp": 0.01026165,
      "balance_loss_clip": 1.04405189,
      "balance_loss_mlp": 1.01876211,
      "epoch": 0.680815246798533,
      "flos": 22127365601280.0,
      "grad_norm": 1.9599068485474205,
      "language_loss": 0.80376077,
      "learning_rate": 9.767054100752536e-07,
      "loss": 0.82544082,
      "num_input_tokens_seen": 121790720,
      "step": 5662,
      "time_per_iteration": 2.475688934326172
    },
    {
      "auxiliary_loss_clip": 0.01128996,
      "auxiliary_loss_mlp": 0.01024863,
      "balance_loss_clip": 1.04531837,
      "balance_loss_mlp": 1.01746607,
      "epoch": 0.6809354896891722,
      "flos": 17201822330880.0,
      "grad_norm": 1.8482289044336226,
      "language_loss": 0.81643987,
      "learning_rate": 9.760361987119584e-07,
      "loss": 0.83797848,
      "num_input_tokens_seen": 121808455,
      "step": 5663,
      "time_per_iteration": 2.4713306427001953
    },
    {
      "auxiliary_loss_clip": 0.011397,
      "auxiliary_loss_mlp": 0.01027554,
      "balance_loss_clip": 1.04399204,
      "balance_loss_mlp": 1.01951385,
      "epoch": 0.6810557325798112,
      "flos": 12458166554880.0,
      "grad_norm": 1.8615498439985667,
      "language_loss": 0.67311943,
      "learning_rate": 9.753671426831592e-07,
      "loss": 0.69479203,
      "num_input_tokens_seen": 121824470,
      "step": 5664,
      "time_per_iteration": 2.43747878074646
    },
    {
      "auxiliary_loss_clip": 0.01148724,
      "auxiliary_loss_mlp": 0.01026749,
      "balance_loss_clip": 1.04486561,
      "balance_loss_mlp": 1.01945019,
      "epoch": 0.6811759754704503,
      "flos": 22156165330560.0,
      "grad_norm": 1.7824829079284061,
      "language_loss": 0.79699397,
      "learning_rate": 9.746982420903483e-07,
      "loss": 0.81874871,
      "num_input_tokens_seen": 121842665,
      "step": 5665,
      "time_per_iteration": 2.4514105319976807
    },
    {
      "auxiliary_loss_clip": 0.0115555,
      "auxiliary_loss_mlp": 0.0102469,
      "balance_loss_clip": 1.05116224,
      "balance_loss_mlp": 1.0178926,
      "epoch": 0.6812962183610894,
      "flos": 17525377065600.0,
      "grad_norm": 1.9972980751986613,
      "language_loss": 0.74940825,
      "learning_rate": 9.740294970349993e-07,
      "loss": 0.77121067,
      "num_input_tokens_seen": 121859080,
      "step": 5666,
      "time_per_iteration": 2.420055627822876
    },
    {
      "auxiliary_loss_clip": 0.01049221,
      "auxiliary_loss_mlp": 0.01001837,
      "balance_loss_clip": 1.01485109,
      "balance_loss_mlp": 1.00085402,
      "epoch": 0.6814164612517285,
      "flos": 60274480855680.0,
      "grad_norm": 0.8743419920956141,
      "language_loss": 0.60887706,
      "learning_rate": 9.733609076185594e-07,
      "loss": 0.62938762,
      "num_input_tokens_seen": 121915485,
      "step": 5667,
      "time_per_iteration": 2.9459009170532227
    },
    {
      "auxiliary_loss_clip": 0.01154799,
      "auxiliary_loss_mlp": 0.01030116,
      "balance_loss_clip": 1.04893303,
      "balance_loss_mlp": 1.02233171,
      "epoch": 0.6815367041423676,
      "flos": 19317750750720.0,
      "grad_norm": 1.7976735443505019,
      "language_loss": 0.83713353,
      "learning_rate": 9.72692473942455e-07,
      "loss": 0.85898274,
      "num_input_tokens_seen": 121932710,
      "step": 5668,
      "time_per_iteration": 2.4389798641204834
    },
    {
      "auxiliary_loss_clip": 0.01120624,
      "auxiliary_loss_mlp": 0.01024956,
      "balance_loss_clip": 1.04871285,
      "balance_loss_mlp": 1.01714134,
      "epoch": 0.6816569470330067,
      "flos": 22161696024960.0,
      "grad_norm": 4.137210942081937,
      "language_loss": 0.77749372,
      "learning_rate": 9.720241961080849e-07,
      "loss": 0.79894954,
      "num_input_tokens_seen": 121952025,
      "step": 5669,
      "time_per_iteration": 3.269195556640625
    },
    {
      "auxiliary_loss_clip": 0.01168316,
      "auxiliary_loss_mlp": 0.01026068,
      "balance_loss_clip": 1.04765201,
      "balance_loss_mlp": 1.01866829,
      "epoch": 0.6817771899236458,
      "flos": 41463501137280.0,
      "grad_norm": 2.2527367203781505,
      "language_loss": 0.73023182,
      "learning_rate": 9.713560742168259e-07,
      "loss": 0.75217569,
      "num_input_tokens_seen": 121974650,
      "step": 5670,
      "time_per_iteration": 2.5809009075164795
    },
    {
      "auxiliary_loss_clip": 0.01125893,
      "auxiliary_loss_mlp": 0.0102939,
      "balance_loss_clip": 1.04355431,
      "balance_loss_mlp": 1.02238309,
      "epoch": 0.6818974328142848,
      "flos": 21106138026240.0,
      "grad_norm": 1.8757381891085267,
      "language_loss": 0.71325529,
      "learning_rate": 9.706881083700333e-07,
      "loss": 0.73480815,
      "num_input_tokens_seen": 121994335,
      "step": 5671,
      "time_per_iteration": 3.3486883640289307
    },
    {
      "auxiliary_loss_clip": 0.01100251,
      "auxiliary_loss_mlp": 0.01028385,
      "balance_loss_clip": 1.04716277,
      "balance_loss_mlp": 1.02052331,
      "epoch": 0.682017675704924,
      "flos": 20441897769600.0,
      "grad_norm": 1.9867108097680781,
      "language_loss": 0.82230723,
      "learning_rate": 9.700202986690357e-07,
      "loss": 0.84359366,
      "num_input_tokens_seen": 122012635,
      "step": 5672,
      "time_per_iteration": 2.5564680099487305
    },
    {
      "auxiliary_loss_clip": 0.0115358,
      "auxiliary_loss_mlp": 0.00762533,
      "balance_loss_clip": 1.04692459,
      "balance_loss_mlp": 1.00061178,
      "epoch": 0.682137918595563,
      "flos": 20044438801920.0,
      "grad_norm": 1.8920798058221098,
      "language_loss": 0.66328001,
      "learning_rate": 9.693526452151413e-07,
      "loss": 0.68244117,
      "num_input_tokens_seen": 122031685,
      "step": 5673,
      "time_per_iteration": 3.2407755851745605
    },
    {
      "auxiliary_loss_clip": 0.01132798,
      "auxiliary_loss_mlp": 0.01023499,
      "balance_loss_clip": 1.04496288,
      "balance_loss_mlp": 1.01537526,
      "epoch": 0.6822581614862021,
      "flos": 31684559063040.0,
      "grad_norm": 1.697455899643641,
      "language_loss": 0.75330949,
      "learning_rate": 9.686851481096305e-07,
      "loss": 0.77487242,
      "num_input_tokens_seen": 122052995,
      "step": 5674,
      "time_per_iteration": 2.5931127071380615
    },
    {
      "auxiliary_loss_clip": 0.01099548,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 1.04274535,
      "balance_loss_mlp": 1.01905751,
      "epoch": 0.6823784043768413,
      "flos": 23477570864640.0,
      "grad_norm": 1.8882355076551773,
      "language_loss": 0.7173987,
      "learning_rate": 9.68017807453762e-07,
      "loss": 0.73866153,
      "num_input_tokens_seen": 122071740,
      "step": 5675,
      "time_per_iteration": 2.5883944034576416
    },
    {
      "auxiliary_loss_clip": 0.01143675,
      "auxiliary_loss_mlp": 0.00762078,
      "balance_loss_clip": 1.04831815,
      "balance_loss_mlp": 1.00057149,
      "epoch": 0.6824986472674803,
      "flos": 14137134024960.0,
      "grad_norm": 1.8058269271248202,
      "language_loss": 0.72952092,
      "learning_rate": 9.673506233487721e-07,
      "loss": 0.74857843,
      "num_input_tokens_seen": 122089705,
      "step": 5676,
      "time_per_iteration": 3.24983811378479
    },
    {
      "auxiliary_loss_clip": 0.01141474,
      "auxiliary_loss_mlp": 0.00761906,
      "balance_loss_clip": 1.04461622,
      "balance_loss_mlp": 1.00050855,
      "epoch": 0.6826188901581194,
      "flos": 21504997624320.0,
      "grad_norm": 1.680776358043912,
      "language_loss": 0.8608079,
      "learning_rate": 9.666835958958717e-07,
      "loss": 0.87984169,
      "num_input_tokens_seen": 122109025,
      "step": 5677,
      "time_per_iteration": 2.4849982261657715
    },
    {
      "auxiliary_loss_clip": 0.01167742,
      "auxiliary_loss_mlp": 0.01020407,
      "balance_loss_clip": 1.04891181,
      "balance_loss_mlp": 1.01343036,
      "epoch": 0.6827391330487584,
      "flos": 20810126044800.0,
      "grad_norm": 2.4835521219678673,
      "language_loss": 0.80907482,
      "learning_rate": 9.660167251962484e-07,
      "loss": 0.83095634,
      "num_input_tokens_seen": 122127385,
      "step": 5678,
      "time_per_iteration": 2.4099223613739014
    },
    {
      "auxiliary_loss_clip": 0.01129298,
      "auxiliary_loss_mlp": 0.01023543,
      "balance_loss_clip": 1.04319572,
      "balance_loss_mlp": 1.01692426,
      "epoch": 0.6828593759393976,
      "flos": 21688788539520.0,
      "grad_norm": 10.157356116017443,
      "language_loss": 0.77760178,
      "learning_rate": 9.653500113510654e-07,
      "loss": 0.7991302,
      "num_input_tokens_seen": 122146500,
      "step": 5679,
      "time_per_iteration": 2.554255485534668
    },
    {
      "auxiliary_loss_clip": 0.01134528,
      "auxiliary_loss_mlp": 0.01027876,
      "balance_loss_clip": 1.04295611,
      "balance_loss_mlp": 1.02032137,
      "epoch": 0.6829796188300367,
      "flos": 25337707557120.0,
      "grad_norm": 2.442300309763335,
      "language_loss": 0.67099756,
      "learning_rate": 9.646834544614627e-07,
      "loss": 0.69262159,
      "num_input_tokens_seen": 122167000,
      "step": 5680,
      "time_per_iteration": 2.537274122238159
    },
    {
      "auxiliary_loss_clip": 0.01132529,
      "auxiliary_loss_mlp": 0.01026343,
      "balance_loss_clip": 1.04588199,
      "balance_loss_mlp": 1.0193994,
      "epoch": 0.6830998617206757,
      "flos": 20704800389760.0,
      "grad_norm": 1.7629091437102198,
      "language_loss": 0.76264429,
      "learning_rate": 9.64017054628558e-07,
      "loss": 0.78423297,
      "num_input_tokens_seen": 122185825,
      "step": 5681,
      "time_per_iteration": 2.467233419418335
    },
    {
      "auxiliary_loss_clip": 0.01114737,
      "auxiliary_loss_mlp": 0.01024885,
      "balance_loss_clip": 1.04224157,
      "balance_loss_mlp": 1.01790273,
      "epoch": 0.6832201046113149,
      "flos": 21726638496000.0,
      "grad_norm": 1.7542470196537288,
      "language_loss": 0.78997815,
      "learning_rate": 9.63350811953441e-07,
      "loss": 0.81137443,
      "num_input_tokens_seen": 122206200,
      "step": 5682,
      "time_per_iteration": 2.5623183250427246
    },
    {
      "auxiliary_loss_clip": 0.01127531,
      "auxiliary_loss_mlp": 0.01022149,
      "balance_loss_clip": 1.04386592,
      "balance_loss_mlp": 1.01493406,
      "epoch": 0.6833403475019539,
      "flos": 19536554448000.0,
      "grad_norm": 2.8896546595659363,
      "language_loss": 0.7039206,
      "learning_rate": 9.626847265371826e-07,
      "loss": 0.72541738,
      "num_input_tokens_seen": 122225520,
      "step": 5683,
      "time_per_iteration": 2.512218475341797
    },
    {
      "auxiliary_loss_clip": 0.01131057,
      "auxiliary_loss_mlp": 0.01025975,
      "balance_loss_clip": 1.04320049,
      "balance_loss_mlp": 1.01885772,
      "epoch": 0.683460590392593,
      "flos": 19352153001600.0,
      "grad_norm": 2.0864397872367526,
      "language_loss": 0.78533518,
      "learning_rate": 9.620187984808262e-07,
      "loss": 0.80690545,
      "num_input_tokens_seen": 122244320,
      "step": 5684,
      "time_per_iteration": 2.486323833465576
    },
    {
      "auxiliary_loss_clip": 0.01139136,
      "auxiliary_loss_mlp": 0.00761744,
      "balance_loss_clip": 1.04664588,
      "balance_loss_mlp": 1.0005579,
      "epoch": 0.6835808332832322,
      "flos": 23288500650240.0,
      "grad_norm": 1.610999900544913,
      "language_loss": 0.85963929,
      "learning_rate": 9.613530278853919e-07,
      "loss": 0.8786481,
      "num_input_tokens_seen": 122264295,
      "step": 5685,
      "time_per_iteration": 2.494279384613037
    },
    {
      "auxiliary_loss_clip": 0.01155453,
      "auxiliary_loss_mlp": 0.01023088,
      "balance_loss_clip": 1.04990995,
      "balance_loss_mlp": 1.01598287,
      "epoch": 0.6837010761738712,
      "flos": 21653416621440.0,
      "grad_norm": 2.021045836357281,
      "language_loss": 0.74426723,
      "learning_rate": 9.60687414851879e-07,
      "loss": 0.76605266,
      "num_input_tokens_seen": 122285300,
      "step": 5686,
      "time_per_iteration": 2.4731569290161133
    },
    {
      "auxiliary_loss_clip": 0.01143967,
      "auxiliary_loss_mlp": 0.01025345,
      "balance_loss_clip": 1.0489378,
      "balance_loss_mlp": 1.01801658,
      "epoch": 0.6838213190645103,
      "flos": 17566387418880.0,
      "grad_norm": 2.3009906422321733,
      "language_loss": 0.77402008,
      "learning_rate": 9.600219594812575e-07,
      "loss": 0.79571319,
      "num_input_tokens_seen": 122303240,
      "step": 5687,
      "time_per_iteration": 2.448185920715332
    },
    {
      "auxiliary_loss_clip": 0.01167513,
      "auxiliary_loss_mlp": 0.01024154,
      "balance_loss_clip": 1.04848981,
      "balance_loss_mlp": 1.01726413,
      "epoch": 0.6839415619551494,
      "flos": 23112538899840.0,
      "grad_norm": 1.692974361196155,
      "language_loss": 0.72805774,
      "learning_rate": 9.593566618744786e-07,
      "loss": 0.74997437,
      "num_input_tokens_seen": 122323390,
      "step": 5688,
      "time_per_iteration": 2.4155213832855225
    },
    {
      "auxiliary_loss_clip": 0.01167908,
      "auxiliary_loss_mlp": 0.01026168,
      "balance_loss_clip": 1.04748166,
      "balance_loss_mlp": 1.01888442,
      "epoch": 0.6840618048457885,
      "flos": 22127868391680.0,
      "grad_norm": 1.810355264466806,
      "language_loss": 0.73995751,
      "learning_rate": 9.58691522132466e-07,
      "loss": 0.76189828,
      "num_input_tokens_seen": 122342200,
      "step": 5689,
      "time_per_iteration": 2.4185643196105957
    },
    {
      "auxiliary_loss_clip": 0.01146562,
      "auxiliary_loss_mlp": 0.01025276,
      "balance_loss_clip": 1.04885101,
      "balance_loss_mlp": 1.01764023,
      "epoch": 0.6841820477364275,
      "flos": 22015898720640.0,
      "grad_norm": 2.397166713152011,
      "language_loss": 0.84819698,
      "learning_rate": 9.58026540356123e-07,
      "loss": 0.86991537,
      "num_input_tokens_seen": 122360465,
      "step": 5690,
      "time_per_iteration": 2.4903764724731445
    },
    {
      "auxiliary_loss_clip": 0.01155678,
      "auxiliary_loss_mlp": 0.01025346,
      "balance_loss_clip": 1.04545617,
      "balance_loss_mlp": 1.01803267,
      "epoch": 0.6843022906270667,
      "flos": 24900531125760.0,
      "grad_norm": 1.6046056119883991,
      "language_loss": 0.86712223,
      "learning_rate": 9.573617166463246e-07,
      "loss": 0.88893247,
      "num_input_tokens_seen": 122381680,
      "step": 5691,
      "time_per_iteration": 2.4692256450653076
    },
    {
      "auxiliary_loss_clip": 0.01142201,
      "auxiliary_loss_mlp": 0.01025309,
      "balance_loss_clip": 1.04382288,
      "balance_loss_mlp": 1.0188067,
      "epoch": 0.6844225335177058,
      "flos": 19969924037760.0,
      "grad_norm": 3.558523135121771,
      "language_loss": 0.60136461,
      "learning_rate": 9.56697051103924e-07,
      "loss": 0.62303966,
      "num_input_tokens_seen": 122399120,
      "step": 5692,
      "time_per_iteration": 2.4762392044067383
    },
    {
      "auxiliary_loss_clip": 0.01137827,
      "auxiliary_loss_mlp": 0.01026207,
      "balance_loss_clip": 1.04455853,
      "balance_loss_mlp": 1.01884627,
      "epoch": 0.6845427764083448,
      "flos": 25883334126720.0,
      "grad_norm": 2.2050805970703644,
      "language_loss": 0.81205273,
      "learning_rate": 9.560325438297522e-07,
      "loss": 0.83369303,
      "num_input_tokens_seen": 122417430,
      "step": 5693,
      "time_per_iteration": 2.4971835613250732
    },
    {
      "auxiliary_loss_clip": 0.01144783,
      "auxiliary_loss_mlp": 0.01023142,
      "balance_loss_clip": 1.05141211,
      "balance_loss_mlp": 1.01629364,
      "epoch": 0.684663019298984,
      "flos": 18880143356160.0,
      "grad_norm": 1.7168429276221087,
      "language_loss": 0.86630297,
      "learning_rate": 9.553681949246127e-07,
      "loss": 0.88798225,
      "num_input_tokens_seen": 122435055,
      "step": 5694,
      "time_per_iteration": 2.5113637447357178
    },
    {
      "auxiliary_loss_clip": 0.01130482,
      "auxiliary_loss_mlp": 0.01026273,
      "balance_loss_clip": 1.0452404,
      "balance_loss_mlp": 1.01836371,
      "epoch": 0.684783262189623,
      "flos": 54193725302400.0,
      "grad_norm": 1.962164242458883,
      "language_loss": 0.74914157,
      "learning_rate": 9.547040044892886e-07,
      "loss": 0.77070916,
      "num_input_tokens_seen": 122462570,
      "step": 5695,
      "time_per_iteration": 3.528773069381714
    },
    {
      "auxiliary_loss_clip": 0.01062137,
      "auxiliary_loss_mlp": 0.01000976,
      "balance_loss_clip": 1.01544106,
      "balance_loss_mlp": 0.9999153,
      "epoch": 0.6849035050802621,
      "flos": 63970264143360.0,
      "grad_norm": 0.8630307867586,
      "language_loss": 0.60133076,
      "learning_rate": 9.540399726245354e-07,
      "loss": 0.62196183,
      "num_input_tokens_seen": 122519275,
      "step": 5696,
      "time_per_iteration": 2.92171573638916
    },
    {
      "auxiliary_loss_clip": 0.01136223,
      "auxiliary_loss_mlp": 0.01026326,
      "balance_loss_clip": 1.04433775,
      "balance_loss_mlp": 1.01846409,
      "epoch": 0.6850237479709013,
      "flos": 25224121774080.0,
      "grad_norm": 1.6878599159076844,
      "language_loss": 0.68831736,
      "learning_rate": 9.533760994310859e-07,
      "loss": 0.70994282,
      "num_input_tokens_seen": 122539675,
      "step": 5697,
      "time_per_iteration": 2.5102057456970215
    },
    {
      "auxiliary_loss_clip": 0.0116935,
      "auxiliary_loss_mlp": 0.01024984,
      "balance_loss_clip": 1.04903245,
      "balance_loss_mlp": 1.01768196,
      "epoch": 0.6851439908615403,
      "flos": 19354128249600.0,
      "grad_norm": 2.040418355002968,
      "language_loss": 0.75328302,
      "learning_rate": 9.527123850096508e-07,
      "loss": 0.77522635,
      "num_input_tokens_seen": 122558035,
      "step": 5698,
      "time_per_iteration": 3.246387004852295
    },
    {
      "auxiliary_loss_clip": 0.01159818,
      "auxiliary_loss_mlp": 0.01022887,
      "balance_loss_clip": 1.04740846,
      "balance_loss_mlp": 1.01582348,
      "epoch": 0.6852642337521794,
      "flos": 23182133500800.0,
      "grad_norm": 1.8724907840886837,
      "language_loss": 0.71947205,
      "learning_rate": 9.520488294609142e-07,
      "loss": 0.74129915,
      "num_input_tokens_seen": 122576815,
      "step": 5699,
      "time_per_iteration": 2.447057008743286
    },
    {
      "auxiliary_loss_clip": 0.01029632,
      "auxiliary_loss_mlp": 0.01002322,
      "balance_loss_clip": 1.01675105,
      "balance_loss_mlp": 1.00125527,
      "epoch": 0.6853844766428185,
      "flos": 62647206583680.0,
      "grad_norm": 0.7445548108741515,
      "language_loss": 0.53871405,
      "learning_rate": 9.513854328855368e-07,
      "loss": 0.55903363,
      "num_input_tokens_seen": 122634690,
      "step": 5700,
      "time_per_iteration": 3.9129080772399902
    },
    {
      "auxiliary_loss_clip": 0.01167198,
      "auxiliary_loss_mlp": 0.01024318,
      "balance_loss_clip": 1.04985428,
      "balance_loss_mlp": 1.01730883,
      "epoch": 0.6855047195334576,
      "flos": 23437242869760.0,
      "grad_norm": 2.0281115486823693,
      "language_loss": 0.81323457,
      "learning_rate": 9.507221953841558e-07,
      "loss": 0.83514977,
      "num_input_tokens_seen": 122652320,
      "step": 5701,
      "time_per_iteration": 2.4152815341949463
    },
    {
      "auxiliary_loss_clip": 0.01158487,
      "auxiliary_loss_mlp": 0.01024381,
      "balance_loss_clip": 1.05084181,
      "balance_loss_mlp": 1.01668656,
      "epoch": 0.6856249624240967,
      "flos": 20664831530880.0,
      "grad_norm": 1.5102995450698231,
      "language_loss": 0.78113216,
      "learning_rate": 9.500591170573824e-07,
      "loss": 0.80296087,
      "num_input_tokens_seen": 122672340,
      "step": 5702,
      "time_per_iteration": 2.4542086124420166
    },
    {
      "auxiliary_loss_clip": 0.0110879,
      "auxiliary_loss_mlp": 0.01023902,
      "balance_loss_clip": 1.04199123,
      "balance_loss_mlp": 1.01666617,
      "epoch": 0.6857452053147358,
      "flos": 17087302794240.0,
      "grad_norm": 1.9622124326126513,
      "language_loss": 0.74071407,
      "learning_rate": 9.493961980058078e-07,
      "loss": 0.76204097,
      "num_input_tokens_seen": 122689935,
      "step": 5703,
      "time_per_iteration": 3.2859106063842773
    },
    {
      "auxiliary_loss_clip": 0.01083469,
      "auxiliary_loss_mlp": 0.0102494,
      "balance_loss_clip": 1.03926325,
      "balance_loss_mlp": 1.01776648,
      "epoch": 0.6858654482053749,
      "flos": 30847266057600.0,
      "grad_norm": 1.9613972706816656,
      "language_loss": 0.67700058,
      "learning_rate": 9.48733438329993e-07,
      "loss": 0.69808471,
      "num_input_tokens_seen": 122710200,
      "step": 5704,
      "time_per_iteration": 2.638317823410034
    },
    {
      "auxiliary_loss_clip": 0.01168297,
      "auxiliary_loss_mlp": 0.00762092,
      "balance_loss_clip": 1.05002117,
      "balance_loss_mlp": 1.0005399,
      "epoch": 0.6859856910960139,
      "flos": 28877314510080.0,
      "grad_norm": 1.7804988203379646,
      "language_loss": 0.74381697,
      "learning_rate": 9.480708381304807e-07,
      "loss": 0.76312089,
      "num_input_tokens_seen": 122731495,
      "step": 5705,
      "time_per_iteration": 2.4973254203796387
    },
    {
      "auxiliary_loss_clip": 0.01109901,
      "auxiliary_loss_mlp": 0.01028459,
      "balance_loss_clip": 1.04580879,
      "balance_loss_mlp": 1.02088904,
      "epoch": 0.6861059339866531,
      "flos": 19354523299200.0,
      "grad_norm": 2.3999947672265005,
      "language_loss": 0.83796859,
      "learning_rate": 9.474083975077858e-07,
      "loss": 0.85935223,
      "num_input_tokens_seen": 122748620,
      "step": 5706,
      "time_per_iteration": 2.5197863578796387
    },
    {
      "auxiliary_loss_clip": 0.01148622,
      "auxiliary_loss_mlp": 0.01022976,
      "balance_loss_clip": 1.04636383,
      "balance_loss_mlp": 1.01590347,
      "epoch": 0.6862261768772921,
      "flos": 22199976944640.0,
      "grad_norm": 9.835527263764096,
      "language_loss": 0.80111837,
      "learning_rate": 9.467461165623994e-07,
      "loss": 0.82283431,
      "num_input_tokens_seen": 122767670,
      "step": 5707,
      "time_per_iteration": 2.442981004714966
    },
    {
      "auxiliary_loss_clip": 0.01158062,
      "auxiliary_loss_mlp": 0.01022686,
      "balance_loss_clip": 1.04757833,
      "balance_loss_mlp": 1.01580453,
      "epoch": 0.6863464197679312,
      "flos": 26285677344000.0,
      "grad_norm": 2.144117148262155,
      "language_loss": 0.79662561,
      "learning_rate": 9.46083995394791e-07,
      "loss": 0.81843305,
      "num_input_tokens_seen": 122785480,
      "step": 5708,
      "time_per_iteration": 2.5060737133026123
    },
    {
      "auxiliary_loss_clip": 0.01155173,
      "auxiliary_loss_mlp": 0.00761378,
      "balance_loss_clip": 1.04716969,
      "balance_loss_mlp": 1.00051546,
      "epoch": 0.6864666626585703,
      "flos": 37815228564480.0,
      "grad_norm": 2.481138689271586,
      "language_loss": 0.63409549,
      "learning_rate": 9.454220341054012e-07,
      "loss": 0.65326095,
      "num_input_tokens_seen": 122810265,
      "step": 5709,
      "time_per_iteration": 2.598345994949341
    },
    {
      "auxiliary_loss_clip": 0.01123225,
      "auxiliary_loss_mlp": 0.01029761,
      "balance_loss_clip": 1.04328811,
      "balance_loss_mlp": 1.02247143,
      "epoch": 0.6865869055492094,
      "flos": 19391152193280.0,
      "grad_norm": 1.9070596138514655,
      "language_loss": 0.80716562,
      "learning_rate": 9.447602327946512e-07,
      "loss": 0.82869542,
      "num_input_tokens_seen": 122828905,
      "step": 5710,
      "time_per_iteration": 2.4986867904663086
    },
    {
      "auxiliary_loss_clip": 0.01139256,
      "auxiliary_loss_mlp": 0.01026968,
      "balance_loss_clip": 1.0451231,
      "balance_loss_mlp": 1.01943982,
      "epoch": 0.6867071484398485,
      "flos": 20375966355840.0,
      "grad_norm": 1.8066430502567807,
      "language_loss": 0.76586872,
      "learning_rate": 9.440985915629338e-07,
      "loss": 0.7875309,
      "num_input_tokens_seen": 122846235,
      "step": 5711,
      "time_per_iteration": 2.4718642234802246
    },
    {
      "auxiliary_loss_clip": 0.01168395,
      "auxiliary_loss_mlp": 0.01024389,
      "balance_loss_clip": 1.05066097,
      "balance_loss_mlp": 1.01735544,
      "epoch": 0.6868273913304875,
      "flos": 15889143801600.0,
      "grad_norm": 2.196203337506372,
      "language_loss": 0.72919637,
      "learning_rate": 9.434371105106223e-07,
      "loss": 0.7511242,
      "num_input_tokens_seen": 122863835,
      "step": 5712,
      "time_per_iteration": 2.3857431411743164
    },
    {
      "auxiliary_loss_clip": 0.0112347,
      "auxiliary_loss_mlp": 0.01026019,
      "balance_loss_clip": 1.04348886,
      "balance_loss_mlp": 1.01850295,
      "epoch": 0.6869476342211267,
      "flos": 24462492768000.0,
      "grad_norm": 1.7184899821878972,
      "language_loss": 0.70556158,
      "learning_rate": 9.427757897380602e-07,
      "loss": 0.72705644,
      "num_input_tokens_seen": 122883235,
      "step": 5713,
      "time_per_iteration": 2.5473453998565674
    },
    {
      "auxiliary_loss_clip": 0.0112479,
      "auxiliary_loss_mlp": 0.01022913,
      "balance_loss_clip": 1.04569185,
      "balance_loss_mlp": 1.01515865,
      "epoch": 0.6870678771117658,
      "flos": 18442571875200.0,
      "grad_norm": 2.286378730540375,
      "language_loss": 0.84600478,
      "learning_rate": 9.421146293455695e-07,
      "loss": 0.86748171,
      "num_input_tokens_seen": 122898975,
      "step": 5714,
      "time_per_iteration": 2.5033187866210938
    },
    {
      "auxiliary_loss_clip": 0.0113827,
      "auxiliary_loss_mlp": 0.0102412,
      "balance_loss_clip": 1.04431129,
      "balance_loss_mlp": 1.01693487,
      "epoch": 0.6871881200024048,
      "flos": 22200371994240.0,
      "grad_norm": 1.788128408753099,
      "language_loss": 0.68465519,
      "learning_rate": 9.414536294334489e-07,
      "loss": 0.70627904,
      "num_input_tokens_seen": 122918995,
      "step": 5715,
      "time_per_iteration": 2.5033304691314697
    },
    {
      "auxiliary_loss_clip": 0.01141822,
      "auxiliary_loss_mlp": 0.01025809,
      "balance_loss_clip": 1.04271805,
      "balance_loss_mlp": 1.01810789,
      "epoch": 0.687308362893044,
      "flos": 22127724737280.0,
      "grad_norm": 2.006388834957188,
      "language_loss": 0.6968714,
      "learning_rate": 9.407927901019708e-07,
      "loss": 0.7185477,
      "num_input_tokens_seen": 122938125,
      "step": 5716,
      "time_per_iteration": 2.4821105003356934
    },
    {
      "auxiliary_loss_clip": 0.01154673,
      "auxiliary_loss_mlp": 0.01023301,
      "balance_loss_clip": 1.047176,
      "balance_loss_mlp": 1.01627409,
      "epoch": 0.687428605783683,
      "flos": 25040546340480.0,
      "grad_norm": 2.7207932885444213,
      "language_loss": 0.76633137,
      "learning_rate": 9.401321114513854e-07,
      "loss": 0.78811109,
      "num_input_tokens_seen": 122957020,
      "step": 5717,
      "time_per_iteration": 2.4805705547332764
    },
    {
      "auxiliary_loss_clip": 0.01170348,
      "auxiliary_loss_mlp": 0.010273,
      "balance_loss_clip": 1.04962957,
      "balance_loss_mlp": 1.0198853,
      "epoch": 0.6875488486743221,
      "flos": 23770063313280.0,
      "grad_norm": 1.8699197166872108,
      "language_loss": 0.74892223,
      "learning_rate": 9.394715935819155e-07,
      "loss": 0.7708987,
      "num_input_tokens_seen": 122977410,
      "step": 5718,
      "time_per_iteration": 2.457313060760498
    },
    {
      "auxiliary_loss_clip": 0.01158266,
      "auxiliary_loss_mlp": 0.01027903,
      "balance_loss_clip": 1.04766273,
      "balance_loss_mlp": 1.0207293,
      "epoch": 0.6876690915649613,
      "flos": 25516937445120.0,
      "grad_norm": 3.5285869913030243,
      "language_loss": 0.61999249,
      "learning_rate": 9.388112365937608e-07,
      "loss": 0.64185417,
      "num_input_tokens_seen": 122996875,
      "step": 5719,
      "time_per_iteration": 2.4832680225372314
    },
    {
      "auxiliary_loss_clip": 0.01127298,
      "auxiliary_loss_mlp": 0.01022468,
      "balance_loss_clip": 1.04493904,
      "balance_loss_mlp": 1.014925,
      "epoch": 0.6877893344556003,
      "flos": 19427996568960.0,
      "grad_norm": 2.416596294166508,
      "language_loss": 0.82506192,
      "learning_rate": 9.381510405870985e-07,
      "loss": 0.84655952,
      "num_input_tokens_seen": 123015890,
      "step": 5720,
      "time_per_iteration": 2.497011661529541
    },
    {
      "auxiliary_loss_clip": 0.01154961,
      "auxiliary_loss_mlp": 0.01024888,
      "balance_loss_clip": 1.0477922,
      "balance_loss_mlp": 1.01751471,
      "epoch": 0.6879095773462394,
      "flos": 18661303745280.0,
      "grad_norm": 2.97680934093026,
      "language_loss": 0.77138931,
      "learning_rate": 9.374910056620791e-07,
      "loss": 0.79318774,
      "num_input_tokens_seen": 123034955,
      "step": 5721,
      "time_per_iteration": 2.4259798526763916
    },
    {
      "auxiliary_loss_clip": 0.01156637,
      "auxiliary_loss_mlp": 0.01028598,
      "balance_loss_clip": 1.04807711,
      "balance_loss_mlp": 1.02099884,
      "epoch": 0.6880298202368785,
      "flos": 20883132437760.0,
      "grad_norm": 2.298995768917676,
      "language_loss": 0.80937731,
      "learning_rate": 9.368311319188293e-07,
      "loss": 0.83122969,
      "num_input_tokens_seen": 123052770,
      "step": 5722,
      "time_per_iteration": 3.2340428829193115
    },
    {
      "auxiliary_loss_clip": 0.01126149,
      "auxiliary_loss_mlp": 0.01023562,
      "balance_loss_clip": 1.0438261,
      "balance_loss_mlp": 1.01648116,
      "epoch": 0.6881500631275176,
      "flos": 30153292318080.0,
      "grad_norm": 1.6718056376974701,
      "language_loss": 0.7943576,
      "learning_rate": 9.361714194574515e-07,
      "loss": 0.81585473,
      "num_input_tokens_seen": 123075105,
      "step": 5723,
      "time_per_iteration": 2.5868897438049316
    },
    {
      "auxiliary_loss_clip": 0.01071363,
      "auxiliary_loss_mlp": 0.01001711,
      "balance_loss_clip": 1.01588881,
      "balance_loss_mlp": 1.00064361,
      "epoch": 0.6882703060181566,
      "flos": 66181537215360.0,
      "grad_norm": 0.730133418288036,
      "language_loss": 0.58289182,
      "learning_rate": 9.355118683780228e-07,
      "loss": 0.60362256,
      "num_input_tokens_seen": 123145175,
      "step": 5724,
      "time_per_iteration": 3.102224349975586
    },
    {
      "auxiliary_loss_clip": 0.01167932,
      "auxiliary_loss_mlp": 0.01025853,
      "balance_loss_clip": 1.04799485,
      "balance_loss_mlp": 1.01851535,
      "epoch": 0.6883905489087958,
      "flos": 18214646123520.0,
      "grad_norm": 2.096248042270591,
      "language_loss": 0.79752886,
      "learning_rate": 9.348524787805987e-07,
      "loss": 0.81946671,
      "num_input_tokens_seen": 123160365,
      "step": 5725,
      "time_per_iteration": 3.5191938877105713
    },
    {
      "auxiliary_loss_clip": 0.01129102,
      "auxiliary_loss_mlp": 0.010215,
      "balance_loss_clip": 1.04172623,
      "balance_loss_mlp": 1.01447535,
      "epoch": 0.6885107917994349,
      "flos": 14056262553600.0,
      "grad_norm": 4.922032093977957,
      "language_loss": 0.8538748,
      "learning_rate": 9.341932507652053e-07,
      "loss": 0.87538081,
      "num_input_tokens_seen": 123174855,
      "step": 5726,
      "time_per_iteration": 2.519352436065674
    },
    {
      "auxiliary_loss_clip": 0.01138951,
      "auxiliary_loss_mlp": 0.01027875,
      "balance_loss_clip": 1.04224718,
      "balance_loss_mlp": 1.01955998,
      "epoch": 0.6886310346900739,
      "flos": 28690722334080.0,
      "grad_norm": 1.905341977876921,
      "language_loss": 0.78584218,
      "learning_rate": 9.335341844318489e-07,
      "loss": 0.80751044,
      "num_input_tokens_seen": 123194995,
      "step": 5727,
      "time_per_iteration": 3.290689468383789
    },
    {
      "auxiliary_loss_clip": 0.01140837,
      "auxiliary_loss_mlp": 0.01027219,
      "balance_loss_clip": 1.04652655,
      "balance_loss_mlp": 1.01994681,
      "epoch": 0.6887512775807131,
      "flos": 24535319592960.0,
      "grad_norm": 1.8871737465022373,
      "language_loss": 0.73136538,
      "learning_rate": 9.328752798805091e-07,
      "loss": 0.75304592,
      "num_input_tokens_seen": 123213465,
      "step": 5728,
      "time_per_iteration": 2.5016701221466064
    },
    {
      "auxiliary_loss_clip": 0.01154397,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.04627395,
      "balance_loss_mlp": 1.0199703,
      "epoch": 0.6888715204713521,
      "flos": 22414363269120.0,
      "grad_norm": 2.2402181236994965,
      "language_loss": 0.76077914,
      "learning_rate": 9.322165372111399e-07,
      "loss": 0.78259498,
      "num_input_tokens_seen": 123231610,
      "step": 5729,
      "time_per_iteration": 3.2580041885375977
    },
    {
      "auxiliary_loss_clip": 0.01123636,
      "auxiliary_loss_mlp": 0.0102226,
      "balance_loss_clip": 1.04590964,
      "balance_loss_mlp": 1.01513469,
      "epoch": 0.6889917633619912,
      "flos": 22054323294720.0,
      "grad_norm": 1.9574166982418773,
      "language_loss": 0.75711697,
      "learning_rate": 9.315579565236747e-07,
      "loss": 0.7785759,
      "num_input_tokens_seen": 123250715,
      "step": 5730,
      "time_per_iteration": 2.5145270824432373
    },
    {
      "auxiliary_loss_clip": 0.01140609,
      "auxiliary_loss_mlp": 0.01028075,
      "balance_loss_clip": 1.0505271,
      "balance_loss_mlp": 1.02004063,
      "epoch": 0.6891120062526304,
      "flos": 23949724164480.0,
      "grad_norm": 1.6966003496311692,
      "language_loss": 0.74528551,
      "learning_rate": 9.308995379180162e-07,
      "loss": 0.7669723,
      "num_input_tokens_seen": 123270270,
      "step": 5731,
      "time_per_iteration": 2.5157313346862793
    },
    {
      "auxiliary_loss_clip": 0.01061384,
      "auxiliary_loss_mlp": 0.0100106,
      "balance_loss_clip": 1.01502061,
      "balance_loss_mlp": 1.00006509,
      "epoch": 0.6892322491432694,
      "flos": 64117354337280.0,
      "grad_norm": 0.7454223445601498,
      "language_loss": 0.59533644,
      "learning_rate": 9.302412814940488e-07,
      "loss": 0.61596096,
      "num_input_tokens_seen": 123333045,
      "step": 5732,
      "time_per_iteration": 3.0954742431640625
    },
    {
      "auxiliary_loss_clip": 0.01138511,
      "auxiliary_loss_mlp": 0.01024738,
      "balance_loss_clip": 1.04325151,
      "balance_loss_mlp": 1.01687288,
      "epoch": 0.6893524920339085,
      "flos": 23002436736000.0,
      "grad_norm": 2.053368614831796,
      "language_loss": 0.71156657,
      "learning_rate": 9.295831873516276e-07,
      "loss": 0.73319906,
      "num_input_tokens_seen": 123352320,
      "step": 5733,
      "time_per_iteration": 2.5057895183563232
    },
    {
      "auxiliary_loss_clip": 0.01167235,
      "auxiliary_loss_mlp": 0.01025925,
      "balance_loss_clip": 1.0490464,
      "balance_loss_mlp": 1.01843834,
      "epoch": 0.6894727349245476,
      "flos": 21396260177280.0,
      "grad_norm": 1.7618763236820842,
      "language_loss": 0.7615453,
      "learning_rate": 9.289252555905873e-07,
      "loss": 0.78347695,
      "num_input_tokens_seen": 123372400,
      "step": 5734,
      "time_per_iteration": 2.433708667755127
    },
    {
      "auxiliary_loss_clip": 0.01155747,
      "auxiliary_loss_mlp": 0.0102275,
      "balance_loss_clip": 1.04883313,
      "balance_loss_mlp": 1.01529956,
      "epoch": 0.6895929778151867,
      "flos": 19865316654720.0,
      "grad_norm": 2.044291818992456,
      "language_loss": 0.76015317,
      "learning_rate": 9.282674863107334e-07,
      "loss": 0.7819382,
      "num_input_tokens_seen": 123390215,
      "step": 5735,
      "time_per_iteration": 2.4485487937927246
    },
    {
      "auxiliary_loss_clip": 0.01151006,
      "auxiliary_loss_mlp": 0.01027335,
      "balance_loss_clip": 1.04805565,
      "balance_loss_mlp": 1.02006578,
      "epoch": 0.6897132207058257,
      "flos": 18179166464640.0,
      "grad_norm": 2.3627322131421318,
      "language_loss": 0.7575624,
      "learning_rate": 9.276098796118488e-07,
      "loss": 0.77934575,
      "num_input_tokens_seen": 123406700,
      "step": 5736,
      "time_per_iteration": 2.4083616733551025
    },
    {
      "auxiliary_loss_clip": 0.01142096,
      "auxiliary_loss_mlp": 0.01024075,
      "balance_loss_clip": 1.04860759,
      "balance_loss_mlp": 1.01702714,
      "epoch": 0.6898334635964649,
      "flos": 32561641359360.0,
      "grad_norm": 1.7887429098307388,
      "language_loss": 0.66648114,
      "learning_rate": 9.269524355936938e-07,
      "loss": 0.6881429,
      "num_input_tokens_seen": 123429880,
      "step": 5737,
      "time_per_iteration": 2.567242383956909
    },
    {
      "auxiliary_loss_clip": 0.01135014,
      "auxiliary_loss_mlp": 0.01022752,
      "balance_loss_clip": 1.04405141,
      "balance_loss_mlp": 1.01578689,
      "epoch": 0.689953706487104,
      "flos": 22819004956800.0,
      "grad_norm": 1.7042803989756066,
      "language_loss": 0.84831685,
      "learning_rate": 9.262951543560002e-07,
      "loss": 0.8698945,
      "num_input_tokens_seen": 123449105,
      "step": 5738,
      "time_per_iteration": 2.5031094551086426
    },
    {
      "auxiliary_loss_clip": 0.01142445,
      "auxiliary_loss_mlp": 0.01031822,
      "balance_loss_clip": 1.05028462,
      "balance_loss_mlp": 1.02417505,
      "epoch": 0.690073949377743,
      "flos": 18515362786560.0,
      "grad_norm": 2.1839234862831414,
      "language_loss": 0.86339325,
      "learning_rate": 9.256380359984795e-07,
      "loss": 0.88513589,
      "num_input_tokens_seen": 123466215,
      "step": 5739,
      "time_per_iteration": 2.4452052116394043
    },
    {
      "auxiliary_loss_clip": 0.01116626,
      "auxiliary_loss_mlp": 0.01026186,
      "balance_loss_clip": 1.03886485,
      "balance_loss_mlp": 1.01899123,
      "epoch": 0.6901941922683821,
      "flos": 34857194716800.0,
      "grad_norm": 2.154249940638445,
      "language_loss": 0.75074643,
      "learning_rate": 9.249810806208139e-07,
      "loss": 0.77217454,
      "num_input_tokens_seen": 123485480,
      "step": 5740,
      "time_per_iteration": 2.664944648742676
    },
    {
      "auxiliary_loss_clip": 0.01109315,
      "auxiliary_loss_mlp": 0.00761566,
      "balance_loss_clip": 1.03845358,
      "balance_loss_mlp": 1.00059927,
      "epoch": 0.6903144351590212,
      "flos": 16253672976000.0,
      "grad_norm": 1.8945977996530499,
      "language_loss": 0.80364668,
      "learning_rate": 9.243242883226627e-07,
      "loss": 0.82235551,
      "num_input_tokens_seen": 123504575,
      "step": 5741,
      "time_per_iteration": 2.528092384338379
    },
    {
      "auxiliary_loss_clip": 0.01156939,
      "auxiliary_loss_mlp": 0.01028227,
      "balance_loss_clip": 1.04451704,
      "balance_loss_mlp": 1.02010906,
      "epoch": 0.6904346780496603,
      "flos": 28035137255040.0,
      "grad_norm": 2.8932824773119874,
      "language_loss": 0.69639575,
      "learning_rate": 9.236676592036628e-07,
      "loss": 0.71824741,
      "num_input_tokens_seen": 123524250,
      "step": 5742,
      "time_per_iteration": 2.5079591274261475
    },
    {
      "auxiliary_loss_clip": 0.01138907,
      "auxiliary_loss_mlp": 0.01023325,
      "balance_loss_clip": 1.04899776,
      "balance_loss_mlp": 1.01615787,
      "epoch": 0.6905549209402994,
      "flos": 23624266008960.0,
      "grad_norm": 1.7007250525889752,
      "language_loss": 0.73777354,
      "learning_rate": 9.230111933634228e-07,
      "loss": 0.75939584,
      "num_input_tokens_seen": 123545845,
      "step": 5743,
      "time_per_iteration": 2.5137429237365723
    },
    {
      "auxiliary_loss_clip": 0.01158403,
      "auxiliary_loss_mlp": 0.01022376,
      "balance_loss_clip": 1.05024183,
      "balance_loss_mlp": 1.0153966,
      "epoch": 0.6906751638309385,
      "flos": 23114945111040.0,
      "grad_norm": 1.6831624748905776,
      "language_loss": 0.80760562,
      "learning_rate": 9.223548909015288e-07,
      "loss": 0.82941341,
      "num_input_tokens_seen": 123567535,
      "step": 5744,
      "time_per_iteration": 2.487330436706543
    },
    {
      "auxiliary_loss_clip": 0.01103924,
      "auxiliary_loss_mlp": 0.01026927,
      "balance_loss_clip": 1.03928256,
      "balance_loss_mlp": 1.01975083,
      "epoch": 0.6907954067215776,
      "flos": 27305468375040.0,
      "grad_norm": 1.818234544003298,
      "language_loss": 0.72195995,
      "learning_rate": 9.216987519175407e-07,
      "loss": 0.74326849,
      "num_input_tokens_seen": 123587710,
      "step": 5745,
      "time_per_iteration": 2.6132993698120117
    },
    {
      "auxiliary_loss_clip": 0.01147898,
      "auxiliary_loss_mlp": 0.0102332,
      "balance_loss_clip": 1.04651523,
      "balance_loss_mlp": 1.01620674,
      "epoch": 0.6909156496122166,
      "flos": 21689399070720.0,
      "grad_norm": 1.7788387209152363,
      "language_loss": 0.6858052,
      "learning_rate": 9.210427765109942e-07,
      "loss": 0.70751739,
      "num_input_tokens_seen": 123607385,
      "step": 5746,
      "time_per_iteration": 2.438671112060547
    },
    {
      "auxiliary_loss_clip": 0.0113983,
      "auxiliary_loss_mlp": 0.01025206,
      "balance_loss_clip": 1.04378462,
      "balance_loss_mlp": 1.0173533,
      "epoch": 0.6910358925028558,
      "flos": 22561453463040.0,
      "grad_norm": 2.953691675612746,
      "language_loss": 0.81158715,
      "learning_rate": 9.20386964781402e-07,
      "loss": 0.83323753,
      "num_input_tokens_seen": 123625405,
      "step": 5747,
      "time_per_iteration": 2.4744808673858643
    },
    {
      "auxiliary_loss_clip": 0.01136451,
      "auxiliary_loss_mlp": 0.01023128,
      "balance_loss_clip": 1.04490495,
      "balance_loss_mlp": 1.01598191,
      "epoch": 0.6911561353934949,
      "flos": 22054107813120.0,
      "grad_norm": 1.849350440588227,
      "language_loss": 0.84274703,
      "learning_rate": 9.197313168282472e-07,
      "loss": 0.86434281,
      "num_input_tokens_seen": 123642850,
      "step": 5748,
      "time_per_iteration": 2.4859039783477783
    },
    {
      "auxiliary_loss_clip": 0.01148937,
      "auxiliary_loss_mlp": 0.01024622,
      "balance_loss_clip": 1.04379201,
      "balance_loss_mlp": 1.01743412,
      "epoch": 0.6912763782841339,
      "flos": 24206557386240.0,
      "grad_norm": 2.2064091290287178,
      "language_loss": 0.7202186,
      "learning_rate": 9.190758327509935e-07,
      "loss": 0.74195421,
      "num_input_tokens_seen": 123661595,
      "step": 5749,
      "time_per_iteration": 3.27254581451416
    },
    {
      "auxiliary_loss_clip": 0.01032514,
      "auxiliary_loss_mlp": 0.00752863,
      "balance_loss_clip": 1.01442003,
      "balance_loss_mlp": 0.99984324,
      "epoch": 0.6913966211747731,
      "flos": 52329641091840.0,
      "grad_norm": 0.932012425441419,
      "language_loss": 0.64492482,
      "learning_rate": 9.184205126490767e-07,
      "loss": 0.66277862,
      "num_input_tokens_seen": 123710490,
      "step": 5750,
      "time_per_iteration": 2.953484296798706
    },
    {
      "auxiliary_loss_clip": 0.01038958,
      "auxiliary_loss_mlp": 0.00752811,
      "balance_loss_clip": 1.0143528,
      "balance_loss_mlp": 0.99991363,
      "epoch": 0.6915168640654121,
      "flos": 66741274851840.0,
      "grad_norm": 1.0796555288360798,
      "language_loss": 0.59702909,
      "learning_rate": 9.177653566219075e-07,
      "loss": 0.61494684,
      "num_input_tokens_seen": 123765215,
      "step": 5751,
      "time_per_iteration": 3.8127617835998535
    },
    {
      "auxiliary_loss_clip": 0.01129051,
      "auxiliary_loss_mlp": 0.01027543,
      "balance_loss_clip": 1.04291868,
      "balance_loss_mlp": 1.02030134,
      "epoch": 0.6916371069560512,
      "flos": 18296523175680.0,
      "grad_norm": 2.7052514069274753,
      "language_loss": 0.762353,
      "learning_rate": 9.171103647688744e-07,
      "loss": 0.78391892,
      "num_input_tokens_seen": 123783955,
      "step": 5752,
      "time_per_iteration": 2.5005910396575928
    },
    {
      "auxiliary_loss_clip": 0.01075087,
      "auxiliary_loss_mlp": 0.01024856,
      "balance_loss_clip": 1.03978825,
      "balance_loss_mlp": 1.01806748,
      "epoch": 0.6917573498466904,
      "flos": 19645794685440.0,
      "grad_norm": 1.930178206733818,
      "language_loss": 0.68999827,
      "learning_rate": 9.164555371893367e-07,
      "loss": 0.7109977,
      "num_input_tokens_seen": 123803885,
      "step": 5753,
      "time_per_iteration": 2.6045751571655273
    },
    {
      "auxiliary_loss_clip": 0.01155808,
      "auxiliary_loss_mlp": 0.00761603,
      "balance_loss_clip": 1.04900646,
      "balance_loss_mlp": 1.00051844,
      "epoch": 0.6918775927373294,
      "flos": 14210319985920.0,
      "grad_norm": 2.094529852681535,
      "language_loss": 0.7520963,
      "learning_rate": 9.158008739826333e-07,
      "loss": 0.77127039,
      "num_input_tokens_seen": 123821485,
      "step": 5754,
      "time_per_iteration": 3.257155418395996
    },
    {
      "auxiliary_loss_clip": 0.01138581,
      "auxiliary_loss_mlp": 0.01027271,
      "balance_loss_clip": 1.04726052,
      "balance_loss_mlp": 1.01980901,
      "epoch": 0.6919978356279685,
      "flos": 23985455218560.0,
      "grad_norm": 1.5857175985167324,
      "language_loss": 0.86595976,
      "learning_rate": 9.151463752480744e-07,
      "loss": 0.8876183,
      "num_input_tokens_seen": 123840215,
      "step": 5755,
      "time_per_iteration": 2.501033306121826
    },
    {
      "auxiliary_loss_clip": 0.01115205,
      "auxiliary_loss_mlp": 0.01027124,
      "balance_loss_clip": 1.04217076,
      "balance_loss_mlp": 1.01991141,
      "epoch": 0.6921180785186076,
      "flos": 23622937205760.0,
      "grad_norm": 1.6344181757249778,
      "language_loss": 0.80368906,
      "learning_rate": 9.144920410849493e-07,
      "loss": 0.82511234,
      "num_input_tokens_seen": 123861450,
      "step": 5756,
      "time_per_iteration": 3.2965164184570312
    },
    {
      "auxiliary_loss_clip": 0.01144526,
      "auxiliary_loss_mlp": 0.01026503,
      "balance_loss_clip": 1.04586649,
      "balance_loss_mlp": 1.01920414,
      "epoch": 0.6922383214092467,
      "flos": 21142623265920.0,
      "grad_norm": 1.6706082176997474,
      "language_loss": 0.80647415,
      "learning_rate": 9.138378715925176e-07,
      "loss": 0.82818443,
      "num_input_tokens_seen": 123880545,
      "step": 5757,
      "time_per_iteration": 2.5535547733306885
    },
    {
      "auxiliary_loss_clip": 0.01134742,
      "auxiliary_loss_mlp": 0.010227,
      "balance_loss_clip": 1.04456425,
      "balance_loss_mlp": 1.01561022,
      "epoch": 0.6923585642998857,
      "flos": 21470667200640.0,
      "grad_norm": 1.5756804331649827,
      "language_loss": 0.80963147,
      "learning_rate": 9.131838668700167e-07,
      "loss": 0.8312059,
      "num_input_tokens_seen": 123900615,
      "step": 5758,
      "time_per_iteration": 2.4891936779022217
    },
    {
      "auxiliary_loss_clip": 0.01126833,
      "auxiliary_loss_mlp": 0.01024972,
      "balance_loss_clip": 1.04350471,
      "balance_loss_mlp": 1.01792026,
      "epoch": 0.6924788071905249,
      "flos": 21105204272640.0,
      "grad_norm": 1.9790880702226041,
      "language_loss": 0.86584604,
      "learning_rate": 9.125300270166598e-07,
      "loss": 0.88736403,
      "num_input_tokens_seen": 123921220,
      "step": 5759,
      "time_per_iteration": 2.544259548187256
    },
    {
      "auxiliary_loss_clip": 0.01133819,
      "auxiliary_loss_mlp": 0.01020283,
      "balance_loss_clip": 1.04431319,
      "balance_loss_mlp": 1.01279998,
      "epoch": 0.692599050081164,
      "flos": 26250018117120.0,
      "grad_norm": 1.7142491767918437,
      "language_loss": 0.85845792,
      "learning_rate": 9.118763521316324e-07,
      "loss": 0.87999892,
      "num_input_tokens_seen": 123941795,
      "step": 5760,
      "time_per_iteration": 2.5486650466918945
    },
    {
      "auxiliary_loss_clip": 0.01168502,
      "auxiliary_loss_mlp": 0.00762082,
      "balance_loss_clip": 1.04765296,
      "balance_loss_mlp": 1.00051713,
      "epoch": 0.692719292971803,
      "flos": 20885215426560.0,
      "grad_norm": 2.4385243358059463,
      "language_loss": 0.76098591,
      "learning_rate": 9.112228423140987e-07,
      "loss": 0.7802918,
      "num_input_tokens_seen": 123960715,
      "step": 5761,
      "time_per_iteration": 2.4318244457244873
    },
    {
      "auxiliary_loss_clip": 0.011448,
      "auxiliary_loss_mlp": 0.01030711,
      "balance_loss_clip": 1.04634619,
      "balance_loss_mlp": 1.02286148,
      "epoch": 0.6928395358624422,
      "flos": 25921938268800.0,
      "grad_norm": 6.885428835020421,
      "language_loss": 0.86322641,
      "learning_rate": 9.105694976631932e-07,
      "loss": 0.88498151,
      "num_input_tokens_seen": 123978625,
      "step": 5762,
      "time_per_iteration": 2.5106899738311768
    },
    {
      "auxiliary_loss_clip": 0.01153497,
      "auxiliary_loss_mlp": 0.01028305,
      "balance_loss_clip": 1.04892206,
      "balance_loss_mlp": 1.02087808,
      "epoch": 0.6929597787530812,
      "flos": 23586559706880.0,
      "grad_norm": 2.375343916997738,
      "language_loss": 0.72661793,
      "learning_rate": 9.099163182780283e-07,
      "loss": 0.74843597,
      "num_input_tokens_seen": 123996780,
      "step": 5763,
      "time_per_iteration": 2.468513011932373
    },
    {
      "auxiliary_loss_clip": 0.01136672,
      "auxiliary_loss_mlp": 0.01027017,
      "balance_loss_clip": 1.04610741,
      "balance_loss_mlp": 1.01919961,
      "epoch": 0.6930800216437203,
      "flos": 18255656476800.0,
      "grad_norm": 3.429194481011056,
      "language_loss": 0.49409071,
      "learning_rate": 9.092633042576916e-07,
      "loss": 0.51572758,
      "num_input_tokens_seen": 124014045,
      "step": 5764,
      "time_per_iteration": 2.4508748054504395
    },
    {
      "auxiliary_loss_clip": 0.0113584,
      "auxiliary_loss_mlp": 0.01027504,
      "balance_loss_clip": 1.04598045,
      "balance_loss_mlp": 1.02026212,
      "epoch": 0.6932002645343595,
      "flos": 29168621809920.0,
      "grad_norm": 1.7174633994847568,
      "language_loss": 0.56394851,
      "learning_rate": 9.086104557012446e-07,
      "loss": 0.58558202,
      "num_input_tokens_seen": 124034615,
      "step": 5765,
      "time_per_iteration": 2.5372023582458496
    },
    {
      "auxiliary_loss_clip": 0.01145246,
      "auxiliary_loss_mlp": 0.01021772,
      "balance_loss_clip": 1.04603648,
      "balance_loss_mlp": 1.01473331,
      "epoch": 0.6933205074249985,
      "flos": 23842746483840.0,
      "grad_norm": 1.8192545164636547,
      "language_loss": 0.65292811,
      "learning_rate": 9.079577727077239e-07,
      "loss": 0.67459834,
      "num_input_tokens_seen": 124053445,
      "step": 5766,
      "time_per_iteration": 2.469566822052002
    },
    {
      "auxiliary_loss_clip": 0.01156079,
      "auxiliary_loss_mlp": 0.01028968,
      "balance_loss_clip": 1.04863262,
      "balance_loss_mlp": 1.02127945,
      "epoch": 0.6934407503156376,
      "flos": 24166696268160.0,
      "grad_norm": 3.126753868699411,
      "language_loss": 0.71937448,
      "learning_rate": 9.073052553761404e-07,
      "loss": 0.741225,
      "num_input_tokens_seen": 124072810,
      "step": 5767,
      "time_per_iteration": 2.4635894298553467
    },
    {
      "auxiliary_loss_clip": 0.01116461,
      "auxiliary_loss_mlp": 0.01024929,
      "balance_loss_clip": 1.04527593,
      "balance_loss_mlp": 1.01679897,
      "epoch": 0.6935609932062767,
      "flos": 20631327120000.0,
      "grad_norm": 1.6614036675546122,
      "language_loss": 0.78268957,
      "learning_rate": 9.066529038054805e-07,
      "loss": 0.80410349,
      "num_input_tokens_seen": 124092875,
      "step": 5768,
      "time_per_iteration": 2.5495917797088623
    },
    {
      "auxiliary_loss_clip": 0.01138751,
      "auxiliary_loss_mlp": 0.01022071,
      "balance_loss_clip": 1.04660332,
      "balance_loss_mlp": 1.01490331,
      "epoch": 0.6936812360969158,
      "flos": 18254184019200.0,
      "grad_norm": 1.815154152159682,
      "language_loss": 0.74178845,
      "learning_rate": 9.060007180947071e-07,
      "loss": 0.76339662,
      "num_input_tokens_seen": 124110930,
      "step": 5769,
      "time_per_iteration": 2.4599761962890625
    },
    {
      "auxiliary_loss_clip": 0.01111126,
      "auxiliary_loss_mlp": 0.01027169,
      "balance_loss_clip": 1.03906393,
      "balance_loss_mlp": 1.01951933,
      "epoch": 0.6938014789875548,
      "flos": 31317336368640.0,
      "grad_norm": 1.851476358020392,
      "language_loss": 0.73212183,
      "learning_rate": 9.053486983427534e-07,
      "loss": 0.75350475,
      "num_input_tokens_seen": 124132180,
      "step": 5770,
      "time_per_iteration": 2.6224005222320557
    },
    {
      "auxiliary_loss_clip": 0.01142913,
      "auxiliary_loss_mlp": 0.01026416,
      "balance_loss_clip": 1.04412901,
      "balance_loss_mlp": 1.01899552,
      "epoch": 0.6939217218781939,
      "flos": 17528429721600.0,
      "grad_norm": 2.009702779252152,
      "language_loss": 0.7055375,
      "learning_rate": 9.046968446485326e-07,
      "loss": 0.72723079,
      "num_input_tokens_seen": 124150585,
      "step": 5771,
      "time_per_iteration": 2.468140125274658
    },
    {
      "auxiliary_loss_clip": 0.01157962,
      "auxiliary_loss_mlp": 0.01028684,
      "balance_loss_clip": 1.04979205,
      "balance_loss_mlp": 1.02054238,
      "epoch": 0.6940419647688331,
      "flos": 18551776199040.0,
      "grad_norm": 8.092004430525884,
      "language_loss": 0.70643914,
      "learning_rate": 9.040451571109295e-07,
      "loss": 0.72830558,
      "num_input_tokens_seen": 124166205,
      "step": 5772,
      "time_per_iteration": 2.4313132762908936
    },
    {
      "auxiliary_loss_clip": 0.01040063,
      "auxiliary_loss_mlp": 0.01002626,
      "balance_loss_clip": 1.02045226,
      "balance_loss_mlp": 1.00154102,
      "epoch": 0.6941622076594721,
      "flos": 66926286829440.0,
      "grad_norm": 0.832527170160163,
      "language_loss": 0.6037572,
      "learning_rate": 9.033936358288042e-07,
      "loss": 0.62418413,
      "num_input_tokens_seen": 124219940,
      "step": 5773,
      "time_per_iteration": 2.988834857940674
    },
    {
      "auxiliary_loss_clip": 0.01170381,
      "auxiliary_loss_mlp": 0.01016035,
      "balance_loss_clip": 1.04958415,
      "balance_loss_mlp": 1.00869155,
      "epoch": 0.6942824505501112,
      "flos": 26578062051840.0,
      "grad_norm": 1.6914707995046343,
      "language_loss": 0.82178885,
      "learning_rate": 9.027422809009937e-07,
      "loss": 0.84365302,
      "num_input_tokens_seen": 124239885,
      "step": 5774,
      "time_per_iteration": 2.4825809001922607
    },
    {
      "auxiliary_loss_clip": 0.01155152,
      "auxiliary_loss_mlp": 0.0102025,
      "balance_loss_clip": 1.04539442,
      "balance_loss_mlp": 1.01265001,
      "epoch": 0.6944026934407503,
      "flos": 21248308056960.0,
      "grad_norm": 3.8448102472157353,
      "language_loss": 0.83407629,
      "learning_rate": 9.020910924263054e-07,
      "loss": 0.85583031,
      "num_input_tokens_seen": 124258410,
      "step": 5775,
      "time_per_iteration": 2.4506568908691406
    },
    {
      "auxiliary_loss_clip": 0.0103796,
      "auxiliary_loss_mlp": 0.01003887,
      "balance_loss_clip": 1.01909065,
      "balance_loss_mlp": 1.0028199,
      "epoch": 0.6945229363313894,
      "flos": 70677191537280.0,
      "grad_norm": 0.8132748977895188,
      "language_loss": 0.58186227,
      "learning_rate": 9.014400705035261e-07,
      "loss": 0.60228074,
      "num_input_tokens_seen": 124315315,
      "step": 5776,
      "time_per_iteration": 3.81909441947937
    },
    {
      "auxiliary_loss_clip": 0.01166827,
      "auxiliary_loss_mlp": 0.01022395,
      "balance_loss_clip": 1.0503571,
      "balance_loss_mlp": 1.01520073,
      "epoch": 0.6946431792220285,
      "flos": 18952934267520.0,
      "grad_norm": 1.9368996722258527,
      "language_loss": 0.76930261,
      "learning_rate": 9.00789215231414e-07,
      "loss": 0.79119486,
      "num_input_tokens_seen": 124333710,
      "step": 5777,
      "time_per_iteration": 3.235917568206787
    },
    {
      "auxiliary_loss_clip": 0.0112462,
      "auxiliary_loss_mlp": 0.00762161,
      "balance_loss_clip": 1.0406965,
      "balance_loss_mlp": 1.00055563,
      "epoch": 0.6947634221126676,
      "flos": 20338834671360.0,
      "grad_norm": 1.7351386482825102,
      "language_loss": 0.81811237,
      "learning_rate": 9.001385267087056e-07,
      "loss": 0.83698022,
      "num_input_tokens_seen": 124352855,
      "step": 5778,
      "time_per_iteration": 2.556151866912842
    },
    {
      "auxiliary_loss_clip": 0.01157447,
      "auxiliary_loss_mlp": 0.01022054,
      "balance_loss_clip": 1.04842675,
      "balance_loss_mlp": 1.01494598,
      "epoch": 0.6948836650033067,
      "flos": 21833723917440.0,
      "grad_norm": 1.4755900027901134,
      "language_loss": 0.70272839,
      "learning_rate": 8.994880050341072e-07,
      "loss": 0.72452343,
      "num_input_tokens_seen": 124372960,
      "step": 5779,
      "time_per_iteration": 2.4567015171051025
    },
    {
      "auxiliary_loss_clip": 0.01135167,
      "auxiliary_loss_mlp": 0.01034608,
      "balance_loss_clip": 1.04640913,
      "balance_loss_mlp": 1.0272944,
      "epoch": 0.6950039078939457,
      "flos": 23657519024640.0,
      "grad_norm": 2.1235507986752165,
      "language_loss": 0.77724135,
      "learning_rate": 8.988376503063026e-07,
      "loss": 0.79893911,
      "num_input_tokens_seen": 124394220,
      "step": 5780,
      "time_per_iteration": 3.2974607944488525
    },
    {
      "auxiliary_loss_clip": 0.01122355,
      "auxiliary_loss_mlp": 0.01026767,
      "balance_loss_clip": 1.04586256,
      "balance_loss_mlp": 1.01874423,
      "epoch": 0.6951241507845849,
      "flos": 21792462168960.0,
      "grad_norm": 1.755732389878013,
      "language_loss": 0.8114289,
      "learning_rate": 8.981874626239521e-07,
      "loss": 0.83292007,
      "num_input_tokens_seen": 124412795,
      "step": 5781,
      "time_per_iteration": 2.547027349472046
    },
    {
      "auxiliary_loss_clip": 0.01155894,
      "auxiliary_loss_mlp": 0.0103125,
      "balance_loss_clip": 1.04939628,
      "balance_loss_mlp": 1.02349496,
      "epoch": 0.695244393675224,
      "flos": 14647568244480.0,
      "grad_norm": 2.431478494265423,
      "language_loss": 0.8810094,
      "learning_rate": 8.975374420856872e-07,
      "loss": 0.90288079,
      "num_input_tokens_seen": 124429690,
      "step": 5782,
      "time_per_iteration": 2.413693428039551
    },
    {
      "auxiliary_loss_clip": 0.01116878,
      "auxiliary_loss_mlp": 0.01021501,
      "balance_loss_clip": 1.04174519,
      "balance_loss_mlp": 1.0145278,
      "epoch": 0.695364636565863,
      "flos": 16873203778560.0,
      "grad_norm": 4.236207942755898,
      "language_loss": 0.72964561,
      "learning_rate": 8.968875887901157e-07,
      "loss": 0.75102937,
      "num_input_tokens_seen": 124447070,
      "step": 5783,
      "time_per_iteration": 3.23355770111084
    },
    {
      "auxiliary_loss_clip": 0.011393,
      "auxiliary_loss_mlp": 0.0102304,
      "balance_loss_clip": 1.04344916,
      "balance_loss_mlp": 1.01535654,
      "epoch": 0.6954848794565022,
      "flos": 19354523299200.0,
      "grad_norm": 1.9328052486000094,
      "language_loss": 0.63046283,
      "learning_rate": 8.9623790283582e-07,
      "loss": 0.6520862,
      "num_input_tokens_seen": 124464950,
      "step": 5784,
      "time_per_iteration": 2.4698426723480225
    },
    {
      "auxiliary_loss_clip": 0.0112813,
      "auxiliary_loss_mlp": 0.01029524,
      "balance_loss_clip": 1.04537439,
      "balance_loss_mlp": 1.02193046,
      "epoch": 0.6956051223471412,
      "flos": 18990209606400.0,
      "grad_norm": 2.340605641380878,
      "language_loss": 0.76530075,
      "learning_rate": 8.955883843213561e-07,
      "loss": 0.78687739,
      "num_input_tokens_seen": 124483965,
      "step": 5785,
      "time_per_iteration": 2.511528253555298
    },
    {
      "auxiliary_loss_clip": 0.01160859,
      "auxiliary_loss_mlp": 0.01027418,
      "balance_loss_clip": 1.04810762,
      "balance_loss_mlp": 1.01949668,
      "epoch": 0.6957253652377803,
      "flos": 16107229226880.0,
      "grad_norm": 1.8928675731567448,
      "language_loss": 0.86788189,
      "learning_rate": 8.949390333452569e-07,
      "loss": 0.88976467,
      "num_input_tokens_seen": 124501910,
      "step": 5786,
      "time_per_iteration": 2.4222774505615234
    },
    {
      "auxiliary_loss_clip": 0.01167613,
      "auxiliary_loss_mlp": 0.01025162,
      "balance_loss_clip": 1.04974055,
      "balance_loss_mlp": 1.01804829,
      "epoch": 0.6958456081284194,
      "flos": 29388646569600.0,
      "grad_norm": 1.6813816244161495,
      "language_loss": 0.67475069,
      "learning_rate": 8.942898500060279e-07,
      "loss": 0.69667846,
      "num_input_tokens_seen": 124521625,
      "step": 5787,
      "time_per_iteration": 2.4769599437713623
    },
    {
      "auxiliary_loss_clip": 0.01118087,
      "auxiliary_loss_mlp": 0.01023143,
      "balance_loss_clip": 1.04379165,
      "balance_loss_mlp": 1.01562631,
      "epoch": 0.6959658510190585,
      "flos": 25154850395520.0,
      "grad_norm": 2.777945724328948,
      "language_loss": 0.71817529,
      "learning_rate": 8.936408344021493e-07,
      "loss": 0.73958755,
      "num_input_tokens_seen": 124538540,
      "step": 5788,
      "time_per_iteration": 2.5563461780548096
    },
    {
      "auxiliary_loss_clip": 0.01150771,
      "auxiliary_loss_mlp": 0.01029629,
      "balance_loss_clip": 1.04954803,
      "balance_loss_mlp": 1.02118313,
      "epoch": 0.6960860939096976,
      "flos": 42814388759040.0,
      "grad_norm": 2.1689027919672967,
      "language_loss": 0.71130282,
      "learning_rate": 8.929919866320765e-07,
      "loss": 0.73310685,
      "num_input_tokens_seen": 124559355,
      "step": 5789,
      "time_per_iteration": 2.675229549407959
    },
    {
      "auxiliary_loss_clip": 0.0113079,
      "auxiliary_loss_mlp": 0.00762327,
      "balance_loss_clip": 1.04247475,
      "balance_loss_mlp": 1.00050497,
      "epoch": 0.6962063368003367,
      "flos": 17566566986880.0,
      "grad_norm": 1.8505713536597757,
      "language_loss": 0.81092548,
      "learning_rate": 8.923433067942385e-07,
      "loss": 0.82985663,
      "num_input_tokens_seen": 124577920,
      "step": 5790,
      "time_per_iteration": 2.5029871463775635
    },
    {
      "auxiliary_loss_clip": 0.01134611,
      "auxiliary_loss_mlp": 0.01027165,
      "balance_loss_clip": 1.04621625,
      "balance_loss_mlp": 1.02006578,
      "epoch": 0.6963265796909758,
      "flos": 21251648021760.0,
      "grad_norm": 1.854768305622176,
      "language_loss": 0.68871766,
      "learning_rate": 8.916947949870417e-07,
      "loss": 0.71033543,
      "num_input_tokens_seen": 124597585,
      "step": 5791,
      "time_per_iteration": 2.5244009494781494
    },
    {
      "auxiliary_loss_clip": 0.0106135,
      "auxiliary_loss_mlp": 0.01002651,
      "balance_loss_clip": 1.01484227,
      "balance_loss_mlp": 1.00163805,
      "epoch": 0.6964468225816148,
      "flos": 68828295801600.0,
      "grad_norm": 0.750887477114594,
      "language_loss": 0.58145362,
      "learning_rate": 8.910464513088615e-07,
      "loss": 0.60209364,
      "num_input_tokens_seen": 124661625,
      "step": 5792,
      "time_per_iteration": 3.1116039752960205
    },
    {
      "auxiliary_loss_clip": 0.01133065,
      "auxiliary_loss_mlp": 0.01022869,
      "balance_loss_clip": 1.0440166,
      "balance_loss_mlp": 1.0150156,
      "epoch": 0.696567065472254,
      "flos": 18950887192320.0,
      "grad_norm": 5.252957896290847,
      "language_loss": 0.7840808,
      "learning_rate": 8.903982758580542e-07,
      "loss": 0.8056401,
      "num_input_tokens_seen": 124680565,
      "step": 5793,
      "time_per_iteration": 2.513719320297241
    },
    {
      "auxiliary_loss_clip": 0.01138294,
      "auxiliary_loss_mlp": 0.01032992,
      "balance_loss_clip": 1.046345,
      "balance_loss_mlp": 1.02561283,
      "epoch": 0.696687308362893,
      "flos": 22856675345280.0,
      "grad_norm": 1.9795699831082307,
      "language_loss": 0.80265558,
      "learning_rate": 8.897502687329457e-07,
      "loss": 0.82436848,
      "num_input_tokens_seen": 124700365,
      "step": 5794,
      "time_per_iteration": 2.4889466762542725
    },
    {
      "auxiliary_loss_clip": 0.01122406,
      "auxiliary_loss_mlp": 0.0102249,
      "balance_loss_clip": 1.04350555,
      "balance_loss_mlp": 1.0154984,
      "epoch": 0.6968075512535321,
      "flos": 24972926987520.0,
      "grad_norm": 2.0478345205776205,
      "language_loss": 0.79856312,
      "learning_rate": 8.891024300318382e-07,
      "loss": 0.82001209,
      "num_input_tokens_seen": 124718935,
      "step": 5795,
      "time_per_iteration": 2.589400291442871
    },
    {
      "auxiliary_loss_clip": 0.01117054,
      "auxiliary_loss_mlp": 0.01023169,
      "balance_loss_clip": 1.04173446,
      "balance_loss_mlp": 1.01642752,
      "epoch": 0.6969277941441713,
      "flos": 21030438113280.0,
      "grad_norm": 2.7166445141360644,
      "language_loss": 0.75870109,
      "learning_rate": 8.884547598530103e-07,
      "loss": 0.78010333,
      "num_input_tokens_seen": 124739505,
      "step": 5796,
      "time_per_iteration": 2.5714428424835205
    },
    {
      "auxiliary_loss_clip": 0.01071477,
      "auxiliary_loss_mlp": 0.01030487,
      "balance_loss_clip": 1.03705394,
      "balance_loss_mlp": 1.02289915,
      "epoch": 0.6970480370348103,
      "flos": 21579404647680.0,
      "grad_norm": 1.8499792818007779,
      "language_loss": 0.75178993,
      "learning_rate": 8.8780725829471e-07,
      "loss": 0.77280957,
      "num_input_tokens_seen": 124757410,
      "step": 5797,
      "time_per_iteration": 2.6393935680389404
    },
    {
      "auxiliary_loss_clip": 0.01168999,
      "auxiliary_loss_mlp": 0.01027174,
      "balance_loss_clip": 1.0486027,
      "balance_loss_mlp": 1.01946163,
      "epoch": 0.6971682799254494,
      "flos": 22419175691520.0,
      "grad_norm": 1.9340346040666845,
      "language_loss": 0.78222102,
      "learning_rate": 8.87159925455165e-07,
      "loss": 0.80418277,
      "num_input_tokens_seen": 124777240,
      "step": 5798,
      "time_per_iteration": 2.430356979370117
    },
    {
      "auxiliary_loss_clip": 0.01122592,
      "auxiliary_loss_mlp": 0.01027879,
      "balance_loss_clip": 1.04484296,
      "balance_loss_mlp": 1.02086091,
      "epoch": 0.6972885228160886,
      "flos": 20005834659840.0,
      "grad_norm": 2.0074781583069417,
      "language_loss": 0.73301578,
      "learning_rate": 8.865127614325738e-07,
      "loss": 0.75452042,
      "num_input_tokens_seen": 124795670,
      "step": 5799,
      "time_per_iteration": 2.520542621612549
    },
    {
      "auxiliary_loss_clip": 0.01132629,
      "auxiliary_loss_mlp": 0.0102719,
      "balance_loss_clip": 1.04342294,
      "balance_loss_mlp": 1.01954317,
      "epoch": 0.6974087657067276,
      "flos": 37853437656960.0,
      "grad_norm": 1.983450800859278,
      "language_loss": 0.66537815,
      "learning_rate": 8.85865766325113e-07,
      "loss": 0.68697637,
      "num_input_tokens_seen": 124819600,
      "step": 5800,
      "time_per_iteration": 2.626162528991699
    },
    {
      "auxiliary_loss_clip": 0.01136049,
      "auxiliary_loss_mlp": 0.01025025,
      "balance_loss_clip": 1.04439139,
      "balance_loss_mlp": 1.01776469,
      "epoch": 0.6975290085973667,
      "flos": 29489267543040.0,
      "grad_norm": 2.192836506003847,
      "language_loss": 0.72344482,
      "learning_rate": 8.852189402309287e-07,
      "loss": 0.74505556,
      "num_input_tokens_seen": 124838785,
      "step": 5801,
      "time_per_iteration": 2.5396711826324463
    },
    {
      "auxiliary_loss_clip": 0.0115447,
      "auxiliary_loss_mlp": 0.01025812,
      "balance_loss_clip": 1.04866207,
      "balance_loss_mlp": 1.01889789,
      "epoch": 0.6976492514880057,
      "flos": 12895630295040.0,
      "grad_norm": 2.2069679831906193,
      "language_loss": 0.74260998,
      "learning_rate": 8.845722832481441e-07,
      "loss": 0.76441282,
      "num_input_tokens_seen": 124854215,
      "step": 5802,
      "time_per_iteration": 3.125596761703491
    },
    {
      "auxiliary_loss_clip": 0.01153587,
      "auxiliary_loss_mlp": 0.01024297,
      "balance_loss_clip": 1.04753709,
      "balance_loss_mlp": 1.01719534,
      "epoch": 0.6977694943786449,
      "flos": 24352929308160.0,
      "grad_norm": 1.85412657963025,
      "language_loss": 0.77529085,
      "learning_rate": 8.83925795474858e-07,
      "loss": 0.79706967,
      "num_input_tokens_seen": 124874340,
      "step": 5803,
      "time_per_iteration": 2.491938352584839
    },
    {
      "auxiliary_loss_clip": 0.01122467,
      "auxiliary_loss_mlp": 0.0102388,
      "balance_loss_clip": 1.04618394,
      "balance_loss_mlp": 1.01614308,
      "epoch": 0.6978897372692839,
      "flos": 29898470257920.0,
      "grad_norm": 3.696652994148292,
      "language_loss": 0.58865392,
      "learning_rate": 8.832794770091414e-07,
      "loss": 0.61011744,
      "num_input_tokens_seen": 124895175,
      "step": 5804,
      "time_per_iteration": 3.4000065326690674
    },
    {
      "auxiliary_loss_clip": 0.01144866,
      "auxiliary_loss_mlp": 0.01025692,
      "balance_loss_clip": 1.04633486,
      "balance_loss_mlp": 1.01811647,
      "epoch": 0.698009980159923,
      "flos": 21761579450880.0,
      "grad_norm": 2.0113103291204126,
      "language_loss": 0.82701099,
      "learning_rate": 8.826333279490401e-07,
      "loss": 0.84871662,
      "num_input_tokens_seen": 124915810,
      "step": 5805,
      "time_per_iteration": 2.49599289894104
    },
    {
      "auxiliary_loss_clip": 0.01143454,
      "auxiliary_loss_mlp": 0.01026419,
      "balance_loss_clip": 1.04748416,
      "balance_loss_mlp": 1.0194962,
      "epoch": 0.6981302230505622,
      "flos": 19857164267520.0,
      "grad_norm": 2.0915443591533593,
      "language_loss": 0.67778647,
      "learning_rate": 8.819873483925748e-07,
      "loss": 0.69948518,
      "num_input_tokens_seen": 124932930,
      "step": 5806,
      "time_per_iteration": 2.4617981910705566
    },
    {
      "auxiliary_loss_clip": 0.01129893,
      "auxiliary_loss_mlp": 0.00761986,
      "balance_loss_clip": 1.04670787,
      "balance_loss_mlp": 1.00050414,
      "epoch": 0.6982504659412012,
      "flos": 22198648141440.0,
      "grad_norm": 1.9298570994482234,
      "language_loss": 0.74184799,
      "learning_rate": 8.81341538437739e-07,
      "loss": 0.7607668,
      "num_input_tokens_seen": 124951220,
      "step": 5807,
      "time_per_iteration": 3.4212121963500977
    },
    {
      "auxiliary_loss_clip": 0.01142352,
      "auxiliary_loss_mlp": 0.01022129,
      "balance_loss_clip": 1.04330921,
      "balance_loss_mlp": 1.0147146,
      "epoch": 0.6983707088318403,
      "flos": 35588479708800.0,
      "grad_norm": 1.9772594302035267,
      "language_loss": 0.68066496,
      "learning_rate": 8.80695898182503e-07,
      "loss": 0.70230979,
      "num_input_tokens_seen": 124972200,
      "step": 5808,
      "time_per_iteration": 2.6189494132995605
    },
    {
      "auxiliary_loss_clip": 0.01060003,
      "auxiliary_loss_mlp": 0.01001457,
      "balance_loss_clip": 1.02090359,
      "balance_loss_mlp": 1.00030029,
      "epoch": 0.6984909517224794,
      "flos": 65440052760960.0,
      "grad_norm": 0.8101711278610646,
      "language_loss": 0.65125275,
      "learning_rate": 8.800504277248093e-07,
      "loss": 0.67186737,
      "num_input_tokens_seen": 125036950,
      "step": 5809,
      "time_per_iteration": 3.8643782138824463
    },
    {
      "auxiliary_loss_clip": 0.01127645,
      "auxiliary_loss_mlp": 0.00762036,
      "balance_loss_clip": 1.05123889,
      "balance_loss_mlp": 1.00052738,
      "epoch": 0.6986111946131185,
      "flos": 18546927863040.0,
      "grad_norm": 2.212635760890346,
      "language_loss": 0.75038385,
      "learning_rate": 8.794051271625753e-07,
      "loss": 0.76928067,
      "num_input_tokens_seen": 125054585,
      "step": 5810,
      "time_per_iteration": 2.5123722553253174
    },
    {
      "auxiliary_loss_clip": 0.01138871,
      "auxiliary_loss_mlp": 0.01024201,
      "balance_loss_clip": 1.0467298,
      "balance_loss_mlp": 1.01727223,
      "epoch": 0.6987314375037575,
      "flos": 23039173370880.0,
      "grad_norm": 2.6801948281864836,
      "language_loss": 0.83186746,
      "learning_rate": 8.787599965936925e-07,
      "loss": 0.85349822,
      "num_input_tokens_seen": 125075515,
      "step": 5811,
      "time_per_iteration": 2.506580114364624
    },
    {
      "auxiliary_loss_clip": 0.01120556,
      "auxiliary_loss_mlp": 0.01023639,
      "balance_loss_clip": 1.04547179,
      "balance_loss_mlp": 1.01669192,
      "epoch": 0.6988516803943967,
      "flos": 38400393029760.0,
      "grad_norm": 1.715966941119164,
      "language_loss": 0.72301733,
      "learning_rate": 8.781150361160261e-07,
      "loss": 0.74445927,
      "num_input_tokens_seen": 125097425,
      "step": 5812,
      "time_per_iteration": 2.6715238094329834
    },
    {
      "auxiliary_loss_clip": 0.01130334,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 1.04548073,
      "balance_loss_mlp": 1.01907253,
      "epoch": 0.6989719232850358,
      "flos": 24096993926400.0,
      "grad_norm": 1.5827104973818182,
      "language_loss": 0.73502684,
      "learning_rate": 8.774702458274181e-07,
      "loss": 0.75659364,
      "num_input_tokens_seen": 125117830,
      "step": 5813,
      "time_per_iteration": 2.5260837078094482
    },
    {
      "auxiliary_loss_clip": 0.01156398,
      "auxiliary_loss_mlp": 0.01025836,
      "balance_loss_clip": 1.04887247,
      "balance_loss_mlp": 1.01809919,
      "epoch": 0.6990921661756748,
      "flos": 14866838818560.0,
      "grad_norm": 2.3991094318439297,
      "language_loss": 0.70330489,
      "learning_rate": 8.768256258256799e-07,
      "loss": 0.72512722,
      "num_input_tokens_seen": 125134455,
      "step": 5814,
      "time_per_iteration": 2.4202916622161865
    },
    {
      "auxiliary_loss_clip": 0.01157571,
      "auxiliary_loss_mlp": 0.01028734,
      "balance_loss_clip": 1.0483619,
      "balance_loss_mlp": 1.02130151,
      "epoch": 0.699212409066314,
      "flos": 20193719725440.0,
      "grad_norm": 1.6305378040662866,
      "language_loss": 0.74216193,
      "learning_rate": 8.76181176208602e-07,
      "loss": 0.76402497,
      "num_input_tokens_seen": 125152555,
      "step": 5815,
      "time_per_iteration": 2.4460389614105225
    },
    {
      "auxiliary_loss_clip": 0.01102664,
      "auxiliary_loss_mlp": 0.01031813,
      "balance_loss_clip": 1.04071593,
      "balance_loss_mlp": 1.02374852,
      "epoch": 0.699332651956953,
      "flos": 19427888828160.0,
      "grad_norm": 1.9788421832959064,
      "language_loss": 0.73437583,
      "learning_rate": 8.755368970739461e-07,
      "loss": 0.75572062,
      "num_input_tokens_seen": 125171915,
      "step": 5816,
      "time_per_iteration": 2.550799608230591
    },
    {
      "auxiliary_loss_clip": 0.01130549,
      "auxiliary_loss_mlp": 0.01026508,
      "balance_loss_clip": 1.04349005,
      "balance_loss_mlp": 1.01863408,
      "epoch": 0.6994528948475921,
      "flos": 16143714466560.0,
      "grad_norm": 3.259344368100055,
      "language_loss": 0.61532134,
      "learning_rate": 8.748927885194479e-07,
      "loss": 0.6368919,
      "num_input_tokens_seen": 125190220,
      "step": 5817,
      "time_per_iteration": 2.5015554428100586
    },
    {
      "auxiliary_loss_clip": 0.01027549,
      "auxiliary_loss_mlp": 0.01004755,
      "balance_loss_clip": 1.01224947,
      "balance_loss_mlp": 1.00376534,
      "epoch": 0.6995731377382313,
      "flos": 64952420699520.0,
      "grad_norm": 0.7905983814474801,
      "language_loss": 0.57417536,
      "learning_rate": 8.742488506428209e-07,
      "loss": 0.5944984,
      "num_input_tokens_seen": 125249310,
      "step": 5818,
      "time_per_iteration": 3.0465736389160156
    },
    {
      "auxiliary_loss_clip": 0.01143125,
      "auxiliary_loss_mlp": 0.00761813,
      "balance_loss_clip": 1.04676628,
      "balance_loss_mlp": 1.00048351,
      "epoch": 0.6996933806288703,
      "flos": 24900136076160.0,
      "grad_norm": 1.8095859629611606,
      "language_loss": 0.78231585,
      "learning_rate": 8.736050835417466e-07,
      "loss": 0.8013652,
      "num_input_tokens_seen": 125269350,
      "step": 5819,
      "time_per_iteration": 2.534228563308716
    },
    {
      "auxiliary_loss_clip": 0.01159359,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 1.04909015,
      "balance_loss_mlp": 1.01872015,
      "epoch": 0.6998136235195094,
      "flos": 20777806782720.0,
      "grad_norm": 1.8192816756510883,
      "language_loss": 0.61549938,
      "learning_rate": 8.729614873138862e-07,
      "loss": 0.63735652,
      "num_input_tokens_seen": 125286985,
      "step": 5820,
      "time_per_iteration": 2.450200319290161
    },
    {
      "auxiliary_loss_clip": 0.01121254,
      "auxiliary_loss_mlp": 0.01026307,
      "balance_loss_clip": 1.0469892,
      "balance_loss_mlp": 1.01854026,
      "epoch": 0.6999338664101485,
      "flos": 23733470332800.0,
      "grad_norm": 1.903322738519752,
      "language_loss": 0.77887797,
      "learning_rate": 8.723180620568716e-07,
      "loss": 0.80035359,
      "num_input_tokens_seen": 125306240,
      "step": 5821,
      "time_per_iteration": 2.588735818862915
    },
    {
      "auxiliary_loss_clip": 0.01143045,
      "auxiliary_loss_mlp": 0.01022418,
      "balance_loss_clip": 1.0449996,
      "balance_loss_mlp": 1.01518822,
      "epoch": 0.7000541093007876,
      "flos": 19864598382720.0,
      "grad_norm": 1.7493931980047281,
      "language_loss": 0.85130024,
      "learning_rate": 8.716748078683116e-07,
      "loss": 0.87295485,
      "num_input_tokens_seen": 125323015,
      "step": 5822,
      "time_per_iteration": 2.47420597076416
    },
    {
      "auxiliary_loss_clip": 0.01073595,
      "auxiliary_loss_mlp": 0.01027216,
      "balance_loss_clip": 1.03788662,
      "balance_loss_mlp": 1.01860905,
      "epoch": 0.7001743521914267,
      "flos": 29679056029440.0,
      "grad_norm": 2.055232501455391,
      "language_loss": 0.68827254,
      "learning_rate": 8.710317248457855e-07,
      "loss": 0.70928067,
      "num_input_tokens_seen": 125342630,
      "step": 5823,
      "time_per_iteration": 2.6623828411102295
    },
    {
      "auxiliary_loss_clip": 0.01136376,
      "auxiliary_loss_mlp": 0.01026819,
      "balance_loss_clip": 1.04667652,
      "balance_loss_mlp": 1.01930285,
      "epoch": 0.7002945950820658,
      "flos": 27489762080640.0,
      "grad_norm": 1.7973682424863193,
      "language_loss": 0.72462505,
      "learning_rate": 8.703888130868482e-07,
      "loss": 0.74625695,
      "num_input_tokens_seen": 125364480,
      "step": 5824,
      "time_per_iteration": 2.5532102584838867
    },
    {
      "auxiliary_loss_clip": 0.01126737,
      "auxiliary_loss_mlp": 0.01023523,
      "balance_loss_clip": 1.0446353,
      "balance_loss_mlp": 1.01682925,
      "epoch": 0.7004148379727049,
      "flos": 22158463800960.0,
      "grad_norm": 2.822393128196926,
      "language_loss": 0.82287407,
      "learning_rate": 8.697460726890307e-07,
      "loss": 0.84437668,
      "num_input_tokens_seen": 125381625,
      "step": 5825,
      "time_per_iteration": 2.5144972801208496
    },
    {
      "auxiliary_loss_clip": 0.011252,
      "auxiliary_loss_mlp": 0.00762111,
      "balance_loss_clip": 1.04229474,
      "balance_loss_mlp": 1.00047016,
      "epoch": 0.7005350808633439,
      "flos": 19423758764160.0,
      "grad_norm": 1.9369837674046624,
      "language_loss": 0.90281129,
      "learning_rate": 8.691035037498354e-07,
      "loss": 0.92168444,
      "num_input_tokens_seen": 125397615,
      "step": 5826,
      "time_per_iteration": 2.5030462741851807
    },
    {
      "auxiliary_loss_clip": 0.01136311,
      "auxiliary_loss_mlp": 0.01024248,
      "balance_loss_clip": 1.04306197,
      "balance_loss_mlp": 1.0170002,
      "epoch": 0.7006553237539831,
      "flos": 23476708938240.0,
      "grad_norm": 1.8795623109171808,
      "language_loss": 0.72476053,
      "learning_rate": 8.684611063667391e-07,
      "loss": 0.74636608,
      "num_input_tokens_seen": 125418080,
      "step": 5827,
      "time_per_iteration": 2.5319175720214844
    },
    {
      "auxiliary_loss_clip": 0.01153478,
      "auxiliary_loss_mlp": 0.01023916,
      "balance_loss_clip": 1.04597783,
      "balance_loss_mlp": 1.0171659,
      "epoch": 0.7007755666446221,
      "flos": 31212872640000.0,
      "grad_norm": 1.7575375963367066,
      "language_loss": 0.76704061,
      "learning_rate": 8.678188806371935e-07,
      "loss": 0.78881454,
      "num_input_tokens_seen": 125440115,
      "step": 5828,
      "time_per_iteration": 2.565089464187622
    },
    {
      "auxiliary_loss_clip": 0.01153655,
      "auxiliary_loss_mlp": 0.01023954,
      "balance_loss_clip": 1.04608214,
      "balance_loss_mlp": 1.01746929,
      "epoch": 0.7008958095352612,
      "flos": 18149899858560.0,
      "grad_norm": 1.627767872174666,
      "language_loss": 0.85306025,
      "learning_rate": 8.671768266586228e-07,
      "loss": 0.87483633,
      "num_input_tokens_seen": 125458240,
      "step": 5829,
      "time_per_iteration": 3.2037551403045654
    },
    {
      "auxiliary_loss_clip": 0.01123475,
      "auxiliary_loss_mlp": 0.01025652,
      "balance_loss_clip": 1.04342556,
      "balance_loss_mlp": 1.01865458,
      "epoch": 0.7010160524259004,
      "flos": 27452307173760.0,
      "grad_norm": 1.675482902238848,
      "language_loss": 0.78192276,
      "learning_rate": 8.665349445284275e-07,
      "loss": 0.80341399,
      "num_input_tokens_seen": 125477980,
      "step": 5830,
      "time_per_iteration": 2.5903918743133545
    },
    {
      "auxiliary_loss_clip": 0.01126801,
      "auxiliary_loss_mlp": 0.01021239,
      "balance_loss_clip": 1.04730821,
      "balance_loss_mlp": 1.01403618,
      "epoch": 0.7011362953165394,
      "flos": 23842064125440.0,
      "grad_norm": 1.459501148550249,
      "language_loss": 0.81086689,
      "learning_rate": 8.658932343439799e-07,
      "loss": 0.83234727,
      "num_input_tokens_seen": 125497765,
      "step": 5831,
      "time_per_iteration": 3.386246919631958
    },
    {
      "auxiliary_loss_clip": 0.01168793,
      "auxiliary_loss_mlp": 0.01028209,
      "balance_loss_clip": 1.04845846,
      "balance_loss_mlp": 1.02040112,
      "epoch": 0.7012565382071785,
      "flos": 24823430582400.0,
      "grad_norm": 1.9015881522778693,
      "language_loss": 0.77469099,
      "learning_rate": 8.65251696202627e-07,
      "loss": 0.79666102,
      "num_input_tokens_seen": 125514145,
      "step": 5832,
      "time_per_iteration": 2.444854259490967
    },
    {
      "auxiliary_loss_clip": 0.01130083,
      "auxiliary_loss_mlp": 0.01024944,
      "balance_loss_clip": 1.04674983,
      "balance_loss_mlp": 1.01758909,
      "epoch": 0.7013767810978175,
      "flos": 21397445326080.0,
      "grad_norm": 2.2945403525867394,
      "language_loss": 0.87699163,
      "learning_rate": 8.646103302016896e-07,
      "loss": 0.89854187,
      "num_input_tokens_seen": 125533115,
      "step": 5833,
      "time_per_iteration": 3.3560352325439453
    },
    {
      "auxiliary_loss_clip": 0.01122077,
      "auxiliary_loss_mlp": 0.01024399,
      "balance_loss_clip": 1.04351306,
      "balance_loss_mlp": 1.01680541,
      "epoch": 0.7014970239884567,
      "flos": 16687150306560.0,
      "grad_norm": 1.8912571964271003,
      "language_loss": 0.88527739,
      "learning_rate": 8.639691364384614e-07,
      "loss": 0.9067421,
      "num_input_tokens_seen": 125550740,
      "step": 5834,
      "time_per_iteration": 2.555995464324951
    },
    {
      "auxiliary_loss_clip": 0.01143218,
      "auxiliary_loss_mlp": 0.01028679,
      "balance_loss_clip": 1.04683483,
      "balance_loss_mlp": 1.02082062,
      "epoch": 0.7016172668790958,
      "flos": 12568268718720.0,
      "grad_norm": 1.9878251352231335,
      "language_loss": 0.7277925,
      "learning_rate": 8.633281150102136e-07,
      "loss": 0.74951148,
      "num_input_tokens_seen": 125567590,
      "step": 5835,
      "time_per_iteration": 2.4555845260620117
    },
    {
      "auxiliary_loss_clip": 0.01140318,
      "auxiliary_loss_mlp": 0.01021705,
      "balance_loss_clip": 1.04754233,
      "balance_loss_mlp": 1.0147016,
      "epoch": 0.7017375097697348,
      "flos": 17452729808640.0,
      "grad_norm": 5.251000715413512,
      "language_loss": 0.67891234,
      "learning_rate": 8.626872660141855e-07,
      "loss": 0.70053256,
      "num_input_tokens_seen": 125585500,
      "step": 5836,
      "time_per_iteration": 3.2454843521118164
    },
    {
      "auxiliary_loss_clip": 0.01112075,
      "auxiliary_loss_mlp": 0.01028037,
      "balance_loss_clip": 1.04408634,
      "balance_loss_mlp": 1.02060997,
      "epoch": 0.701857752660374,
      "flos": 18513028402560.0,
      "grad_norm": 1.971044312174865,
      "language_loss": 0.7454384,
      "learning_rate": 8.620465895475957e-07,
      "loss": 0.7668395,
      "num_input_tokens_seen": 125603720,
      "step": 5837,
      "time_per_iteration": 2.5258865356445312
    },
    {
      "auxiliary_loss_clip": 0.01107985,
      "auxiliary_loss_mlp": 0.01025099,
      "balance_loss_clip": 1.04362595,
      "balance_loss_mlp": 1.01782727,
      "epoch": 0.701977995551013,
      "flos": 24425971614720.0,
      "grad_norm": 1.584783907692154,
      "language_loss": 0.75290573,
      "learning_rate": 8.614060857076333e-07,
      "loss": 0.77423656,
      "num_input_tokens_seen": 125624390,
      "step": 5838,
      "time_per_iteration": 2.5844318866729736
    },
    {
      "auxiliary_loss_clip": 0.01134269,
      "auxiliary_loss_mlp": 0.0102868,
      "balance_loss_clip": 1.04340684,
      "balance_loss_mlp": 1.02098477,
      "epoch": 0.7020982384416521,
      "flos": 23002759958400.0,
      "grad_norm": 1.9210371169033498,
      "language_loss": 0.74943495,
      "learning_rate": 8.60765754591462e-07,
      "loss": 0.7710644,
      "num_input_tokens_seen": 125644085,
      "step": 5839,
      "time_per_iteration": 2.496074914932251
    },
    {
      "auxiliary_loss_clip": 0.0116628,
      "auxiliary_loss_mlp": 0.01024093,
      "balance_loss_clip": 1.04787827,
      "balance_loss_mlp": 1.01695824,
      "epoch": 0.7022184813322913,
      "flos": 20449080489600.0,
      "grad_norm": 2.405025307998839,
      "language_loss": 0.72841823,
      "learning_rate": 8.601255962962211e-07,
      "loss": 0.75032198,
      "num_input_tokens_seen": 125663095,
      "step": 5840,
      "time_per_iteration": 2.4173026084899902
    },
    {
      "auxiliary_loss_clip": 0.0116535,
      "auxiliary_loss_mlp": 0.0102879,
      "balance_loss_clip": 1.05135846,
      "balance_loss_mlp": 1.0206008,
      "epoch": 0.7023387242229303,
      "flos": 19790514581760.0,
      "grad_norm": 2.4169471783817555,
      "language_loss": 0.72169554,
      "learning_rate": 8.594856109190194e-07,
      "loss": 0.74363697,
      "num_input_tokens_seen": 125680125,
      "step": 5841,
      "time_per_iteration": 2.4390385150909424
    },
    {
      "auxiliary_loss_clip": 0.01168743,
      "auxiliary_loss_mlp": 0.01023853,
      "balance_loss_clip": 1.04885232,
      "balance_loss_mlp": 1.01621222,
      "epoch": 0.7024589671135694,
      "flos": 33259278286080.0,
      "grad_norm": 1.6536870667754038,
      "language_loss": 0.68811655,
      "learning_rate": 8.588457985569446e-07,
      "loss": 0.71004248,
      "num_input_tokens_seen": 125703035,
      "step": 5842,
      "time_per_iteration": 2.551112413406372
    },
    {
      "auxiliary_loss_clip": 0.01170963,
      "auxiliary_loss_mlp": 0.01027417,
      "balance_loss_clip": 1.04891908,
      "balance_loss_mlp": 1.01988006,
      "epoch": 0.7025792100042085,
      "flos": 19098982967040.0,
      "grad_norm": 2.325769154993244,
      "language_loss": 0.72026587,
      "learning_rate": 8.582061593070542e-07,
      "loss": 0.74224973,
      "num_input_tokens_seen": 125723765,
      "step": 5843,
      "time_per_iteration": 2.433300733566284
    },
    {
      "auxiliary_loss_clip": 0.01169784,
      "auxiliary_loss_mlp": 0.00761986,
      "balance_loss_clip": 1.04952431,
      "balance_loss_mlp": 1.00046539,
      "epoch": 0.7026994528948476,
      "flos": 18952611045120.0,
      "grad_norm": 2.09562273630407,
      "language_loss": 0.77006441,
      "learning_rate": 8.57566693266383e-07,
      "loss": 0.78938204,
      "num_input_tokens_seen": 125741455,
      "step": 5844,
      "time_per_iteration": 2.410703659057617
    },
    {
      "auxiliary_loss_clip": 0.01145536,
      "auxiliary_loss_mlp": 0.00762762,
      "balance_loss_clip": 1.04570544,
      "balance_loss_mlp": 1.00048518,
      "epoch": 0.7028196957854866,
      "flos": 19536662188800.0,
      "grad_norm": 6.044492178560601,
      "language_loss": 0.69362265,
      "learning_rate": 8.569274005319354e-07,
      "loss": 0.71270561,
      "num_input_tokens_seen": 125759855,
      "step": 5845,
      "time_per_iteration": 2.4974935054779053
    },
    {
      "auxiliary_loss_clip": 0.01150789,
      "auxiliary_loss_mlp": 0.01026063,
      "balance_loss_clip": 1.04639792,
      "balance_loss_mlp": 1.01847196,
      "epoch": 0.7029399386761258,
      "flos": 20845318394880.0,
      "grad_norm": 1.8238291304949386,
      "language_loss": 0.79649365,
      "learning_rate": 8.562882812006913e-07,
      "loss": 0.81826216,
      "num_input_tokens_seen": 125777345,
      "step": 5846,
      "time_per_iteration": 2.4432694911956787
    },
    {
      "auxiliary_loss_clip": 0.01165123,
      "auxiliary_loss_mlp": 0.01027856,
      "balance_loss_clip": 1.04709482,
      "balance_loss_mlp": 1.02019072,
      "epoch": 0.7030601815667649,
      "flos": 22055005653120.0,
      "grad_norm": 2.237983222467098,
      "language_loss": 0.77741587,
      "learning_rate": 8.556493353696066e-07,
      "loss": 0.79934567,
      "num_input_tokens_seen": 125796345,
      "step": 5847,
      "time_per_iteration": 2.4328713417053223
    },
    {
      "auxiliary_loss_clip": 0.01158836,
      "auxiliary_loss_mlp": 0.00762288,
      "balance_loss_clip": 1.05026376,
      "balance_loss_mlp": 1.0005362,
      "epoch": 0.7031804244574039,
      "flos": 27198742089600.0,
      "grad_norm": 2.322622129372933,
      "language_loss": 0.68586165,
      "learning_rate": 8.550105631356077e-07,
      "loss": 0.70507288,
      "num_input_tokens_seen": 125816070,
      "step": 5848,
      "time_per_iteration": 2.4995882511138916
    },
    {
      "auxiliary_loss_clip": 0.01121023,
      "auxiliary_loss_mlp": 0.01026495,
      "balance_loss_clip": 1.04183519,
      "balance_loss_mlp": 1.01860905,
      "epoch": 0.7033006673480431,
      "flos": 22379853277440.0,
      "grad_norm": 1.845409076447715,
      "language_loss": 0.77179873,
      "learning_rate": 8.543719645955961e-07,
      "loss": 0.79327393,
      "num_input_tokens_seen": 125834400,
      "step": 5849,
      "time_per_iteration": 2.524508476257324
    },
    {
      "auxiliary_loss_clip": 0.01141703,
      "auxiliary_loss_mlp": 0.01022721,
      "balance_loss_clip": 1.04646909,
      "balance_loss_mlp": 1.01544976,
      "epoch": 0.7034209102386821,
      "flos": 24715986024960.0,
      "grad_norm": 1.5584195849754012,
      "language_loss": 0.74670684,
      "learning_rate": 8.537335398464467e-07,
      "loss": 0.76835114,
      "num_input_tokens_seen": 125854720,
      "step": 5850,
      "time_per_iteration": 2.5125372409820557
    },
    {
      "auxiliary_loss_clip": 0.01139523,
      "auxiliary_loss_mlp": 0.01028532,
      "balance_loss_clip": 1.04275489,
      "balance_loss_mlp": 1.02102232,
      "epoch": 0.7035411531293212,
      "flos": 22556174163840.0,
      "grad_norm": 2.5716388290327483,
      "language_loss": 0.85361779,
      "learning_rate": 8.53095288985007e-07,
      "loss": 0.87529838,
      "num_input_tokens_seen": 125868455,
      "step": 5851,
      "time_per_iteration": 2.4696247577667236
    },
    {
      "auxiliary_loss_clip": 0.01166293,
      "auxiliary_loss_mlp": 0.01022844,
      "balance_loss_clip": 1.04903579,
      "balance_loss_mlp": 1.01557183,
      "epoch": 0.7036613960199604,
      "flos": 22674967418880.0,
      "grad_norm": 1.6514324453961566,
      "language_loss": 0.82382655,
      "learning_rate": 8.524572121081009e-07,
      "loss": 0.84571797,
      "num_input_tokens_seen": 125888555,
      "step": 5852,
      "time_per_iteration": 2.426699161529541
    },
    {
      "auxiliary_loss_clip": 0.01159088,
      "auxiliary_loss_mlp": 0.01028098,
      "balance_loss_clip": 1.04752481,
      "balance_loss_mlp": 1.02080822,
      "epoch": 0.7037816389105994,
      "flos": 22492146170880.0,
      "grad_norm": 2.485132491332381,
      "language_loss": 0.62506866,
      "learning_rate": 8.518193093125232e-07,
      "loss": 0.64694047,
      "num_input_tokens_seen": 125907610,
      "step": 5853,
      "time_per_iteration": 2.4579017162323
    },
    {
      "auxiliary_loss_clip": 0.01145507,
      "auxiliary_loss_mlp": 0.01024203,
      "balance_loss_clip": 1.04743564,
      "balance_loss_mlp": 1.01741982,
      "epoch": 0.7039018818012385,
      "flos": 27087490690560.0,
      "grad_norm": 1.6433352001801482,
      "language_loss": 0.80866694,
      "learning_rate": 8.511815806950436e-07,
      "loss": 0.83036405,
      "num_input_tokens_seen": 125928640,
      "step": 5854,
      "time_per_iteration": 2.591630697250366
    },
    {
      "auxiliary_loss_clip": 0.01153622,
      "auxiliary_loss_mlp": 0.01023292,
      "balance_loss_clip": 1.04581761,
      "balance_loss_mlp": 1.0159843,
      "epoch": 0.7040221246918776,
      "flos": 17749819198080.0,
      "grad_norm": 1.6436640800644307,
      "language_loss": 0.78003961,
      "learning_rate": 8.505440263524044e-07,
      "loss": 0.80180871,
      "num_input_tokens_seen": 125947485,
      "step": 5855,
      "time_per_iteration": 3.2241318225860596
    },
    {
      "auxiliary_loss_clip": 0.01155432,
      "auxiliary_loss_mlp": 0.01022679,
      "balance_loss_clip": 1.04542816,
      "balance_loss_mlp": 1.01461482,
      "epoch": 0.7041423675825167,
      "flos": 16279851012480.0,
      "grad_norm": 3.934870077160221,
      "language_loss": 0.88130814,
      "learning_rate": 8.49906646381322e-07,
      "loss": 0.90308917,
      "num_input_tokens_seen": 125960320,
      "step": 5856,
      "time_per_iteration": 2.444519281387329
    },
    {
      "auxiliary_loss_clip": 0.01130476,
      "auxiliary_loss_mlp": 0.01022495,
      "balance_loss_clip": 1.04649091,
      "balance_loss_mlp": 1.01571178,
      "epoch": 0.7042626104731557,
      "flos": 25483181639040.0,
      "grad_norm": 2.630008703780646,
      "language_loss": 0.72072744,
      "learning_rate": 8.492694408784884e-07,
      "loss": 0.74225724,
      "num_input_tokens_seen": 125980575,
      "step": 5857,
      "time_per_iteration": 2.5613455772399902
    },
    {
      "auxiliary_loss_clip": 0.01158781,
      "auxiliary_loss_mlp": 0.01027697,
      "balance_loss_clip": 1.04797339,
      "balance_loss_mlp": 1.02060962,
      "epoch": 0.7043828533637949,
      "flos": 17857622891520.0,
      "grad_norm": 3.749287466560704,
      "language_loss": 0.62531662,
      "learning_rate": 8.486324099405642e-07,
      "loss": 0.64718139,
      "num_input_tokens_seen": 125997420,
      "step": 5858,
      "time_per_iteration": 3.2527029514312744
    },
    {
      "auxiliary_loss_clip": 0.0115246,
      "auxiliary_loss_mlp": 0.01026192,
      "balance_loss_clip": 1.0462575,
      "balance_loss_mlp": 1.01937926,
      "epoch": 0.704503096254434,
      "flos": 29494259533440.0,
      "grad_norm": 1.6526191757965571,
      "language_loss": 0.74981928,
      "learning_rate": 8.479955536641887e-07,
      "loss": 0.77160579,
      "num_input_tokens_seen": 126018915,
      "step": 5859,
      "time_per_iteration": 3.3527369499206543
    },
    {
      "auxiliary_loss_clip": 0.0113179,
      "auxiliary_loss_mlp": 0.01026169,
      "balance_loss_clip": 1.04039085,
      "balance_loss_mlp": 1.01920414,
      "epoch": 0.704623339145073,
      "flos": 30920739327360.0,
      "grad_norm": 1.9982653628858016,
      "language_loss": 0.66674972,
      "learning_rate": 8.473588721459716e-07,
      "loss": 0.68832934,
      "num_input_tokens_seen": 126038825,
      "step": 5860,
      "time_per_iteration": 2.5654003620147705
    },
    {
      "auxiliary_loss_clip": 0.01157067,
      "auxiliary_loss_mlp": 0.01038139,
      "balance_loss_clip": 1.05032432,
      "balance_loss_mlp": 1.02970552,
      "epoch": 0.7047435820357122,
      "flos": 23914747296000.0,
      "grad_norm": 2.081898189509378,
      "language_loss": 0.70755208,
      "learning_rate": 8.467223654824967e-07,
      "loss": 0.72950417,
      "num_input_tokens_seen": 126058280,
      "step": 5861,
      "time_per_iteration": 2.479520320892334
    },
    {
      "auxiliary_loss_clip": 0.011484,
      "auxiliary_loss_mlp": 0.01025183,
      "balance_loss_clip": 1.04636121,
      "balance_loss_mlp": 1.01795602,
      "epoch": 0.7048638249263512,
      "flos": 46494010926720.0,
      "grad_norm": 1.795873877181477,
      "language_loss": 0.62471318,
      "learning_rate": 8.460860337703233e-07,
      "loss": 0.64644897,
      "num_input_tokens_seen": 126078885,
      "step": 5862,
      "time_per_iteration": 3.4283645153045654
    },
    {
      "auxiliary_loss_clip": 0.0111548,
      "auxiliary_loss_mlp": 0.01030576,
      "balance_loss_clip": 1.04255819,
      "balance_loss_mlp": 1.02251458,
      "epoch": 0.7049840678169903,
      "flos": 21689219502720.0,
      "grad_norm": 1.743779418961563,
      "language_loss": 0.70548517,
      "learning_rate": 8.454498771059797e-07,
      "loss": 0.7269457,
      "num_input_tokens_seen": 126098260,
      "step": 5863,
      "time_per_iteration": 2.507906198501587
    },
    {
      "auxiliary_loss_clip": 0.01106172,
      "auxiliary_loss_mlp": 0.01024287,
      "balance_loss_clip": 1.04242611,
      "balance_loss_mlp": 1.01622832,
      "epoch": 0.7051043107076294,
      "flos": 18405081054720.0,
      "grad_norm": 2.095799603692248,
      "language_loss": 0.83618104,
      "learning_rate": 8.448138955859725e-07,
      "loss": 0.85748565,
      "num_input_tokens_seen": 126114845,
      "step": 5864,
      "time_per_iteration": 2.513786792755127
    },
    {
      "auxiliary_loss_clip": 0.01141321,
      "auxiliary_loss_mlp": 0.01025458,
      "balance_loss_clip": 1.04597449,
      "balance_loss_mlp": 1.01797152,
      "epoch": 0.7052245535982685,
      "flos": 19319043640320.0,
      "grad_norm": 1.8849638653558245,
      "language_loss": 0.90191555,
      "learning_rate": 8.44178089306778e-07,
      "loss": 0.92358327,
      "num_input_tokens_seen": 126132780,
      "step": 5865,
      "time_per_iteration": 2.4798450469970703
    },
    {
      "auxiliary_loss_clip": 0.01167238,
      "auxiliary_loss_mlp": 0.01023918,
      "balance_loss_clip": 1.0483036,
      "balance_loss_mlp": 1.01717961,
      "epoch": 0.7053447964889076,
      "flos": 19062138591360.0,
      "grad_norm": 1.7480958491746597,
      "language_loss": 0.7696026,
      "learning_rate": 8.4354245836485e-07,
      "loss": 0.79151416,
      "num_input_tokens_seen": 126151225,
      "step": 5866,
      "time_per_iteration": 2.4040088653564453
    },
    {
      "auxiliary_loss_clip": 0.01129014,
      "auxiliary_loss_mlp": 0.0102406,
      "balance_loss_clip": 1.04613924,
      "balance_loss_mlp": 1.01572466,
      "epoch": 0.7054650393795466,
      "flos": 27379228953600.0,
      "grad_norm": 1.4810631398635954,
      "language_loss": 0.72855306,
      "learning_rate": 8.429070028566108e-07,
      "loss": 0.7500838,
      "num_input_tokens_seen": 126172535,
      "step": 5867,
      "time_per_iteration": 2.573024272918701
    },
    {
      "auxiliary_loss_clip": 0.01153429,
      "auxiliary_loss_mlp": 0.01028346,
      "balance_loss_clip": 1.04831207,
      "balance_loss_mlp": 1.0206387,
      "epoch": 0.7055852822701858,
      "flos": 16102201322880.0,
      "grad_norm": 1.826659008747168,
      "language_loss": 0.75051713,
      "learning_rate": 8.422717228784586e-07,
      "loss": 0.77233487,
      "num_input_tokens_seen": 126189410,
      "step": 5868,
      "time_per_iteration": 2.441877603530884
    },
    {
      "auxiliary_loss_clip": 0.01113021,
      "auxiliary_loss_mlp": 0.01026678,
      "balance_loss_clip": 1.04725575,
      "balance_loss_mlp": 1.01892662,
      "epoch": 0.7057055251608249,
      "flos": 11692299744000.0,
      "grad_norm": 1.8004013758980293,
      "language_loss": 0.69582367,
      "learning_rate": 8.416366185267663e-07,
      "loss": 0.7172206,
      "num_input_tokens_seen": 126206910,
      "step": 5869,
      "time_per_iteration": 2.510385036468506
    },
    {
      "auxiliary_loss_clip": 0.01154063,
      "auxiliary_loss_mlp": 0.0102178,
      "balance_loss_clip": 1.04589319,
      "balance_loss_mlp": 1.01464534,
      "epoch": 0.7058257680514639,
      "flos": 22711560399360.0,
      "grad_norm": 1.6831362460816388,
      "language_loss": 0.78001696,
      "learning_rate": 8.410016898978778e-07,
      "loss": 0.80177534,
      "num_input_tokens_seen": 126224385,
      "step": 5870,
      "time_per_iteration": 2.459655523300171
    },
    {
      "auxiliary_loss_clip": 0.01112938,
      "auxiliary_loss_mlp": 0.01025937,
      "balance_loss_clip": 1.04558218,
      "balance_loss_mlp": 1.0186981,
      "epoch": 0.7059460109421031,
      "flos": 17529543043200.0,
      "grad_norm": 1.6492320170949961,
      "language_loss": 0.78770697,
      "learning_rate": 8.403669370881115e-07,
      "loss": 0.80909574,
      "num_input_tokens_seen": 126243120,
      "step": 5871,
      "time_per_iteration": 2.5525572299957275
    },
    {
      "auxiliary_loss_clip": 0.01168931,
      "auxiliary_loss_mlp": 0.01026675,
      "balance_loss_clip": 1.04996181,
      "balance_loss_mlp": 1.01989484,
      "epoch": 0.7060662538327421,
      "flos": 23544687427200.0,
      "grad_norm": 1.8383117626121872,
      "language_loss": 0.78571761,
      "learning_rate": 8.397323601937587e-07,
      "loss": 0.80767369,
      "num_input_tokens_seen": 126263020,
      "step": 5872,
      "time_per_iteration": 2.429551362991333
    },
    {
      "auxiliary_loss_clip": 0.01120571,
      "auxiliary_loss_mlp": 0.01026093,
      "balance_loss_clip": 1.04356551,
      "balance_loss_mlp": 1.01910996,
      "epoch": 0.7061864967233812,
      "flos": 30260736875520.0,
      "grad_norm": 1.745887519231487,
      "language_loss": 0.76808316,
      "learning_rate": 8.390979593110838e-07,
      "loss": 0.78954977,
      "num_input_tokens_seen": 126285150,
      "step": 5873,
      "time_per_iteration": 2.6032698154449463
    },
    {
      "auxiliary_loss_clip": 0.01145353,
      "auxiliary_loss_mlp": 0.01025955,
      "balance_loss_clip": 1.04846871,
      "balance_loss_mlp": 1.01808167,
      "epoch": 0.7063067396140204,
      "flos": 20701460424960.0,
      "grad_norm": 1.4806571208470214,
      "language_loss": 0.81541359,
      "learning_rate": 8.384637345363262e-07,
      "loss": 0.83712661,
      "num_input_tokens_seen": 126304340,
      "step": 5874,
      "time_per_iteration": 2.5697927474975586
    },
    {
      "auxiliary_loss_clip": 0.01132994,
      "auxiliary_loss_mlp": 0.01023677,
      "balance_loss_clip": 1.04261243,
      "balance_loss_mlp": 1.01652193,
      "epoch": 0.7064269825046594,
      "flos": 32266168081920.0,
      "grad_norm": 1.7318408462134016,
      "language_loss": 0.766119,
      "learning_rate": 8.378296859656964e-07,
      "loss": 0.78768569,
      "num_input_tokens_seen": 126325495,
      "step": 5875,
      "time_per_iteration": 2.5778212547302246
    },
    {
      "auxiliary_loss_clip": 0.01141256,
      "auxiliary_loss_mlp": 0.01028521,
      "balance_loss_clip": 1.04652238,
      "balance_loss_mlp": 1.02108204,
      "epoch": 0.7065472253952985,
      "flos": 30227124723840.0,
      "grad_norm": 2.0260462488986963,
      "language_loss": 0.68625867,
      "learning_rate": 8.371958136953792e-07,
      "loss": 0.70795649,
      "num_input_tokens_seen": 126345525,
      "step": 5876,
      "time_per_iteration": 2.557030439376831
    },
    {
      "auxiliary_loss_clip": 0.0112916,
      "auxiliary_loss_mlp": 0.01028571,
      "balance_loss_clip": 1.04245698,
      "balance_loss_mlp": 1.02067327,
      "epoch": 0.7066674682859376,
      "flos": 16216720859520.0,
      "grad_norm": 3.4497262046621087,
      "language_loss": 0.66013145,
      "learning_rate": 8.365621178215326e-07,
      "loss": 0.68170869,
      "num_input_tokens_seen": 126361995,
      "step": 5877,
      "time_per_iteration": 2.5502867698669434
    },
    {
      "auxiliary_loss_clip": 0.01148615,
      "auxiliary_loss_mlp": 0.01025465,
      "balance_loss_clip": 1.04501164,
      "balance_loss_mlp": 1.01883674,
      "epoch": 0.7067877111765767,
      "flos": 14830461319680.0,
      "grad_norm": 2.467832362969757,
      "language_loss": 0.75492555,
      "learning_rate": 8.359285984402871e-07,
      "loss": 0.77666628,
      "num_input_tokens_seen": 126379260,
      "step": 5878,
      "time_per_iteration": 2.432802438735962
    },
    {
      "auxiliary_loss_clip": 0.01134576,
      "auxiliary_loss_mlp": 0.01022677,
      "balance_loss_clip": 1.04621029,
      "balance_loss_mlp": 1.01581955,
      "epoch": 0.7069079540672157,
      "flos": 25440196037760.0,
      "grad_norm": 1.8604467206663504,
      "language_loss": 0.73830485,
      "learning_rate": 8.352952556477489e-07,
      "loss": 0.75987744,
      "num_input_tokens_seen": 126397170,
      "step": 5879,
      "time_per_iteration": 2.5171470642089844
    },
    {
      "auxiliary_loss_clip": 0.01153866,
      "auxiliary_loss_mlp": 0.01026467,
      "balance_loss_clip": 1.04867709,
      "balance_loss_mlp": 1.01946938,
      "epoch": 0.7070281969578549,
      "flos": 24607751368320.0,
      "grad_norm": 1.9111363703637,
      "language_loss": 0.76861405,
      "learning_rate": 8.34662089539993e-07,
      "loss": 0.79041731,
      "num_input_tokens_seen": 126416680,
      "step": 5880,
      "time_per_iteration": 2.476982593536377
    },
    {
      "auxiliary_loss_clip": 0.01166145,
      "auxiliary_loss_mlp": 0.01024269,
      "balance_loss_clip": 1.04948759,
      "balance_loss_mlp": 1.01721179,
      "epoch": 0.707148439848494,
      "flos": 26724469887360.0,
      "grad_norm": 9.695301230249642,
      "language_loss": 0.79246104,
      "learning_rate": 8.340291002130722e-07,
      "loss": 0.81436527,
      "num_input_tokens_seen": 126435870,
      "step": 5881,
      "time_per_iteration": 2.459730625152588
    },
    {
      "auxiliary_loss_clip": 0.01170055,
      "auxiliary_loss_mlp": 0.0102628,
      "balance_loss_clip": 1.04911423,
      "balance_loss_mlp": 1.01866269,
      "epoch": 0.707268682739133,
      "flos": 15085750256640.0,
      "grad_norm": 4.180355208202724,
      "language_loss": 0.79438704,
      "learning_rate": 8.3339628776301e-07,
      "loss": 0.8163504,
      "num_input_tokens_seen": 126454010,
      "step": 5882,
      "time_per_iteration": 3.126051425933838
    },
    {
      "auxiliary_loss_clip": 0.01167083,
      "auxiliary_loss_mlp": 0.0102498,
      "balance_loss_clip": 1.0481497,
      "balance_loss_mlp": 1.0180068,
      "epoch": 0.7073889256297722,
      "flos": 34313148345600.0,
      "grad_norm": 1.829479874034662,
      "language_loss": 0.57265085,
      "learning_rate": 8.327636522858033e-07,
      "loss": 0.59457147,
      "num_input_tokens_seen": 126473615,
      "step": 5883,
      "time_per_iteration": 2.517216920852661
    },
    {
      "auxiliary_loss_clip": 0.01113286,
      "auxiliary_loss_mlp": 0.0102884,
      "balance_loss_clip": 1.04569483,
      "balance_loss_mlp": 1.02137113,
      "epoch": 0.7075091685204112,
      "flos": 20083940784000.0,
      "grad_norm": 1.8732293239577875,
      "language_loss": 0.77574593,
      "learning_rate": 8.321311938774225e-07,
      "loss": 0.79716718,
      "num_input_tokens_seen": 126492705,
      "step": 5884,
      "time_per_iteration": 3.3915157318115234
    },
    {
      "auxiliary_loss_clip": 0.01171782,
      "auxiliary_loss_mlp": 0.01026484,
      "balance_loss_clip": 1.0495218,
      "balance_loss_mlp": 1.01899755,
      "epoch": 0.7076294114110503,
      "flos": 20777124424320.0,
      "grad_norm": 1.9646135379962832,
      "language_loss": 0.79203027,
      "learning_rate": 8.314989126338104e-07,
      "loss": 0.81401294,
      "num_input_tokens_seen": 126512715,
      "step": 5885,
      "time_per_iteration": 3.2675116062164307
    },
    {
      "auxiliary_loss_clip": 0.01157339,
      "auxiliary_loss_mlp": 0.01025987,
      "balance_loss_clip": 1.04732287,
      "balance_loss_mlp": 1.01873934,
      "epoch": 0.7077496543016895,
      "flos": 17967689141760.0,
      "grad_norm": 1.684375479227643,
      "language_loss": 0.84470969,
      "learning_rate": 8.308668086508847e-07,
      "loss": 0.86654294,
      "num_input_tokens_seen": 126530795,
      "step": 5886,
      "time_per_iteration": 2.439687967300415
    },
    {
      "auxiliary_loss_clip": 0.01128684,
      "auxiliary_loss_mlp": 0.01020038,
      "balance_loss_clip": 1.04210651,
      "balance_loss_mlp": 1.01254249,
      "epoch": 0.7078698971923285,
      "flos": 45478098564480.0,
      "grad_norm": 1.881127755599909,
      "language_loss": 0.73955005,
      "learning_rate": 8.302348820245342e-07,
      "loss": 0.76103729,
      "num_input_tokens_seen": 126553360,
      "step": 5887,
      "time_per_iteration": 2.731684446334839
    },
    {
      "auxiliary_loss_clip": 0.01126637,
      "auxiliary_loss_mlp": 0.01025945,
      "balance_loss_clip": 1.04322624,
      "balance_loss_mlp": 1.01799381,
      "epoch": 0.7079901400829676,
      "flos": 26943704547840.0,
      "grad_norm": 2.409356345252419,
      "language_loss": 0.70149148,
      "learning_rate": 8.296031328506232e-07,
      "loss": 0.72301733,
      "num_input_tokens_seen": 126573110,
      "step": 5888,
      "time_per_iteration": 2.565338134765625
    },
    {
      "auxiliary_loss_clip": 0.01141519,
      "auxiliary_loss_mlp": 0.01024775,
      "balance_loss_clip": 1.0468049,
      "balance_loss_mlp": 1.01764059,
      "epoch": 0.7081103829736067,
      "flos": 24423206267520.0,
      "grad_norm": 1.9519650635385772,
      "language_loss": 0.75844556,
      "learning_rate": 8.289715612249857e-07,
      "loss": 0.78010851,
      "num_input_tokens_seen": 126593725,
      "step": 5889,
      "time_per_iteration": 3.256669044494629
    },
    {
      "auxiliary_loss_clip": 0.0113752,
      "auxiliary_loss_mlp": 0.01026351,
      "balance_loss_clip": 1.04622734,
      "balance_loss_mlp": 1.01879311,
      "epoch": 0.7082306258642458,
      "flos": 18543300589440.0,
      "grad_norm": 2.4845261798980314,
      "language_loss": 0.77542293,
      "learning_rate": 8.283401672434305e-07,
      "loss": 0.79706168,
      "num_input_tokens_seen": 126608950,
      "step": 5890,
      "time_per_iteration": 2.437650203704834
    },
    {
      "auxiliary_loss_clip": 0.01138333,
      "auxiliary_loss_mlp": 0.01026034,
      "balance_loss_clip": 1.04798603,
      "balance_loss_mlp": 1.01906633,
      "epoch": 0.7083508687548848,
      "flos": 23477534951040.0,
      "grad_norm": 2.413514482741669,
      "language_loss": 0.70276546,
      "learning_rate": 8.277089510017412e-07,
      "loss": 0.72440922,
      "num_input_tokens_seen": 126629755,
      "step": 5891,
      "time_per_iteration": 2.5006752014160156
    },
    {
      "auxiliary_loss_clip": 0.011395,
      "auxiliary_loss_mlp": 0.01025214,
      "balance_loss_clip": 1.0491358,
      "balance_loss_mlp": 1.01809764,
      "epoch": 0.708471111645524,
      "flos": 22419463000320.0,
      "grad_norm": 5.2911139257654405,
      "language_loss": 0.8237235,
      "learning_rate": 8.270779125956719e-07,
      "loss": 0.84537065,
      "num_input_tokens_seen": 126650135,
      "step": 5892,
      "time_per_iteration": 2.4883430004119873
    },
    {
      "auxiliary_loss_clip": 0.01107796,
      "auxiliary_loss_mlp": 0.01024325,
      "balance_loss_clip": 1.04331398,
      "balance_loss_mlp": 1.01713359,
      "epoch": 0.7085913545361631,
      "flos": 20922885815040.0,
      "grad_norm": 2.6514971254165056,
      "language_loss": 0.80068707,
      "learning_rate": 8.264470521209505e-07,
      "loss": 0.82200825,
      "num_input_tokens_seen": 126668500,
      "step": 5893,
      "time_per_iteration": 2.5412471294403076
    },
    {
      "auxiliary_loss_clip": 0.01145371,
      "auxiliary_loss_mlp": 0.01023477,
      "balance_loss_clip": 1.04493821,
      "balance_loss_mlp": 1.01636338,
      "epoch": 0.7087115974268021,
      "flos": 15012384727680.0,
      "grad_norm": 2.3041720460929405,
      "language_loss": 0.76861179,
      "learning_rate": 8.258163696732785e-07,
      "loss": 0.79030031,
      "num_input_tokens_seen": 126686090,
      "step": 5894,
      "time_per_iteration": 2.422011375427246
    },
    {
      "auxiliary_loss_clip": 0.01148852,
      "auxiliary_loss_mlp": 0.01024945,
      "balance_loss_clip": 1.04607105,
      "balance_loss_mlp": 1.01791513,
      "epoch": 0.7088318403174413,
      "flos": 21539040739200.0,
      "grad_norm": 1.857016607301194,
      "language_loss": 0.7716186,
      "learning_rate": 8.251858653483288e-07,
      "loss": 0.7933566,
      "num_input_tokens_seen": 126704255,
      "step": 5895,
      "time_per_iteration": 2.457542657852173
    },
    {
      "auxiliary_loss_clip": 0.01154831,
      "auxiliary_loss_mlp": 0.01023376,
      "balance_loss_clip": 1.04921913,
      "balance_loss_mlp": 1.01609182,
      "epoch": 0.7089520832080803,
      "flos": 15516785462400.0,
      "grad_norm": 2.015062678844766,
      "language_loss": 0.85957199,
      "learning_rate": 8.245555392417501e-07,
      "loss": 0.88135409,
      "num_input_tokens_seen": 126718910,
      "step": 5896,
      "time_per_iteration": 2.3973546028137207
    },
    {
      "auxiliary_loss_clip": 0.01098049,
      "auxiliary_loss_mlp": 0.01021564,
      "balance_loss_clip": 1.03941357,
      "balance_loss_mlp": 1.01419091,
      "epoch": 0.7090723260987194,
      "flos": 20412667077120.0,
      "grad_norm": 2.6988756889881804,
      "language_loss": 0.78762686,
      "learning_rate": 8.239253914491613e-07,
      "loss": 0.80882299,
      "num_input_tokens_seen": 126737235,
      "step": 5897,
      "time_per_iteration": 2.5295279026031494
    },
    {
      "auxiliary_loss_clip": 0.01122608,
      "auxiliary_loss_mlp": 0.01020553,
      "balance_loss_clip": 1.04569817,
      "balance_loss_mlp": 1.01364517,
      "epoch": 0.7091925689893585,
      "flos": 25668337271040.0,
      "grad_norm": 1.7954880170059735,
      "language_loss": 0.75259316,
      "learning_rate": 8.232954220661556e-07,
      "loss": 0.77402484,
      "num_input_tokens_seen": 126759970,
      "step": 5898,
      "time_per_iteration": 2.5826380252838135
    },
    {
      "auxiliary_loss_clip": 0.01169178,
      "auxiliary_loss_mlp": 0.01028165,
      "balance_loss_clip": 1.05151212,
      "balance_loss_mlp": 1.02137613,
      "epoch": 0.7093128118799976,
      "flos": 24206629213440.0,
      "grad_norm": 2.4662082860745635,
      "language_loss": 0.70080054,
      "learning_rate": 8.226656311882989e-07,
      "loss": 0.72277403,
      "num_input_tokens_seen": 126779280,
      "step": 5899,
      "time_per_iteration": 2.4339444637298584
    },
    {
      "auxiliary_loss_clip": 0.01151128,
      "auxiliary_loss_mlp": 0.01026805,
      "balance_loss_clip": 1.04739881,
      "balance_loss_mlp": 1.01989055,
      "epoch": 0.7094330547706367,
      "flos": 16646786398080.0,
      "grad_norm": 2.0857110750536445,
      "language_loss": 0.77122843,
      "learning_rate": 8.22036018911129e-07,
      "loss": 0.79300779,
      "num_input_tokens_seen": 126797310,
      "step": 5900,
      "time_per_iteration": 2.435462713241577
    },
    {
      "auxiliary_loss_clip": 0.01172454,
      "auxiliary_loss_mlp": 0.01028306,
      "balance_loss_clip": 1.04938328,
      "balance_loss_mlp": 1.0207299,
      "epoch": 0.7095532976612757,
      "flos": 16283370545280.0,
      "grad_norm": 2.164938336487819,
      "language_loss": 0.80656147,
      "learning_rate": 8.214065853301599e-07,
      "loss": 0.82856911,
      "num_input_tokens_seen": 126812840,
      "step": 5901,
      "time_per_iteration": 2.388033390045166
    },
    {
      "auxiliary_loss_clip": 0.01063006,
      "auxiliary_loss_mlp": 0.01003132,
      "balance_loss_clip": 1.01616704,
      "balance_loss_mlp": 1.00214851,
      "epoch": 0.7096735405519149,
      "flos": 70722080559360.0,
      "grad_norm": 0.8472975926833304,
      "language_loss": 0.58255839,
      "learning_rate": 8.207773305408734e-07,
      "loss": 0.60321987,
      "num_input_tokens_seen": 126880060,
      "step": 5902,
      "time_per_iteration": 3.1554813385009766
    },
    {
      "auxiliary_loss_clip": 0.01119159,
      "auxiliary_loss_mlp": 0.0102758,
      "balance_loss_clip": 1.04238808,
      "balance_loss_mlp": 1.0198493,
      "epoch": 0.709793783442554,
      "flos": 23621500661760.0,
      "grad_norm": 1.965219374983623,
      "language_loss": 0.79957139,
      "learning_rate": 8.201482546387288e-07,
      "loss": 0.82103878,
      "num_input_tokens_seen": 126899535,
      "step": 5903,
      "time_per_iteration": 2.6002721786499023
    },
    {
      "auxiliary_loss_clip": 0.01153218,
      "auxiliary_loss_mlp": 0.0102567,
      "balance_loss_clip": 1.04794693,
      "balance_loss_mlp": 1.01869631,
      "epoch": 0.709914026333193,
      "flos": 25993472204160.0,
      "grad_norm": 1.6392482300275395,
      "language_loss": 0.91766375,
      "learning_rate": 8.195193577191553e-07,
      "loss": 0.93945265,
      "num_input_tokens_seen": 126921365,
      "step": 5904,
      "time_per_iteration": 2.4943082332611084
    },
    {
      "auxiliary_loss_clip": 0.01147404,
      "auxiliary_loss_mlp": 0.00761936,
      "balance_loss_clip": 1.0465194,
      "balance_loss_mlp": 1.0004282,
      "epoch": 0.7100342692238322,
      "flos": 24861531934080.0,
      "grad_norm": 2.201979540955928,
      "language_loss": 0.84544945,
      "learning_rate": 8.188906398775579e-07,
      "loss": 0.86454284,
      "num_input_tokens_seen": 126941910,
      "step": 5905,
      "time_per_iteration": 2.5196001529693604
    },
    {
      "auxiliary_loss_clip": 0.01168288,
      "auxiliary_loss_mlp": 0.00762281,
      "balance_loss_clip": 1.04793715,
      "balance_loss_mlp": 1.00042605,
      "epoch": 0.7101545121144712,
      "flos": 24932203943040.0,
      "grad_norm": 1.965040495960102,
      "language_loss": 0.69046426,
      "learning_rate": 8.18262101209311e-07,
      "loss": 0.70976996,
      "num_input_tokens_seen": 126961120,
      "step": 5906,
      "time_per_iteration": 2.4749648571014404
    },
    {
      "auxiliary_loss_clip": 0.011577,
      "auxiliary_loss_mlp": 0.01022723,
      "balance_loss_clip": 1.04731846,
      "balance_loss_mlp": 1.0156002,
      "epoch": 0.7102747550051103,
      "flos": 23768842250880.0,
      "grad_norm": 1.7456079028346174,
      "language_loss": 0.70122093,
      "learning_rate": 8.176337418097626e-07,
      "loss": 0.72302514,
      "num_input_tokens_seen": 126981590,
      "step": 5907,
      "time_per_iteration": 2.4929699897766113
    },
    {
      "auxiliary_loss_clip": 0.01153564,
      "auxiliary_loss_mlp": 0.00761701,
      "balance_loss_clip": 1.04880261,
      "balance_loss_mlp": 1.00035095,
      "epoch": 0.7103949978957494,
      "flos": 15303907509120.0,
      "grad_norm": 2.6451944249928823,
      "language_loss": 0.79995382,
      "learning_rate": 8.170055617742364e-07,
      "loss": 0.81910646,
      "num_input_tokens_seen": 126998870,
      "step": 5908,
      "time_per_iteration": 2.4380545616149902
    },
    {
      "auxiliary_loss_clip": 0.01133699,
      "auxiliary_loss_mlp": 0.01030153,
      "balance_loss_clip": 1.04359865,
      "balance_loss_mlp": 1.02233922,
      "epoch": 0.7105152407863885,
      "flos": 22638805401600.0,
      "grad_norm": 1.769385070674959,
      "language_loss": 0.71064436,
      "learning_rate": 8.163775611980252e-07,
      "loss": 0.73228288,
      "num_input_tokens_seen": 127017980,
      "step": 5909,
      "time_per_iteration": 3.2582204341888428
    },
    {
      "auxiliary_loss_clip": 0.01140579,
      "auxiliary_loss_mlp": 0.01026256,
      "balance_loss_clip": 1.04775047,
      "balance_loss_mlp": 1.01936603,
      "epoch": 0.7106354836770276,
      "flos": 17238594879360.0,
      "grad_norm": 1.5974972136353895,
      "language_loss": 0.78581631,
      "learning_rate": 8.157497401763982e-07,
      "loss": 0.80748469,
      "num_input_tokens_seen": 127035645,
      "step": 5910,
      "time_per_iteration": 2.4502665996551514
    },
    {
      "auxiliary_loss_clip": 0.01151934,
      "auxiliary_loss_mlp": 0.01023923,
      "balance_loss_clip": 1.04712069,
      "balance_loss_mlp": 1.01661563,
      "epoch": 0.7107557265676667,
      "flos": 20193647898240.0,
      "grad_norm": 1.8255890527228482,
      "language_loss": 0.78039765,
      "learning_rate": 8.151220988045935e-07,
      "loss": 0.80215621,
      "num_input_tokens_seen": 127054900,
      "step": 5911,
      "time_per_iteration": 3.2621326446533203
    },
    {
      "auxiliary_loss_clip": 0.01153461,
      "auxiliary_loss_mlp": 0.01023164,
      "balance_loss_clip": 1.04785657,
      "balance_loss_mlp": 1.01639557,
      "epoch": 0.7108759694583058,
      "flos": 21507080613120.0,
      "grad_norm": 2.0565607215947623,
      "language_loss": 0.82941008,
      "learning_rate": 8.144946371778234e-07,
      "loss": 0.85117626,
      "num_input_tokens_seen": 127075010,
      "step": 5912,
      "time_per_iteration": 3.3095526695251465
    },
    {
      "auxiliary_loss_clip": 0.01140143,
      "auxiliary_loss_mlp": 0.00762795,
      "balance_loss_clip": 1.04746783,
      "balance_loss_mlp": 1.0004282,
      "epoch": 0.7109962123489448,
      "flos": 24061909317120.0,
      "grad_norm": 1.8420633798382486,
      "language_loss": 0.78162861,
      "learning_rate": 8.138673553912751e-07,
      "loss": 0.80065787,
      "num_input_tokens_seen": 127095570,
      "step": 5913,
      "time_per_iteration": 2.4990622997283936
    },
    {
      "auxiliary_loss_clip": 0.01111225,
      "auxiliary_loss_mlp": 0.01023491,
      "balance_loss_clip": 1.04306483,
      "balance_loss_mlp": 1.01618683,
      "epoch": 0.711116455239584,
      "flos": 30480474326400.0,
      "grad_norm": 10.918322215441364,
      "language_loss": 0.57274139,
      "learning_rate": 8.132402535401059e-07,
      "loss": 0.59408861,
      "num_input_tokens_seen": 127116825,
      "step": 5914,
      "time_per_iteration": 2.619413375854492
    },
    {
      "auxiliary_loss_clip": 0.01153458,
      "auxiliary_loss_mlp": 0.0102762,
      "balance_loss_clip": 1.04951155,
      "balance_loss_mlp": 1.0201335,
      "epoch": 0.711236698130223,
      "flos": 25045610158080.0,
      "grad_norm": 2.109172351053258,
      "language_loss": 0.74345809,
      "learning_rate": 8.126133317194465e-07,
      "loss": 0.76526886,
      "num_input_tokens_seen": 127137015,
      "step": 5915,
      "time_per_iteration": 3.184783458709717
    },
    {
      "auxiliary_loss_clip": 0.01108946,
      "auxiliary_loss_mlp": 0.01027381,
      "balance_loss_clip": 1.04166222,
      "balance_loss_mlp": 1.01949489,
      "epoch": 0.7113569410208621,
      "flos": 24206701040640.0,
      "grad_norm": 1.9780733247385713,
      "language_loss": 0.74532568,
      "learning_rate": 8.11986590024401e-07,
      "loss": 0.76668894,
      "num_input_tokens_seen": 127156755,
      "step": 5916,
      "time_per_iteration": 2.5804224014282227
    },
    {
      "auxiliary_loss_clip": 0.01147855,
      "auxiliary_loss_mlp": 0.01027545,
      "balance_loss_clip": 1.05148816,
      "balance_loss_mlp": 1.01954293,
      "epoch": 0.7114771839115013,
      "flos": 35439306526080.0,
      "grad_norm": 1.876163098924071,
      "language_loss": 0.69152963,
      "learning_rate": 8.113600285500442e-07,
      "loss": 0.71328366,
      "num_input_tokens_seen": 127176965,
      "step": 5917,
      "time_per_iteration": 2.600675106048584
    },
    {
      "auxiliary_loss_clip": 0.01168909,
      "auxiliary_loss_mlp": 0.01019595,
      "balance_loss_clip": 1.04831445,
      "balance_loss_mlp": 1.01253796,
      "epoch": 0.7115974268021403,
      "flos": 21099458096640.0,
      "grad_norm": 1.8739102300953667,
      "language_loss": 0.74540234,
      "learning_rate": 8.107336473914268e-07,
      "loss": 0.76728743,
      "num_input_tokens_seen": 127195595,
      "step": 5918,
      "time_per_iteration": 2.414710283279419
    },
    {
      "auxiliary_loss_clip": 0.010498,
      "auxiliary_loss_mlp": 0.01001163,
      "balance_loss_clip": 1.01632953,
      "balance_loss_mlp": 1.00010765,
      "epoch": 0.7117176696927794,
      "flos": 56752866616320.0,
      "grad_norm": 0.7701010514575031,
      "language_loss": 0.55774122,
      "learning_rate": 8.101074466435694e-07,
      "loss": 0.57825083,
      "num_input_tokens_seen": 127255070,
      "step": 5919,
      "time_per_iteration": 3.0018229484558105
    },
    {
      "auxiliary_loss_clip": 0.01147664,
      "auxiliary_loss_mlp": 0.01025741,
      "balance_loss_clip": 1.04545951,
      "balance_loss_mlp": 1.01873732,
      "epoch": 0.7118379125834186,
      "flos": 15925269905280.0,
      "grad_norm": 1.7143730186604327,
      "language_loss": 0.6781137,
      "learning_rate": 8.094814264014662e-07,
      "loss": 0.69984782,
      "num_input_tokens_seen": 127273825,
      "step": 5920,
      "time_per_iteration": 2.423405885696411
    },
    {
      "auxiliary_loss_clip": 0.01170842,
      "auxiliary_loss_mlp": 0.01028259,
      "balance_loss_clip": 1.04891109,
      "balance_loss_mlp": 1.02046835,
      "epoch": 0.7119581554740576,
      "flos": 20193360589440.0,
      "grad_norm": 2.06826062705643,
      "language_loss": 0.81433731,
      "learning_rate": 8.088555867600844e-07,
      "loss": 0.83632827,
      "num_input_tokens_seen": 127289990,
      "step": 5921,
      "time_per_iteration": 2.425159454345703
    },
    {
      "auxiliary_loss_clip": 0.01124207,
      "auxiliary_loss_mlp": 0.01022925,
      "balance_loss_clip": 1.0438298,
      "balance_loss_mlp": 1.01627922,
      "epoch": 0.7120783983646967,
      "flos": 34715383822080.0,
      "grad_norm": 1.645086391183544,
      "language_loss": 0.60422426,
      "learning_rate": 8.08229927814362e-07,
      "loss": 0.62569559,
      "num_input_tokens_seen": 127312880,
      "step": 5922,
      "time_per_iteration": 2.6355979442596436
    },
    {
      "auxiliary_loss_clip": 0.01121827,
      "auxiliary_loss_mlp": 0.01022627,
      "balance_loss_clip": 1.04188216,
      "balance_loss_mlp": 1.01566815,
      "epoch": 0.7121986412553358,
      "flos": 26359114700160.0,
      "grad_norm": 1.6856458353800652,
      "language_loss": 0.65223628,
      "learning_rate": 8.076044496592134e-07,
      "loss": 0.67368084,
      "num_input_tokens_seen": 127334730,
      "step": 5923,
      "time_per_iteration": 2.558905601501465
    },
    {
      "auxiliary_loss_clip": 0.01140975,
      "auxiliary_loss_mlp": 0.01026422,
      "balance_loss_clip": 1.04757428,
      "balance_loss_mlp": 1.01953149,
      "epoch": 0.7123188841459749,
      "flos": 11145344371200.0,
      "grad_norm": 2.9472909354827648,
      "language_loss": 0.779719,
      "learning_rate": 8.069791523895204e-07,
      "loss": 0.80139291,
      "num_input_tokens_seen": 127351180,
      "step": 5924,
      "time_per_iteration": 2.444303274154663
    },
    {
      "auxiliary_loss_clip": 0.01113495,
      "auxiliary_loss_mlp": 0.01027021,
      "balance_loss_clip": 1.03968561,
      "balance_loss_mlp": 1.02003574,
      "epoch": 0.7124391270366139,
      "flos": 20811670329600.0,
      "grad_norm": 2.7339564776865966,
      "language_loss": 0.77795053,
      "learning_rate": 8.063540361001422e-07,
      "loss": 0.79935575,
      "num_input_tokens_seen": 127369750,
      "step": 5925,
      "time_per_iteration": 2.509763240814209
    },
    {
      "auxiliary_loss_clip": 0.01120847,
      "auxiliary_loss_mlp": 0.01026442,
      "balance_loss_clip": 1.04367721,
      "balance_loss_mlp": 1.0185082,
      "epoch": 0.7125593699272531,
      "flos": 17603734584960.0,
      "grad_norm": 1.8695298956361248,
      "language_loss": 0.79380929,
      "learning_rate": 8.057291008859069e-07,
      "loss": 0.81528217,
      "num_input_tokens_seen": 127387910,
      "step": 5926,
      "time_per_iteration": 2.5028536319732666
    },
    {
      "auxiliary_loss_clip": 0.01151277,
      "auxiliary_loss_mlp": 0.01027847,
      "balance_loss_clip": 1.04630232,
      "balance_loss_mlp": 1.02089465,
      "epoch": 0.7126796128178922,
      "flos": 28654057526400.0,
      "grad_norm": 1.892980246859141,
      "language_loss": 0.68136835,
      "learning_rate": 8.051043468416187e-07,
      "loss": 0.70315963,
      "num_input_tokens_seen": 127409160,
      "step": 5927,
      "time_per_iteration": 2.5853114128112793
    },
    {
      "auxiliary_loss_clip": 0.01167046,
      "auxiliary_loss_mlp": 0.0102243,
      "balance_loss_clip": 1.04982388,
      "balance_loss_mlp": 1.01541996,
      "epoch": 0.7127998557085312,
      "flos": 16034438315520.0,
      "grad_norm": 1.739864015659535,
      "language_loss": 0.82251871,
      "learning_rate": 8.044797740620506e-07,
      "loss": 0.84441346,
      "num_input_tokens_seen": 127427765,
      "step": 5928,
      "time_per_iteration": 2.4110355377197266
    },
    {
      "auxiliary_loss_clip": 0.01107474,
      "auxiliary_loss_mlp": 0.01027235,
      "balance_loss_clip": 1.04434967,
      "balance_loss_mlp": 1.02057743,
      "epoch": 0.7129200985991703,
      "flos": 23403271582080.0,
      "grad_norm": 2.3382013158819914,
      "language_loss": 0.78935671,
      "learning_rate": 8.038553826419494e-07,
      "loss": 0.81070387,
      "num_input_tokens_seen": 127446475,
      "step": 5929,
      "time_per_iteration": 2.5456135272979736
    },
    {
      "auxiliary_loss_clip": 0.01166213,
      "auxiliary_loss_mlp": 0.01021946,
      "balance_loss_clip": 1.04738116,
      "balance_loss_mlp": 1.01466799,
      "epoch": 0.7130403414898094,
      "flos": 21397445326080.0,
      "grad_norm": 1.5925983082648503,
      "language_loss": 0.80878288,
      "learning_rate": 8.032311726760364e-07,
      "loss": 0.83066452,
      "num_input_tokens_seen": 127467695,
      "step": 5930,
      "time_per_iteration": 2.4311695098876953
    },
    {
      "auxiliary_loss_clip": 0.01117748,
      "auxiliary_loss_mlp": 0.01022347,
      "balance_loss_clip": 1.04487777,
      "balance_loss_mlp": 1.01435423,
      "epoch": 0.7131605843804485,
      "flos": 74739045306240.0,
      "grad_norm": 1.6652256721273955,
      "language_loss": 0.6909281,
      "learning_rate": 8.026071442590022e-07,
      "loss": 0.71232903,
      "num_input_tokens_seen": 127494590,
      "step": 5931,
      "time_per_iteration": 2.9107630252838135
    },
    {
      "auxiliary_loss_clip": 0.01155695,
      "auxiliary_loss_mlp": 0.01023063,
      "balance_loss_clip": 1.05101502,
      "balance_loss_mlp": 1.0163188,
      "epoch": 0.7132808272710875,
      "flos": 18368739469440.0,
      "grad_norm": 1.8645366465907327,
      "language_loss": 0.80614018,
      "learning_rate": 8.019832974855134e-07,
      "loss": 0.82792771,
      "num_input_tokens_seen": 127512550,
      "step": 5932,
      "time_per_iteration": 2.414710760116577
    },
    {
      "auxiliary_loss_clip": 0.01124711,
      "auxiliary_loss_mlp": 0.01021925,
      "balance_loss_clip": 1.04586852,
      "balance_loss_mlp": 1.01435828,
      "epoch": 0.7134010701617267,
      "flos": 23253380127360.0,
      "grad_norm": 2.397313765032879,
      "language_loss": 0.82587063,
      "learning_rate": 8.013596324502052e-07,
      "loss": 0.84733701,
      "num_input_tokens_seen": 127531015,
      "step": 5933,
      "time_per_iteration": 2.5310628414154053
    },
    {
      "auxiliary_loss_clip": 0.01146503,
      "auxiliary_loss_mlp": 0.01022212,
      "balance_loss_clip": 1.04686999,
      "balance_loss_mlp": 1.01549208,
      "epoch": 0.7135213130523658,
      "flos": 23653137565440.0,
      "grad_norm": 1.6789758185620631,
      "language_loss": 0.7860254,
      "learning_rate": 8.007361492476872e-07,
      "loss": 0.80771255,
      "num_input_tokens_seen": 127550340,
      "step": 5934,
      "time_per_iteration": 2.4680466651916504
    },
    {
      "auxiliary_loss_clip": 0.01135552,
      "auxiliary_loss_mlp": 0.01025064,
      "balance_loss_clip": 1.04570079,
      "balance_loss_mlp": 1.0172503,
      "epoch": 0.7136415559430048,
      "flos": 24790644443520.0,
      "grad_norm": 1.626311560765611,
      "language_loss": 0.7903989,
      "learning_rate": 8.001128479725426e-07,
      "loss": 0.81200504,
      "num_input_tokens_seen": 127572245,
      "step": 5935,
      "time_per_iteration": 2.542800188064575
    },
    {
      "auxiliary_loss_clip": 0.01102892,
      "auxiliary_loss_mlp": 0.01022208,
      "balance_loss_clip": 1.03877807,
      "balance_loss_mlp": 1.01485837,
      "epoch": 0.713761798833644,
      "flos": 18296954138880.0,
      "grad_norm": 1.6236523823109696,
      "language_loss": 0.81287438,
      "learning_rate": 7.994897287193248e-07,
      "loss": 0.8341254,
      "num_input_tokens_seen": 127591625,
      "step": 5936,
      "time_per_iteration": 3.272777557373047
    },
    {
      "auxiliary_loss_clip": 0.01156622,
      "auxiliary_loss_mlp": 0.01028473,
      "balance_loss_clip": 1.04727864,
      "balance_loss_mlp": 1.02078366,
      "epoch": 0.713882041724283,
      "flos": 15558262692480.0,
      "grad_norm": 12.020479750260124,
      "language_loss": 0.83558595,
      "learning_rate": 7.988667915825605e-07,
      "loss": 0.8574369,
      "num_input_tokens_seen": 127608690,
      "step": 5937,
      "time_per_iteration": 3.234461545944214
    },
    {
      "auxiliary_loss_clip": 0.01139117,
      "auxiliary_loss_mlp": 0.01025049,
      "balance_loss_clip": 1.04517698,
      "balance_loss_mlp": 1.01760483,
      "epoch": 0.7140022846149221,
      "flos": 24061011477120.0,
      "grad_norm": 2.190641409999294,
      "language_loss": 0.75256544,
      "learning_rate": 7.982440366567491e-07,
      "loss": 0.77420712,
      "num_input_tokens_seen": 127627180,
      "step": 5938,
      "time_per_iteration": 2.487715482711792
    },
    {
      "auxiliary_loss_clip": 0.01148037,
      "auxiliary_loss_mlp": 0.01023534,
      "balance_loss_clip": 1.04544067,
      "balance_loss_mlp": 1.01628566,
      "epoch": 0.7141225275055613,
      "flos": 27891710248320.0,
      "grad_norm": 1.798389852428495,
      "language_loss": 0.75367653,
      "learning_rate": 7.97621464036361e-07,
      "loss": 0.77539229,
      "num_input_tokens_seen": 127648940,
      "step": 5939,
      "time_per_iteration": 3.3337647914886475
    },
    {
      "auxiliary_loss_clip": 0.01156462,
      "auxiliary_loss_mlp": 0.01023416,
      "balance_loss_clip": 1.04747534,
      "balance_loss_mlp": 1.01582873,
      "epoch": 0.7142427703962003,
      "flos": 19682603147520.0,
      "grad_norm": 1.612917090763218,
      "language_loss": 0.68116152,
      "learning_rate": 7.969990738158417e-07,
      "loss": 0.70296025,
      "num_input_tokens_seen": 127667350,
      "step": 5940,
      "time_per_iteration": 2.427645206451416
    },
    {
      "auxiliary_loss_clip": 0.0115693,
      "auxiliary_loss_mlp": 0.01025578,
      "balance_loss_clip": 1.04932261,
      "balance_loss_mlp": 1.01804423,
      "epoch": 0.7143630132868394,
      "flos": 21032377447680.0,
      "grad_norm": 2.197820580097943,
      "language_loss": 0.85094392,
      "learning_rate": 7.963768660896062e-07,
      "loss": 0.87276906,
      "num_input_tokens_seen": 127685760,
      "step": 5941,
      "time_per_iteration": 3.181696653366089
    },
    {
      "auxiliary_loss_clip": 0.01156318,
      "auxiliary_loss_mlp": 0.01025819,
      "balance_loss_clip": 1.04737258,
      "balance_loss_mlp": 1.01796341,
      "epoch": 0.7144832561774785,
      "flos": 24129923719680.0,
      "grad_norm": 2.1006198240667793,
      "language_loss": 0.82558203,
      "learning_rate": 7.957548409520432e-07,
      "loss": 0.84740341,
      "num_input_tokens_seen": 127704985,
      "step": 5942,
      "time_per_iteration": 2.446225643157959
    },
    {
      "auxiliary_loss_clip": 0.01125794,
      "auxiliary_loss_mlp": 0.01020061,
      "balance_loss_clip": 1.04293668,
      "balance_loss_mlp": 1.01313806,
      "epoch": 0.7146034990681176,
      "flos": 16325817442560.0,
      "grad_norm": 3.5090619472832847,
      "language_loss": 0.84121609,
      "learning_rate": 7.951329984975135e-07,
      "loss": 0.86267465,
      "num_input_tokens_seen": 127721925,
      "step": 5943,
      "time_per_iteration": 2.471788167953491
    },
    {
      "auxiliary_loss_clip": 0.01041298,
      "auxiliary_loss_mlp": 0.01001847,
      "balance_loss_clip": 1.01241088,
      "balance_loss_mlp": 1.00081563,
      "epoch": 0.7147237419587567,
      "flos": 69627164232960.0,
      "grad_norm": 0.7138321493804686,
      "language_loss": 0.54321826,
      "learning_rate": 7.94511338820349e-07,
      "loss": 0.56364971,
      "num_input_tokens_seen": 127784230,
      "step": 5944,
      "time_per_iteration": 3.0834860801696777
    },
    {
      "auxiliary_loss_clip": 0.01140484,
      "auxiliary_loss_mlp": 0.00762358,
      "balance_loss_clip": 1.04580259,
      "balance_loss_mlp": 1.0003413,
      "epoch": 0.7148439848493958,
      "flos": 22266806198400.0,
      "grad_norm": 2.297531839271508,
      "language_loss": 0.78409398,
      "learning_rate": 7.938898620148575e-07,
      "loss": 0.80312246,
      "num_input_tokens_seen": 127801990,
      "step": 5945,
      "time_per_iteration": 2.484203815460205
    },
    {
      "auxiliary_loss_clip": 0.01139539,
      "auxiliary_loss_mlp": 0.0102522,
      "balance_loss_clip": 1.0464921,
      "balance_loss_mlp": 1.01804686,
      "epoch": 0.7149642277400349,
      "flos": 17931383470080.0,
      "grad_norm": 1.9867961615757628,
      "language_loss": 0.70704776,
      "learning_rate": 7.932685681753135e-07,
      "loss": 0.72869533,
      "num_input_tokens_seen": 127819270,
      "step": 5946,
      "time_per_iteration": 2.4448025226593018
    },
    {
      "auxiliary_loss_clip": 0.01164645,
      "auxiliary_loss_mlp": 0.01021362,
      "balance_loss_clip": 1.04844403,
      "balance_loss_mlp": 1.01468062,
      "epoch": 0.7150844706306739,
      "flos": 31681937370240.0,
      "grad_norm": 2.025471147651801,
      "language_loss": 0.62650669,
      "learning_rate": 7.92647457395969e-07,
      "loss": 0.64836669,
      "num_input_tokens_seen": 127841095,
      "step": 5947,
      "time_per_iteration": 2.5029704570770264
    },
    {
      "auxiliary_loss_clip": 0.01104954,
      "auxiliary_loss_mlp": 0.01026908,
      "balance_loss_clip": 1.03982544,
      "balance_loss_mlp": 1.01941872,
      "epoch": 0.7152047135213131,
      "flos": 10926217451520.0,
      "grad_norm": 2.033030334656612,
      "language_loss": 0.74778056,
      "learning_rate": 7.920265297710444e-07,
      "loss": 0.76909918,
      "num_input_tokens_seen": 127858485,
      "step": 5948,
      "time_per_iteration": 2.55680775642395
    },
    {
      "auxiliary_loss_clip": 0.01155035,
      "auxiliary_loss_mlp": 0.01028357,
      "balance_loss_clip": 1.04873919,
      "balance_loss_mlp": 1.02104664,
      "epoch": 0.7153249564119522,
      "flos": 20995640812800.0,
      "grad_norm": 1.8020935211084854,
      "language_loss": 0.73254424,
      "learning_rate": 7.914057853947363e-07,
      "loss": 0.75437814,
      "num_input_tokens_seen": 127877665,
      "step": 5949,
      "time_per_iteration": 2.4449634552001953
    },
    {
      "auxiliary_loss_clip": 0.01126886,
      "auxiliary_loss_mlp": 0.01030701,
      "balance_loss_clip": 1.04588938,
      "balance_loss_mlp": 1.02309573,
      "epoch": 0.7154451993025912,
      "flos": 24243114453120.0,
      "grad_norm": 2.3097265687939466,
      "language_loss": 0.62464869,
      "learning_rate": 7.907852243612089e-07,
      "loss": 0.6462245,
      "num_input_tokens_seen": 127898070,
      "step": 5950,
      "time_per_iteration": 2.5632383823394775
    },
    {
      "auxiliary_loss_clip": 0.01137629,
      "auxiliary_loss_mlp": 0.01024004,
      "balance_loss_clip": 1.04442382,
      "balance_loss_mlp": 1.01710463,
      "epoch": 0.7155654421932304,
      "flos": 23330947547520.0,
      "grad_norm": 2.4445920557105323,
      "language_loss": 0.72456896,
      "learning_rate": 7.901648467646009e-07,
      "loss": 0.7461853,
      "num_input_tokens_seen": 127917010,
      "step": 5951,
      "time_per_iteration": 2.4919145107269287
    },
    {
      "auxiliary_loss_clip": 0.01170737,
      "auxiliary_loss_mlp": 0.01028093,
      "balance_loss_clip": 1.05073297,
      "balance_loss_mlp": 1.02089024,
      "epoch": 0.7156856850838694,
      "flos": 22711883621760.0,
      "grad_norm": 1.513611795855191,
      "language_loss": 0.72323102,
      "learning_rate": 7.895446526990244e-07,
      "loss": 0.74521929,
      "num_input_tokens_seen": 127937025,
      "step": 5952,
      "time_per_iteration": 2.431321859359741
    },
    {
      "auxiliary_loss_clip": 0.0112192,
      "auxiliary_loss_mlp": 0.01026386,
      "balance_loss_clip": 1.04544544,
      "balance_loss_mlp": 1.01892948,
      "epoch": 0.7158059279745085,
      "flos": 19865424395520.0,
      "grad_norm": 1.4918191436637456,
      "language_loss": 0.75417376,
      "learning_rate": 7.889246422585609e-07,
      "loss": 0.77565682,
      "num_input_tokens_seen": 127956410,
      "step": 5953,
      "time_per_iteration": 2.5446574687957764
    },
    {
      "auxiliary_loss_clip": 0.01169615,
      "auxiliary_loss_mlp": 0.01022095,
      "balance_loss_clip": 1.05043077,
      "balance_loss_mlp": 1.01537204,
      "epoch": 0.7159261708651476,
      "flos": 24134772055680.0,
      "grad_norm": 2.0535115772754033,
      "language_loss": 0.73640311,
      "learning_rate": 7.883048155372675e-07,
      "loss": 0.75832021,
      "num_input_tokens_seen": 127974925,
      "step": 5954,
      "time_per_iteration": 2.438392400741577
    },
    {
      "auxiliary_loss_clip": 0.0114473,
      "auxiliary_loss_mlp": 0.01023021,
      "balance_loss_clip": 1.04652762,
      "balance_loss_mlp": 1.01592767,
      "epoch": 0.7160464137557867,
      "flos": 16983198201600.0,
      "grad_norm": 2.2442108113923274,
      "language_loss": 0.71274406,
      "learning_rate": 7.876851726291698e-07,
      "loss": 0.73442155,
      "num_input_tokens_seen": 127993225,
      "step": 5955,
      "time_per_iteration": 2.462009906768799
    },
    {
      "auxiliary_loss_clip": 0.01129355,
      "auxiliary_loss_mlp": 0.01023029,
      "balance_loss_clip": 1.04402018,
      "balance_loss_mlp": 1.01588583,
      "epoch": 0.7161666566464258,
      "flos": 25228251838080.0,
      "grad_norm": 1.9405708162473427,
      "language_loss": 0.78466296,
      "learning_rate": 7.870657136282666e-07,
      "loss": 0.8061868,
      "num_input_tokens_seen": 128012085,
      "step": 5956,
      "time_per_iteration": 2.586334705352783
    },
    {
      "auxiliary_loss_clip": 0.0114964,
      "auxiliary_loss_mlp": 0.01026719,
      "balance_loss_clip": 1.04631352,
      "balance_loss_mlp": 1.01942682,
      "epoch": 0.7162868995370649,
      "flos": 26468390851200.0,
      "grad_norm": 1.5535642698682754,
      "language_loss": 0.81967443,
      "learning_rate": 7.86446438628531e-07,
      "loss": 0.84143806,
      "num_input_tokens_seen": 128033155,
      "step": 5957,
      "time_per_iteration": 2.480459213256836
    },
    {
      "auxiliary_loss_clip": 0.010693,
      "auxiliary_loss_mlp": 0.01001776,
      "balance_loss_clip": 1.01337409,
      "balance_loss_mlp": 1.00078666,
      "epoch": 0.716407142427704,
      "flos": 69998912040960.0,
      "grad_norm": 0.7706519337822517,
      "language_loss": 0.56893235,
      "learning_rate": 7.858273477239059e-07,
      "loss": 0.58964312,
      "num_input_tokens_seen": 128101575,
      "step": 5958,
      "time_per_iteration": 3.028585433959961
    },
    {
      "auxiliary_loss_clip": 0.01097445,
      "auxiliary_loss_mlp": 0.01025699,
      "balance_loss_clip": 1.04128027,
      "balance_loss_mlp": 1.01806951,
      "epoch": 0.716527385318343,
      "flos": 20740459616640.0,
      "grad_norm": 3.981616231933726,
      "language_loss": 0.71342969,
      "learning_rate": 7.852084410083067e-07,
      "loss": 0.73466116,
      "num_input_tokens_seen": 128120395,
      "step": 5959,
      "time_per_iteration": 2.529221773147583
    },
    {
      "auxiliary_loss_clip": 0.01135568,
      "auxiliary_loss_mlp": 0.01023634,
      "balance_loss_clip": 1.04644775,
      "balance_loss_mlp": 1.01680613,
      "epoch": 0.7166476282089821,
      "flos": 25371966153600.0,
      "grad_norm": 1.5823201993328826,
      "language_loss": 0.63586152,
      "learning_rate": 7.84589718575621e-07,
      "loss": 0.6574536,
      "num_input_tokens_seen": 128140840,
      "step": 5960,
      "time_per_iteration": 2.499711513519287
    },
    {
      "auxiliary_loss_clip": 0.01139584,
      "auxiliary_loss_mlp": 0.01024692,
      "balance_loss_clip": 1.04140449,
      "balance_loss_mlp": 1.01726508,
      "epoch": 0.7167678710996213,
      "flos": 24133730561280.0,
      "grad_norm": 2.051914502092473,
      "language_loss": 0.69125974,
      "learning_rate": 7.83971180519708e-07,
      "loss": 0.71290255,
      "num_input_tokens_seen": 128159695,
      "step": 5961,
      "time_per_iteration": 2.485414743423462
    },
    {
      "auxiliary_loss_clip": 0.01171054,
      "auxiliary_loss_mlp": 0.01020219,
      "balance_loss_clip": 1.05058181,
      "balance_loss_mlp": 1.01269412,
      "epoch": 0.7168881139902603,
      "flos": 30226586019840.0,
      "grad_norm": 1.9783952506312736,
      "language_loss": 0.75783879,
      "learning_rate": 7.833528269344008e-07,
      "loss": 0.77975154,
      "num_input_tokens_seen": 128179600,
      "step": 5962,
      "time_per_iteration": 2.4719698429107666
    },
    {
      "auxiliary_loss_clip": 0.01127714,
      "auxiliary_loss_mlp": 0.01027687,
      "balance_loss_clip": 1.0478543,
      "balance_loss_mlp": 1.02006364,
      "epoch": 0.7170083568808994,
      "flos": 14606414236800.0,
      "grad_norm": 2.19147052371818,
      "language_loss": 0.77354544,
      "learning_rate": 7.827346579135023e-07,
      "loss": 0.7950995,
      "num_input_tokens_seen": 128196940,
      "step": 5963,
      "time_per_iteration": 3.235485553741455
    },
    {
      "auxiliary_loss_clip": 0.01132614,
      "auxiliary_loss_mlp": 0.01023625,
      "balance_loss_clip": 1.04109097,
      "balance_loss_mlp": 1.01607275,
      "epoch": 0.7171285997715385,
      "flos": 23331091201920.0,
      "grad_norm": 1.918318510181046,
      "language_loss": 0.83056885,
      "learning_rate": 7.821166735507885e-07,
      "loss": 0.85213125,
      "num_input_tokens_seen": 128215970,
      "step": 5964,
      "time_per_iteration": 3.3208627700805664
    },
    {
      "auxiliary_loss_clip": 0.01166054,
      "auxiliary_loss_mlp": 0.0102293,
      "balance_loss_clip": 1.04807389,
      "balance_loss_mlp": 1.01586342,
      "epoch": 0.7172488426621776,
      "flos": 16543543731840.0,
      "grad_norm": 1.7061114409008664,
      "language_loss": 0.68515599,
      "learning_rate": 7.81498873940007e-07,
      "loss": 0.70704591,
      "num_input_tokens_seen": 128233185,
      "step": 5965,
      "time_per_iteration": 3.24603533744812
    },
    {
      "auxiliary_loss_clip": 0.01157576,
      "auxiliary_loss_mlp": 0.01020295,
      "balance_loss_clip": 1.04477262,
      "balance_loss_mlp": 1.01268053,
      "epoch": 0.7173690855528166,
      "flos": 26541612725760.0,
      "grad_norm": 2.198894462808227,
      "language_loss": 0.77404654,
      "learning_rate": 7.808812591748768e-07,
      "loss": 0.79582524,
      "num_input_tokens_seen": 128253565,
      "step": 5966,
      "time_per_iteration": 2.4849202632904053
    },
    {
      "auxiliary_loss_clip": 0.01120585,
      "auxiliary_loss_mlp": 0.01022783,
      "balance_loss_clip": 1.04278624,
      "balance_loss_mlp": 1.01538587,
      "epoch": 0.7174893284434558,
      "flos": 22784099915520.0,
      "grad_norm": 1.9995247814891484,
      "language_loss": 0.65141243,
      "learning_rate": 7.802638293490915e-07,
      "loss": 0.67284608,
      "num_input_tokens_seen": 128273210,
      "step": 5967,
      "time_per_iteration": 2.511413812637329
    },
    {
      "auxiliary_loss_clip": 0.01141701,
      "auxiliary_loss_mlp": 0.01024165,
      "balance_loss_clip": 1.04485655,
      "balance_loss_mlp": 1.01720881,
      "epoch": 0.7176095713340949,
      "flos": 23293564467840.0,
      "grad_norm": 1.607539206124552,
      "language_loss": 0.76828629,
      "learning_rate": 7.796465845563123e-07,
      "loss": 0.78994495,
      "num_input_tokens_seen": 128292085,
      "step": 5968,
      "time_per_iteration": 3.2865138053894043
    },
    {
      "auxiliary_loss_clip": 0.01133691,
      "auxiliary_loss_mlp": 0.00762008,
      "balance_loss_clip": 1.04482985,
      "balance_loss_mlp": 1.00039935,
      "epoch": 0.7177298142247339,
      "flos": 25591631777280.0,
      "grad_norm": 1.9185757109855452,
      "language_loss": 0.79618633,
      "learning_rate": 7.790295248901766e-07,
      "loss": 0.81514329,
      "num_input_tokens_seen": 128313215,
      "step": 5969,
      "time_per_iteration": 2.513559103012085
    },
    {
      "auxiliary_loss_clip": 0.01154186,
      "auxiliary_loss_mlp": 0.01023751,
      "balance_loss_clip": 1.04812098,
      "balance_loss_mlp": 1.01649463,
      "epoch": 0.7178500571153731,
      "flos": 31652778504960.0,
      "grad_norm": 3.232229592781444,
      "language_loss": 0.62354678,
      "learning_rate": 7.784126504442902e-07,
      "loss": 0.64532614,
      "num_input_tokens_seen": 128336445,
      "step": 5970,
      "time_per_iteration": 2.550555944442749
    },
    {
      "auxiliary_loss_clip": 0.01116124,
      "auxiliary_loss_mlp": 0.01019245,
      "balance_loss_clip": 1.04328299,
      "balance_loss_mlp": 1.01193142,
      "epoch": 0.7179703000060121,
      "flos": 19427242383360.0,
      "grad_norm": 1.431572761934135,
      "language_loss": 0.67669272,
      "learning_rate": 7.777959613122351e-07,
      "loss": 0.69804645,
      "num_input_tokens_seen": 128356270,
      "step": 5971,
      "time_per_iteration": 2.496440887451172
    },
    {
      "auxiliary_loss_clip": 0.01131966,
      "auxiliary_loss_mlp": 0.01025844,
      "balance_loss_clip": 1.04554319,
      "balance_loss_mlp": 1.01872134,
      "epoch": 0.7180905428966512,
      "flos": 28839249072000.0,
      "grad_norm": 1.8246313015915578,
      "language_loss": 0.77922994,
      "learning_rate": 7.771794575875604e-07,
      "loss": 0.80080807,
      "num_input_tokens_seen": 128378140,
      "step": 5972,
      "time_per_iteration": 2.5360546112060547
    },
    {
      "auxiliary_loss_clip": 0.01156815,
      "auxiliary_loss_mlp": 0.01029284,
      "balance_loss_clip": 1.05040479,
      "balance_loss_mlp": 1.02143455,
      "epoch": 0.7182107857872904,
      "flos": 20047563285120.0,
      "grad_norm": 2.3693136504902492,
      "language_loss": 0.77600729,
      "learning_rate": 7.765631393637888e-07,
      "loss": 0.79786831,
      "num_input_tokens_seen": 128396335,
      "step": 5973,
      "time_per_iteration": 2.480060577392578
    },
    {
      "auxiliary_loss_clip": 0.01148786,
      "auxiliary_loss_mlp": 0.01022247,
      "balance_loss_clip": 1.04402542,
      "balance_loss_mlp": 1.01473045,
      "epoch": 0.7183310286779294,
      "flos": 22747686503040.0,
      "grad_norm": 2.7891524672140107,
      "language_loss": 0.48781639,
      "learning_rate": 7.75947006734417e-07,
      "loss": 0.50952673,
      "num_input_tokens_seen": 128414115,
      "step": 5974,
      "time_per_iteration": 2.4523487091064453
    },
    {
      "auxiliary_loss_clip": 0.01166098,
      "auxiliary_loss_mlp": 0.01025511,
      "balance_loss_clip": 1.04692411,
      "balance_loss_mlp": 1.01825094,
      "epoch": 0.7184512715685685,
      "flos": 17158262112000.0,
      "grad_norm": 1.9762087871114307,
      "language_loss": 0.82911795,
      "learning_rate": 7.753310597929101e-07,
      "loss": 0.85103405,
      "num_input_tokens_seen": 128430755,
      "step": 5975,
      "time_per_iteration": 2.363180160522461
    },
    {
      "auxiliary_loss_clip": 0.01069045,
      "auxiliary_loss_mlp": 0.01001057,
      "balance_loss_clip": 1.01330996,
      "balance_loss_mlp": 1.00007999,
      "epoch": 0.7185715144592076,
      "flos": 65509611448320.0,
      "grad_norm": 0.7588439268693543,
      "language_loss": 0.55146426,
      "learning_rate": 7.747152986327095e-07,
      "loss": 0.57216531,
      "num_input_tokens_seen": 128491300,
      "step": 5976,
      "time_per_iteration": 2.930744171142578
    },
    {
      "auxiliary_loss_clip": 0.01115246,
      "auxiliary_loss_mlp": 0.01025427,
      "balance_loss_clip": 1.04405522,
      "balance_loss_mlp": 1.01831925,
      "epoch": 0.7186917573498467,
      "flos": 16180522928640.0,
      "grad_norm": 1.7968503572306591,
      "language_loss": 0.68002009,
      "learning_rate": 7.740997233472228e-07,
      "loss": 0.70142674,
      "num_input_tokens_seen": 128508920,
      "step": 5977,
      "time_per_iteration": 2.4975147247314453
    },
    {
      "auxiliary_loss_clip": 0.0113971,
      "auxiliary_loss_mlp": 0.01019987,
      "balance_loss_clip": 1.04523945,
      "balance_loss_mlp": 1.01359463,
      "epoch": 0.7188120002404857,
      "flos": 29242274647680.0,
      "grad_norm": 2.028477166157526,
      "language_loss": 0.70545155,
      "learning_rate": 7.734843340298329e-07,
      "loss": 0.72704852,
      "num_input_tokens_seen": 128528745,
      "step": 5978,
      "time_per_iteration": 2.556546211242676
    },
    {
      "auxiliary_loss_clip": 0.01143963,
      "auxiliary_loss_mlp": 0.01029059,
      "balance_loss_clip": 1.04546499,
      "balance_loss_mlp": 1.02142322,
      "epoch": 0.7189322431311249,
      "flos": 33401161008000.0,
      "grad_norm": 1.8227730176582633,
      "language_loss": 0.7514649,
      "learning_rate": 7.72869130773895e-07,
      "loss": 0.77319515,
      "num_input_tokens_seen": 128549345,
      "step": 5979,
      "time_per_iteration": 2.580125093460083
    },
    {
      "auxiliary_loss_clip": 0.0106055,
      "auxiliary_loss_mlp": 0.01001273,
      "balance_loss_clip": 1.01362038,
      "balance_loss_mlp": 1.00025368,
      "epoch": 0.719052486021764,
      "flos": 61351263792000.0,
      "grad_norm": 0.7841740392847185,
      "language_loss": 0.59397566,
      "learning_rate": 7.722541136727343e-07,
      "loss": 0.61459386,
      "num_input_tokens_seen": 128605360,
      "step": 5980,
      "time_per_iteration": 2.910393238067627
    },
    {
      "auxiliary_loss_clip": 0.01153463,
      "auxiliary_loss_mlp": 0.0102267,
      "balance_loss_clip": 1.0480113,
      "balance_loss_mlp": 1.015306,
      "epoch": 0.719172728912403,
      "flos": 15596795007360.0,
      "grad_norm": 4.4258872686997925,
      "language_loss": 0.80874044,
      "learning_rate": 7.716392828196483e-07,
      "loss": 0.83050179,
      "num_input_tokens_seen": 128623160,
      "step": 5981,
      "time_per_iteration": 2.4103119373321533
    },
    {
      "auxiliary_loss_clip": 0.01154508,
      "auxiliary_loss_mlp": 0.01026728,
      "balance_loss_clip": 1.04875374,
      "balance_loss_mlp": 1.01968312,
      "epoch": 0.7192929718030422,
      "flos": 15553162961280.0,
      "grad_norm": 2.3907294685851266,
      "language_loss": 0.76855034,
      "learning_rate": 7.710246383079064e-07,
      "loss": 0.79036266,
      "num_input_tokens_seen": 128638545,
      "step": 5982,
      "time_per_iteration": 2.41166090965271
    },
    {
      "auxiliary_loss_clip": 0.01141437,
      "auxiliary_loss_mlp": 0.01024494,
      "balance_loss_clip": 1.04382706,
      "balance_loss_mlp": 1.01710343,
      "epoch": 0.7194132146936812,
      "flos": 21862487733120.0,
      "grad_norm": 2.7910990426610023,
      "language_loss": 0.92155468,
      "learning_rate": 7.704101802307492e-07,
      "loss": 0.94321406,
      "num_input_tokens_seen": 128650845,
      "step": 5983,
      "time_per_iteration": 2.452523708343506
    },
    {
      "auxiliary_loss_clip": 0.01117713,
      "auxiliary_loss_mlp": 0.01029865,
      "balance_loss_clip": 1.04321396,
      "balance_loss_mlp": 1.02182698,
      "epoch": 0.7195334575843203,
      "flos": 27338900958720.0,
      "grad_norm": 2.079848905110411,
      "language_loss": 0.86884528,
      "learning_rate": 7.697959086813912e-07,
      "loss": 0.89032108,
      "num_input_tokens_seen": 128667010,
      "step": 5984,
      "time_per_iteration": 2.530970335006714
    },
    {
      "auxiliary_loss_clip": 0.01117591,
      "auxiliary_loss_mlp": 0.01025737,
      "balance_loss_clip": 1.04199338,
      "balance_loss_mlp": 1.01849818,
      "epoch": 0.7196537004749595,
      "flos": 18770615809920.0,
      "grad_norm": 1.7202020209522562,
      "language_loss": 0.79658598,
      "learning_rate": 7.691818237530145e-07,
      "loss": 0.81801927,
      "num_input_tokens_seen": 128685870,
      "step": 5985,
      "time_per_iteration": 2.4783823490142822
    },
    {
      "auxiliary_loss_clip": 0.01124029,
      "auxiliary_loss_mlp": 0.01024653,
      "balance_loss_clip": 1.0443089,
      "balance_loss_mlp": 1.0174644,
      "epoch": 0.7197739433655985,
      "flos": 24531009960960.0,
      "grad_norm": 3.347365544878862,
      "language_loss": 0.77560627,
      "learning_rate": 7.685679255387774e-07,
      "loss": 0.79709309,
      "num_input_tokens_seen": 128704185,
      "step": 5986,
      "time_per_iteration": 2.5717742443084717
    },
    {
      "auxiliary_loss_clip": 0.01137741,
      "auxiliary_loss_mlp": 0.01025392,
      "balance_loss_clip": 1.04616594,
      "balance_loss_mlp": 1.01807249,
      "epoch": 0.7198941862562376,
      "flos": 18040587793920.0,
      "grad_norm": 1.9515849246077535,
      "language_loss": 0.7704761,
      "learning_rate": 7.679542141318065e-07,
      "loss": 0.79210746,
      "num_input_tokens_seen": 128721290,
      "step": 5987,
      "time_per_iteration": 2.460951566696167
    },
    {
      "auxiliary_loss_clip": 0.0112759,
      "auxiliary_loss_mlp": 0.0102375,
      "balance_loss_clip": 1.04280448,
      "balance_loss_mlp": 1.01686835,
      "epoch": 0.7200144291468767,
      "flos": 29022393542400.0,
      "grad_norm": 2.985615211994362,
      "language_loss": 0.75820959,
      "learning_rate": 7.673406896252013e-07,
      "loss": 0.77972293,
      "num_input_tokens_seen": 128742665,
      "step": 5988,
      "time_per_iteration": 2.5501222610473633
    },
    {
      "auxiliary_loss_clip": 0.01123739,
      "auxiliary_loss_mlp": 0.01028472,
      "balance_loss_clip": 1.04211879,
      "balance_loss_mlp": 1.02016938,
      "epoch": 0.7201346720375158,
      "flos": 25374264624000.0,
      "grad_norm": 1.7704955440651207,
      "language_loss": 0.7838248,
      "learning_rate": 7.667273521120347e-07,
      "loss": 0.80534697,
      "num_input_tokens_seen": 128762225,
      "step": 5989,
      "time_per_iteration": 3.335773229598999
    },
    {
      "auxiliary_loss_clip": 0.01130774,
      "auxiliary_loss_mlp": 0.01028129,
      "balance_loss_clip": 1.04635978,
      "balance_loss_mlp": 1.02109003,
      "epoch": 0.7202549149281549,
      "flos": 14355614499840.0,
      "grad_norm": 1.9877454207093774,
      "language_loss": 0.79988301,
      "learning_rate": 7.661142016853468e-07,
      "loss": 0.82147205,
      "num_input_tokens_seen": 128779585,
      "step": 5990,
      "time_per_iteration": 3.3239893913269043
    },
    {
      "auxiliary_loss_clip": 0.01110229,
      "auxiliary_loss_mlp": 0.01025868,
      "balance_loss_clip": 1.042992,
      "balance_loss_mlp": 1.01868272,
      "epoch": 0.7203751578187939,
      "flos": 23001682550400.0,
      "grad_norm": 1.7743199523224717,
      "language_loss": 0.74498689,
      "learning_rate": 7.655012384381543e-07,
      "loss": 0.76634789,
      "num_input_tokens_seen": 128799070,
      "step": 5991,
      "time_per_iteration": 2.561940908432007
    },
    {
      "auxiliary_loss_clip": 0.01139675,
      "auxiliary_loss_mlp": 0.01028656,
      "balance_loss_clip": 1.05005169,
      "balance_loss_mlp": 1.02133083,
      "epoch": 0.7204954007094331,
      "flos": 23692424065920.0,
      "grad_norm": 1.6525945971799834,
      "language_loss": 0.81969774,
      "learning_rate": 7.648884624634415e-07,
      "loss": 0.84138101,
      "num_input_tokens_seen": 128817620,
      "step": 5992,
      "time_per_iteration": 3.3387362957000732
    },
    {
      "auxiliary_loss_clip": 0.01148888,
      "auxiliary_loss_mlp": 0.01026357,
      "balance_loss_clip": 1.04598236,
      "balance_loss_mlp": 1.01917481,
      "epoch": 0.7206156436000721,
      "flos": 16253026531200.0,
      "grad_norm": 1.7731871169080444,
      "language_loss": 0.88784838,
      "learning_rate": 7.642758738541683e-07,
      "loss": 0.90960085,
      "num_input_tokens_seen": 128834200,
      "step": 5993,
      "time_per_iteration": 2.4055371284484863
    },
    {
      "auxiliary_loss_clip": 0.01058743,
      "auxiliary_loss_mlp": 0.01001506,
      "balance_loss_clip": 1.01376402,
      "balance_loss_mlp": 1.00054014,
      "epoch": 0.7207358864907112,
      "flos": 54377806504320.0,
      "grad_norm": 0.7572707582816807,
      "language_loss": 0.60749763,
      "learning_rate": 7.636634727032621e-07,
      "loss": 0.62810016,
      "num_input_tokens_seen": 128891305,
      "step": 5994,
      "time_per_iteration": 2.8905177116394043
    },
    {
      "auxiliary_loss_clip": 0.01127889,
      "auxiliary_loss_mlp": 0.0102503,
      "balance_loss_clip": 1.04140258,
      "balance_loss_mlp": 1.01720393,
      "epoch": 0.7208561293813504,
      "flos": 19135540033920.0,
      "grad_norm": 1.8721676484559238,
      "language_loss": 0.78841472,
      "learning_rate": 7.630512591036231e-07,
      "loss": 0.80994391,
      "num_input_tokens_seen": 128910615,
      "step": 5995,
      "time_per_iteration": 3.284900188446045
    },
    {
      "auxiliary_loss_clip": 0.01157233,
      "auxiliary_loss_mlp": 0.01029274,
      "balance_loss_clip": 1.04973745,
      "balance_loss_mlp": 1.0220201,
      "epoch": 0.7209763722719894,
      "flos": 17748526308480.0,
      "grad_norm": 2.3773216578585736,
      "language_loss": 0.64676392,
      "learning_rate": 7.624392331481255e-07,
      "loss": 0.66862893,
      "num_input_tokens_seen": 128928270,
      "step": 5996,
      "time_per_iteration": 2.4248831272125244
    },
    {
      "auxiliary_loss_clip": 0.0105985,
      "auxiliary_loss_mlp": 0.01001382,
      "balance_loss_clip": 1.01456189,
      "balance_loss_mlp": 1.00031495,
      "epoch": 0.7210966151626285,
      "flos": 66819488716800.0,
      "grad_norm": 0.7493092913616837,
      "language_loss": 0.51836169,
      "learning_rate": 7.618273949296115e-07,
      "loss": 0.53897393,
      "num_input_tokens_seen": 128987780,
      "step": 5997,
      "time_per_iteration": 2.9322750568389893
    },
    {
      "auxiliary_loss_clip": 0.01133361,
      "auxiliary_loss_mlp": 0.0102533,
      "balance_loss_clip": 1.04344153,
      "balance_loss_mlp": 1.01722407,
      "epoch": 0.7212168580532676,
      "flos": 21141869080320.0,
      "grad_norm": 1.8981321493140701,
      "language_loss": 0.68664795,
      "learning_rate": 7.612157445408987e-07,
      "loss": 0.70823485,
      "num_input_tokens_seen": 129005590,
      "step": 5998,
      "time_per_iteration": 2.4826321601867676
    },
    {
      "auxiliary_loss_clip": 0.01148237,
      "auxiliary_loss_mlp": 0.01028395,
      "balance_loss_clip": 1.05186796,
      "balance_loss_mlp": 1.02081037,
      "epoch": 0.7213371009439067,
      "flos": 22345738335360.0,
      "grad_norm": 2.226406481358763,
      "language_loss": 0.74367869,
      "learning_rate": 7.606042820747716e-07,
      "loss": 0.76544499,
      "num_input_tokens_seen": 129021995,
      "step": 5999,
      "time_per_iteration": 2.4945120811462402
    },
    {
      "auxiliary_loss_clip": 0.01146962,
      "auxiliary_loss_mlp": 0.01024081,
      "balance_loss_clip": 1.05046725,
      "balance_loss_mlp": 1.01663053,
      "epoch": 0.7214573438345457,
      "flos": 18515901490560.0,
      "grad_norm": 2.1359100509397706,
      "language_loss": 0.85404289,
      "learning_rate": 7.599930076239889e-07,
      "loss": 0.87575334,
      "num_input_tokens_seen": 129039280,
      "step": 6000,
      "time_per_iteration": 2.472728967666626
    },
    {
      "auxiliary_loss_clip": 0.01115423,
      "auxiliary_loss_mlp": 0.00761979,
      "balance_loss_clip": 1.04511642,
      "balance_loss_mlp": 1.00037766,
      "epoch": 0.7215775867251849,
      "flos": 35736108606720.0,
      "grad_norm": 2.006742038219633,
      "language_loss": 0.70400584,
      "learning_rate": 7.593819212812818e-07,
      "loss": 0.72277993,
      "num_input_tokens_seen": 129060860,
      "step": 6001,
      "time_per_iteration": 2.66943621635437
    },
    {
      "auxiliary_loss_clip": 0.01153777,
      "auxiliary_loss_mlp": 0.01022394,
      "balance_loss_clip": 1.04854417,
      "balance_loss_mlp": 1.01536119,
      "epoch": 0.721697829615824,
      "flos": 20372410909440.0,
      "grad_norm": 5.027456021302868,
      "language_loss": 0.71434486,
      "learning_rate": 7.587710231393508e-07,
      "loss": 0.73610651,
      "num_input_tokens_seen": 129079215,
      "step": 6002,
      "time_per_iteration": 2.4332945346832275
    },
    {
      "auxiliary_loss_clip": 0.01071424,
      "auxiliary_loss_mlp": 0.01021402,
      "balance_loss_clip": 1.03706491,
      "balance_loss_mlp": 1.01458025,
      "epoch": 0.721818072506463,
      "flos": 20229809915520.0,
      "grad_norm": 1.909431944797739,
      "language_loss": 0.8379997,
      "learning_rate": 7.581603132908685e-07,
      "loss": 0.85892797,
      "num_input_tokens_seen": 129097185,
      "step": 6003,
      "time_per_iteration": 2.5940792560577393
    },
    {
      "auxiliary_loss_clip": 0.01121517,
      "auxiliary_loss_mlp": 0.01022629,
      "balance_loss_clip": 1.04460168,
      "balance_loss_mlp": 1.01512146,
      "epoch": 0.7219383153971022,
      "flos": 18186887888640.0,
      "grad_norm": 1.9426869239432827,
      "language_loss": 0.78648496,
      "learning_rate": 7.575497918284795e-07,
      "loss": 0.80792642,
      "num_input_tokens_seen": 129114730,
      "step": 6004,
      "time_per_iteration": 2.501044511795044
    },
    {
      "auxiliary_loss_clip": 0.01171574,
      "auxiliary_loss_mlp": 0.01029238,
      "balance_loss_clip": 1.04900825,
      "balance_loss_mlp": 1.02178466,
      "epoch": 0.7220585582877412,
      "flos": 17342124854400.0,
      "grad_norm": 1.9389326334801453,
      "language_loss": 0.74713039,
      "learning_rate": 7.569394588447984e-07,
      "loss": 0.76913851,
      "num_input_tokens_seen": 129131745,
      "step": 6005,
      "time_per_iteration": 2.451655387878418
    },
    {
      "auxiliary_loss_clip": 0.01145827,
      "auxiliary_loss_mlp": 0.01025989,
      "balance_loss_clip": 1.04499614,
      "balance_loss_mlp": 1.01891947,
      "epoch": 0.7221788011783803,
      "flos": 16976338704000.0,
      "grad_norm": 2.1875925920998402,
      "language_loss": 0.77720118,
      "learning_rate": 7.563293144324146e-07,
      "loss": 0.79891932,
      "num_input_tokens_seen": 129147295,
      "step": 6006,
      "time_per_iteration": 2.4157588481903076
    },
    {
      "auxiliary_loss_clip": 0.01168018,
      "auxiliary_loss_mlp": 0.0102617,
      "balance_loss_clip": 1.05020463,
      "balance_loss_mlp": 1.01928282,
      "epoch": 0.7222990440690195,
      "flos": 26286359702400.0,
      "grad_norm": 1.7745956050819711,
      "language_loss": 0.80347037,
      "learning_rate": 7.557193586838834e-07,
      "loss": 0.82541227,
      "num_input_tokens_seen": 129162660,
      "step": 6007,
      "time_per_iteration": 2.452289581298828
    },
    {
      "auxiliary_loss_clip": 0.01144258,
      "auxiliary_loss_mlp": 0.01024422,
      "balance_loss_clip": 1.04560399,
      "balance_loss_mlp": 1.0174005,
      "epoch": 0.7224192869596585,
      "flos": 17601687509760.0,
      "grad_norm": 2.063032941343346,
      "language_loss": 0.70465046,
      "learning_rate": 7.551095916917371e-07,
      "loss": 0.72633725,
      "num_input_tokens_seen": 129179990,
      "step": 6008,
      "time_per_iteration": 2.4540679454803467
    },
    {
      "auxiliary_loss_clip": 0.01137213,
      "auxiliary_loss_mlp": 0.01027759,
      "balance_loss_clip": 1.04519868,
      "balance_loss_mlp": 1.01980138,
      "epoch": 0.7225395298502976,
      "flos": 12932331016320.0,
      "grad_norm": 3.9488347740190664,
      "language_loss": 0.66441917,
      "learning_rate": 7.545000135484758e-07,
      "loss": 0.68606889,
      "num_input_tokens_seen": 129197425,
      "step": 6009,
      "time_per_iteration": 2.4877817630767822
    },
    {
      "auxiliary_loss_clip": 0.01169243,
      "auxiliary_loss_mlp": 0.00761983,
      "balance_loss_clip": 1.05022001,
      "balance_loss_mlp": 1.0003767,
      "epoch": 0.7226597727409367,
      "flos": 29643899592960.0,
      "grad_norm": 4.110110059471532,
      "language_loss": 0.62876147,
      "learning_rate": 7.538906243465714e-07,
      "loss": 0.64807373,
      "num_input_tokens_seen": 129217560,
      "step": 6010,
      "time_per_iteration": 2.475687265396118
    },
    {
      "auxiliary_loss_clip": 0.01170204,
      "auxiliary_loss_mlp": 0.01024402,
      "balance_loss_clip": 1.05039358,
      "balance_loss_mlp": 1.01715159,
      "epoch": 0.7227800156315758,
      "flos": 13771635183360.0,
      "grad_norm": 2.1839692433528075,
      "language_loss": 0.78865159,
      "learning_rate": 7.5328142417847e-07,
      "loss": 0.81059766,
      "num_input_tokens_seen": 129234325,
      "step": 6011,
      "time_per_iteration": 2.3999295234680176
    },
    {
      "auxiliary_loss_clip": 0.0114998,
      "auxiliary_loss_mlp": 0.01030784,
      "balance_loss_clip": 1.04466867,
      "balance_loss_mlp": 1.02412021,
      "epoch": 0.7229002585222148,
      "flos": 20301882554880.0,
      "grad_norm": 1.7848167117486438,
      "language_loss": 0.69578463,
      "learning_rate": 7.526724131365838e-07,
      "loss": 0.71759224,
      "num_input_tokens_seen": 129255280,
      "step": 6012,
      "time_per_iteration": 2.4556541442871094
    },
    {
      "auxiliary_loss_clip": 0.01138734,
      "auxiliary_loss_mlp": 0.01028338,
      "balance_loss_clip": 1.04858291,
      "balance_loss_mlp": 1.02079248,
      "epoch": 0.723020501412854,
      "flos": 16581250033920.0,
      "grad_norm": 1.9976839115489198,
      "language_loss": 0.70107841,
      "learning_rate": 7.520635913133017e-07,
      "loss": 0.72274917,
      "num_input_tokens_seen": 129273910,
      "step": 6013,
      "time_per_iteration": 2.4425346851348877
    },
    {
      "auxiliary_loss_clip": 0.01159931,
      "auxiliary_loss_mlp": 0.01028621,
      "balance_loss_clip": 1.04900575,
      "balance_loss_mlp": 1.0204854,
      "epoch": 0.7231407443034931,
      "flos": 28548300908160.0,
      "grad_norm": 1.8870891315437903,
      "language_loss": 0.82634008,
      "learning_rate": 7.514549588009798e-07,
      "loss": 0.84822559,
      "num_input_tokens_seen": 129294785,
      "step": 6014,
      "time_per_iteration": 2.4922797679901123
    },
    {
      "auxiliary_loss_clip": 0.01141797,
      "auxiliary_loss_mlp": 0.01025867,
      "balance_loss_clip": 1.04628873,
      "balance_loss_mlp": 1.01902401,
      "epoch": 0.7232609871941321,
      "flos": 30008536508160.0,
      "grad_norm": 3.5782944014282987,
      "language_loss": 0.70600319,
      "learning_rate": 7.508465156919492e-07,
      "loss": 0.72767979,
      "num_input_tokens_seen": 129318295,
      "step": 6015,
      "time_per_iteration": 2.5513219833374023
    },
    {
      "auxiliary_loss_clip": 0.01141146,
      "auxiliary_loss_mlp": 0.01027078,
      "balance_loss_clip": 1.04583597,
      "balance_loss_mlp": 1.01952028,
      "epoch": 0.7233812300847713,
      "flos": 16654005031680.0,
      "grad_norm": 2.3502057885378798,
      "language_loss": 0.61568433,
      "learning_rate": 7.502382620785083e-07,
      "loss": 0.63736653,
      "num_input_tokens_seen": 129334845,
      "step": 6016,
      "time_per_iteration": 3.2132537364959717
    },
    {
      "auxiliary_loss_clip": 0.01028106,
      "auxiliary_loss_mlp": 0.01004215,
      "balance_loss_clip": 1.01288819,
      "balance_loss_mlp": 1.00329721,
      "epoch": 0.7235014729754103,
      "flos": 67258784050560.0,
      "grad_norm": 0.8044529059662425,
      "language_loss": 0.62498724,
      "learning_rate": 7.496301980529289e-07,
      "loss": 0.64531052,
      "num_input_tokens_seen": 129398055,
      "step": 6017,
      "time_per_iteration": 3.9700419902801514
    },
    {
      "auxiliary_loss_clip": 0.01170349,
      "auxiliary_loss_mlp": 0.01025866,
      "balance_loss_clip": 1.0501442,
      "balance_loss_mlp": 1.01863289,
      "epoch": 0.7236217158660494,
      "flos": 26943237671040.0,
      "grad_norm": 5.784379325592039,
      "language_loss": 0.74407017,
      "learning_rate": 7.490223237074547e-07,
      "loss": 0.76603234,
      "num_input_tokens_seen": 129417765,
      "step": 6018,
      "time_per_iteration": 2.4488601684570312
    },
    {
      "auxiliary_loss_clip": 0.01124949,
      "auxiliary_loss_mlp": 0.01026676,
      "balance_loss_clip": 1.04261518,
      "balance_loss_mlp": 1.01899862,
      "epoch": 0.7237419587566886,
      "flos": 29423372042880.0,
      "grad_norm": 2.0386632418468675,
      "language_loss": 0.65892655,
      "learning_rate": 7.484146391342989e-07,
      "loss": 0.68044281,
      "num_input_tokens_seen": 129437560,
      "step": 6019,
      "time_per_iteration": 3.402531862258911
    },
    {
      "auxiliary_loss_clip": 0.01133786,
      "auxiliary_loss_mlp": 0.01026306,
      "balance_loss_clip": 1.04389572,
      "balance_loss_mlp": 1.01911151,
      "epoch": 0.7238622016473276,
      "flos": 17821496787840.0,
      "grad_norm": 3.1553578170541843,
      "language_loss": 0.56824768,
      "learning_rate": 7.478071444256484e-07,
      "loss": 0.58984864,
      "num_input_tokens_seen": 129455320,
      "step": 6020,
      "time_per_iteration": 2.4340176582336426
    },
    {
      "auxiliary_loss_clip": 0.01136763,
      "auxiliary_loss_mlp": 0.01025542,
      "balance_loss_clip": 1.04581952,
      "balance_loss_mlp": 1.01824927,
      "epoch": 0.7239824445379667,
      "flos": 25739117020800.0,
      "grad_norm": 1.669516718253167,
      "language_loss": 0.79221809,
      "learning_rate": 7.471998396736579e-07,
      "loss": 0.8138411,
      "num_input_tokens_seen": 129475700,
      "step": 6021,
      "time_per_iteration": 3.297128677368164
    },
    {
      "auxiliary_loss_clip": 0.01130012,
      "auxiliary_loss_mlp": 0.01022486,
      "balance_loss_clip": 1.04740191,
      "balance_loss_mlp": 1.01543784,
      "epoch": 0.7241026874286057,
      "flos": 23148916398720.0,
      "grad_norm": 1.6938870151479555,
      "language_loss": 0.75689989,
      "learning_rate": 7.465927249704549e-07,
      "loss": 0.77842486,
      "num_input_tokens_seen": 129493585,
      "step": 6022,
      "time_per_iteration": 2.5260355472564697
    },
    {
      "auxiliary_loss_clip": 0.01153484,
      "auxiliary_loss_mlp": 0.01024684,
      "balance_loss_clip": 1.04785001,
      "balance_loss_mlp": 1.01752806,
      "epoch": 0.7242229303192449,
      "flos": 20266905686400.0,
      "grad_norm": 2.010529393297914,
      "language_loss": 0.77115488,
      "learning_rate": 7.459858004081398e-07,
      "loss": 0.79293656,
      "num_input_tokens_seen": 129511555,
      "step": 6023,
      "time_per_iteration": 2.4605953693389893
    },
    {
      "auxiliary_loss_clip": 0.01029351,
      "auxiliary_loss_mlp": 0.01003216,
      "balance_loss_clip": 1.01182222,
      "balance_loss_mlp": 1.00210714,
      "epoch": 0.724343173209884,
      "flos": 62311659684480.0,
      "grad_norm": 0.6565992248594862,
      "language_loss": 0.58060646,
      "learning_rate": 7.453790660787815e-07,
      "loss": 0.60093212,
      "num_input_tokens_seen": 129579650,
      "step": 6024,
      "time_per_iteration": 3.1811561584472656
    },
    {
      "auxiliary_loss_clip": 0.01142533,
      "auxiliary_loss_mlp": 0.0102236,
      "balance_loss_clip": 1.04668736,
      "balance_loss_mlp": 1.01466513,
      "epoch": 0.724463416100523,
      "flos": 35006403813120.0,
      "grad_norm": 2.487729595746856,
      "language_loss": 0.63646984,
      "learning_rate": 7.447725220744214e-07,
      "loss": 0.65811872,
      "num_input_tokens_seen": 129601895,
      "step": 6025,
      "time_per_iteration": 2.607048988342285
    },
    {
      "auxiliary_loss_clip": 0.01168389,
      "auxiliary_loss_mlp": 0.01028281,
      "balance_loss_clip": 1.04786289,
      "balance_loss_mlp": 1.02077329,
      "epoch": 0.7245836589911622,
      "flos": 21871968923520.0,
      "grad_norm": 2.191457515657719,
      "language_loss": 0.77638066,
      "learning_rate": 7.441661684870717e-07,
      "loss": 0.79834729,
      "num_input_tokens_seen": 129622150,
      "step": 6026,
      "time_per_iteration": 2.4498631954193115
    },
    {
      "auxiliary_loss_clip": 0.01169349,
      "auxiliary_loss_mlp": 0.01020306,
      "balance_loss_clip": 1.0504781,
      "balance_loss_mlp": 1.01339173,
      "epoch": 0.7247039018818012,
      "flos": 23006494972800.0,
      "grad_norm": 1.6908496084677738,
      "language_loss": 0.81659722,
      "learning_rate": 7.435600054087152e-07,
      "loss": 0.8384937,
      "num_input_tokens_seen": 129644315,
      "step": 6027,
      "time_per_iteration": 2.456315279006958
    },
    {
      "auxiliary_loss_clip": 0.01172203,
      "auxiliary_loss_mlp": 0.01029281,
      "balance_loss_clip": 1.05223274,
      "balance_loss_mlp": 1.02198195,
      "epoch": 0.7248241447724403,
      "flos": 31722588587520.0,
      "grad_norm": 1.7903916572246275,
      "language_loss": 0.74406439,
      "learning_rate": 7.42954032931308e-07,
      "loss": 0.76607919,
      "num_input_tokens_seen": 129665355,
      "step": 6028,
      "time_per_iteration": 2.5084571838378906
    },
    {
      "auxiliary_loss_clip": 0.01140779,
      "auxiliary_loss_mlp": 0.0102533,
      "balance_loss_clip": 1.04539967,
      "balance_loss_mlp": 1.01845133,
      "epoch": 0.7249443876630794,
      "flos": 34896984007680.0,
      "grad_norm": 1.8133674041723753,
      "language_loss": 0.74583489,
      "learning_rate": 7.423482511467733e-07,
      "loss": 0.76749599,
      "num_input_tokens_seen": 129686125,
      "step": 6029,
      "time_per_iteration": 2.6122748851776123
    },
    {
      "auxiliary_loss_clip": 0.01087051,
      "auxiliary_loss_mlp": 0.0102633,
      "balance_loss_clip": 1.04073763,
      "balance_loss_mlp": 1.01939464,
      "epoch": 0.7250646305537185,
      "flos": 26359294268160.0,
      "grad_norm": 2.028596163551907,
      "language_loss": 0.64529043,
      "learning_rate": 7.417426601470099e-07,
      "loss": 0.66642416,
      "num_input_tokens_seen": 129706485,
      "step": 6030,
      "time_per_iteration": 2.619476795196533
    },
    {
      "auxiliary_loss_clip": 0.01156154,
      "auxiliary_loss_mlp": 0.01025234,
      "balance_loss_clip": 1.04832625,
      "balance_loss_mlp": 1.01741385,
      "epoch": 0.7251848734443576,
      "flos": 30081614728320.0,
      "grad_norm": 2.068492348662004,
      "language_loss": 0.78497219,
      "learning_rate": 7.411372600238841e-07,
      "loss": 0.80678606,
      "num_input_tokens_seen": 129727100,
      "step": 6031,
      "time_per_iteration": 2.520246982574463
    },
    {
      "auxiliary_loss_clip": 0.01168934,
      "auxiliary_loss_mlp": 0.01027507,
      "balance_loss_clip": 1.04935288,
      "balance_loss_mlp": 1.02019346,
      "epoch": 0.7253051163349967,
      "flos": 17785262943360.0,
      "grad_norm": 1.9766609161773634,
      "language_loss": 0.73686266,
      "learning_rate": 7.405320508692346e-07,
      "loss": 0.75882709,
      "num_input_tokens_seen": 129745840,
      "step": 6032,
      "time_per_iteration": 2.3928418159484863
    },
    {
      "auxiliary_loss_clip": 0.01165948,
      "auxiliary_loss_mlp": 0.01021786,
      "balance_loss_clip": 1.04986668,
      "balance_loss_mlp": 1.01487768,
      "epoch": 0.7254253592256358,
      "flos": 12641346938880.0,
      "grad_norm": 1.7691461649455498,
      "language_loss": 0.75384581,
      "learning_rate": 7.399270327748727e-07,
      "loss": 0.77572316,
      "num_input_tokens_seen": 129763500,
      "step": 6033,
      "time_per_iteration": 2.407437801361084
    },
    {
      "auxiliary_loss_clip": 0.01126669,
      "auxiliary_loss_mlp": 0.00760791,
      "balance_loss_clip": 1.04334259,
      "balance_loss_mlp": 1.00032187,
      "epoch": 0.7255456021162748,
      "flos": 27199208966400.0,
      "grad_norm": 1.8752201488629237,
      "language_loss": 0.74161017,
      "learning_rate": 7.39322205832577e-07,
      "loss": 0.76048476,
      "num_input_tokens_seen": 129784390,
      "step": 6034,
      "time_per_iteration": 2.5662126541137695
    },
    {
      "auxiliary_loss_clip": 0.01136063,
      "auxiliary_loss_mlp": 0.01021565,
      "balance_loss_clip": 1.04540586,
      "balance_loss_mlp": 1.01435924,
      "epoch": 0.725665845006914,
      "flos": 21288205088640.0,
      "grad_norm": 1.9693143105615372,
      "language_loss": 0.80809402,
      "learning_rate": 7.387175701341009e-07,
      "loss": 0.82967037,
      "num_input_tokens_seen": 129803060,
      "step": 6035,
      "time_per_iteration": 2.4947454929351807
    },
    {
      "auxiliary_loss_clip": 0.01153514,
      "auxiliary_loss_mlp": 0.01022176,
      "balance_loss_clip": 1.04659915,
      "balance_loss_mlp": 1.01502991,
      "epoch": 0.7257860878975531,
      "flos": 16033684129920.0,
      "grad_norm": 10.596058797726988,
      "language_loss": 0.72169155,
      "learning_rate": 7.381131257711659e-07,
      "loss": 0.7434485,
      "num_input_tokens_seen": 129820165,
      "step": 6036,
      "time_per_iteration": 2.428905725479126
    },
    {
      "auxiliary_loss_clip": 0.01140422,
      "auxiliary_loss_mlp": 0.01025967,
      "balance_loss_clip": 1.05238068,
      "balance_loss_mlp": 1.0189997,
      "epoch": 0.7259063307881921,
      "flos": 12129943052160.0,
      "grad_norm": 1.7709084545011553,
      "language_loss": 0.83624882,
      "learning_rate": 7.375088728354677e-07,
      "loss": 0.85791272,
      "num_input_tokens_seen": 129835195,
      "step": 6037,
      "time_per_iteration": 2.4450795650482178
    },
    {
      "auxiliary_loss_clip": 0.01129009,
      "auxiliary_loss_mlp": 0.01022151,
      "balance_loss_clip": 1.04504681,
      "balance_loss_mlp": 1.01491451,
      "epoch": 0.7260265736788313,
      "flos": 30443845432320.0,
      "grad_norm": 1.6581015970387019,
      "language_loss": 0.67050523,
      "learning_rate": 7.369048114186691e-07,
      "loss": 0.69201684,
      "num_input_tokens_seen": 129856240,
      "step": 6038,
      "time_per_iteration": 2.5816047191619873
    },
    {
      "auxiliary_loss_clip": 0.01135205,
      "auxiliary_loss_mlp": 0.00761333,
      "balance_loss_clip": 1.04711318,
      "balance_loss_mlp": 1.00029314,
      "epoch": 0.7261468165694703,
      "flos": 21142264129920.0,
      "grad_norm": 1.7516089499832659,
      "language_loss": 0.83453536,
      "learning_rate": 7.363009416124055e-07,
      "loss": 0.85350072,
      "num_input_tokens_seen": 129875565,
      "step": 6039,
      "time_per_iteration": 2.508725166320801
    },
    {
      "auxiliary_loss_clip": 0.01130544,
      "auxiliary_loss_mlp": 0.01023985,
      "balance_loss_clip": 1.04731607,
      "balance_loss_mlp": 1.01664138,
      "epoch": 0.7262670594601094,
      "flos": 22306308180480.0,
      "grad_norm": 2.158664072081652,
      "language_loss": 0.62693131,
      "learning_rate": 7.356972635082852e-07,
      "loss": 0.6484766,
      "num_input_tokens_seen": 129894420,
      "step": 6040,
      "time_per_iteration": 2.508397340774536
    },
    {
      "auxiliary_loss_clip": 0.01112205,
      "auxiliary_loss_mlp": 0.01026033,
      "balance_loss_clip": 1.04809213,
      "balance_loss_mlp": 1.01895809,
      "epoch": 0.7263873023507486,
      "flos": 25335049950720.0,
      "grad_norm": 1.8823020402572586,
      "language_loss": 0.75217694,
      "learning_rate": 7.35093777197884e-07,
      "loss": 0.77355933,
      "num_input_tokens_seen": 129914490,
      "step": 6041,
      "time_per_iteration": 2.575162887573242
    },
    {
      "auxiliary_loss_clip": 0.01139885,
      "auxiliary_loss_mlp": 0.01020626,
      "balance_loss_clip": 1.04828906,
      "balance_loss_mlp": 1.01376522,
      "epoch": 0.7265075452413876,
      "flos": 23878621192320.0,
      "grad_norm": 2.175094438950915,
      "language_loss": 0.86124527,
      "learning_rate": 7.344904827727525e-07,
      "loss": 0.88285047,
      "num_input_tokens_seen": 129931670,
      "step": 6042,
      "time_per_iteration": 2.4986205101013184
    },
    {
      "auxiliary_loss_clip": 0.0112713,
      "auxiliary_loss_mlp": 0.01023941,
      "balance_loss_clip": 1.04285884,
      "balance_loss_mlp": 1.01655054,
      "epoch": 0.7266277881320267,
      "flos": 28724549967360.0,
      "grad_norm": 2.993181674063456,
      "language_loss": 0.73743117,
      "learning_rate": 7.338873803244076e-07,
      "loss": 0.75894189,
      "num_input_tokens_seen": 129946905,
      "step": 6043,
      "time_per_iteration": 3.282459020614624
    },
    {
      "auxiliary_loss_clip": 0.01135136,
      "auxiliary_loss_mlp": 0.0102525,
      "balance_loss_clip": 1.04605055,
      "balance_loss_mlp": 1.01845217,
      "epoch": 0.7267480310226658,
      "flos": 24863507182080.0,
      "grad_norm": 1.7229417007540275,
      "language_loss": 0.80872029,
      "learning_rate": 7.332844699443401e-07,
      "loss": 0.83032417,
      "num_input_tokens_seen": 129965505,
      "step": 6044,
      "time_per_iteration": 3.36826491355896
    },
    {
      "auxiliary_loss_clip": 0.01102204,
      "auxiliary_loss_mlp": 0.01024867,
      "balance_loss_clip": 1.04048014,
      "balance_loss_mlp": 1.01815832,
      "epoch": 0.7268682739133049,
      "flos": 27198490694400.0,
      "grad_norm": 1.9023147716971842,
      "language_loss": 0.75172096,
      "learning_rate": 7.326817517240121e-07,
      "loss": 0.77299166,
      "num_input_tokens_seen": 129987210,
      "step": 6045,
      "time_per_iteration": 2.589280366897583
    },
    {
      "auxiliary_loss_clip": 0.01154823,
      "auxiliary_loss_mlp": 0.00760961,
      "balance_loss_clip": 1.0481807,
      "balance_loss_mlp": 1.00028038,
      "epoch": 0.7269885168039439,
      "flos": 33508138688640.0,
      "grad_norm": 2.189282586823286,
      "language_loss": 0.83345866,
      "learning_rate": 7.320792257548545e-07,
      "loss": 0.85261655,
      "num_input_tokens_seen": 130008385,
      "step": 6046,
      "time_per_iteration": 3.3918864727020264
    },
    {
      "auxiliary_loss_clip": 0.01144733,
      "auxiliary_loss_mlp": 0.01022623,
      "balance_loss_clip": 1.04714429,
      "balance_loss_mlp": 1.0151602,
      "epoch": 0.7271087596945831,
      "flos": 24313750548480.0,
      "grad_norm": 1.8360867878056761,
      "language_loss": 0.76349485,
      "learning_rate": 7.314768921282704e-07,
      "loss": 0.78516841,
      "num_input_tokens_seen": 130029040,
      "step": 6047,
      "time_per_iteration": 2.504153251647949
    },
    {
      "auxiliary_loss_clip": 0.01156181,
      "auxiliary_loss_mlp": 0.01023949,
      "balance_loss_clip": 1.04748607,
      "balance_loss_mlp": 1.01679075,
      "epoch": 0.7272290025852222,
      "flos": 23805147922560.0,
      "grad_norm": 2.825462046756991,
      "language_loss": 0.72337544,
      "learning_rate": 7.30874750935633e-07,
      "loss": 0.74517673,
      "num_input_tokens_seen": 130048725,
      "step": 6048,
      "time_per_iteration": 3.2359619140625
    },
    {
      "auxiliary_loss_clip": 0.01126845,
      "auxiliary_loss_mlp": 0.01023657,
      "balance_loss_clip": 1.04684186,
      "balance_loss_mlp": 1.01652217,
      "epoch": 0.7273492454758612,
      "flos": 16720367408640.0,
      "grad_norm": 4.23380308243205,
      "language_loss": 0.7874791,
      "learning_rate": 7.30272802268286e-07,
      "loss": 0.8089841,
      "num_input_tokens_seen": 130065720,
      "step": 6049,
      "time_per_iteration": 2.487602949142456
    },
    {
      "auxiliary_loss_clip": 0.01071103,
      "auxiliary_loss_mlp": 0.01021433,
      "balance_loss_clip": 1.03615773,
      "balance_loss_mlp": 1.01452446,
      "epoch": 0.7274694883665004,
      "flos": 28031330413440.0,
      "grad_norm": 2.808644567431553,
      "language_loss": 0.7596457,
      "learning_rate": 7.29671046217547e-07,
      "loss": 0.78057104,
      "num_input_tokens_seen": 130084830,
      "step": 6050,
      "time_per_iteration": 2.5993053913116455
    },
    {
      "auxiliary_loss_clip": 0.01128471,
      "auxiliary_loss_mlp": 0.01028324,
      "balance_loss_clip": 1.0463599,
      "balance_loss_mlp": 1.02125168,
      "epoch": 0.7275897312571394,
      "flos": 30372706546560.0,
      "grad_norm": 1.825345955677869,
      "language_loss": 0.81345701,
      "learning_rate": 7.290694828746988e-07,
      "loss": 0.83502495,
      "num_input_tokens_seen": 130104495,
      "step": 6051,
      "time_per_iteration": 2.577407121658325
    },
    {
      "auxiliary_loss_clip": 0.01129597,
      "auxiliary_loss_mlp": 0.01024737,
      "balance_loss_clip": 1.04344559,
      "balance_loss_mlp": 1.01746202,
      "epoch": 0.7277099741477785,
      "flos": 19204775498880.0,
      "grad_norm": 1.964649200838904,
      "language_loss": 0.85851324,
      "learning_rate": 7.284681123310004e-07,
      "loss": 0.88005662,
      "num_input_tokens_seen": 130123210,
      "step": 6052,
      "time_per_iteration": 2.51304030418396
    },
    {
      "auxiliary_loss_clip": 0.01152796,
      "auxiliary_loss_mlp": 0.01028245,
      "balance_loss_clip": 1.04791343,
      "balance_loss_mlp": 1.02045512,
      "epoch": 0.7278302170384175,
      "flos": 20667884186880.0,
      "grad_norm": 2.98817717559006,
      "language_loss": 0.79554474,
      "learning_rate": 7.27866934677678e-07,
      "loss": 0.81735516,
      "num_input_tokens_seen": 130142880,
      "step": 6053,
      "time_per_iteration": 2.4361069202423096
    },
    {
      "auxiliary_loss_clip": 0.01108135,
      "auxiliary_loss_mlp": 0.01022581,
      "balance_loss_clip": 1.04340911,
      "balance_loss_mlp": 1.01555705,
      "epoch": 0.7279504599290567,
      "flos": 19093200877440.0,
      "grad_norm": 1.6930859080041893,
      "language_loss": 0.78456926,
      "learning_rate": 7.272659500059297e-07,
      "loss": 0.80587649,
      "num_input_tokens_seen": 130160220,
      "step": 6054,
      "time_per_iteration": 2.5480964183807373
    },
    {
      "auxiliary_loss_clip": 0.0114776,
      "auxiliary_loss_mlp": 0.01034422,
      "balance_loss_clip": 1.04690599,
      "balance_loss_mlp": 1.02736211,
      "epoch": 0.7280707028196958,
      "flos": 19062174504960.0,
      "grad_norm": 2.661770738953473,
      "language_loss": 0.80178165,
      "learning_rate": 7.266651584069264e-07,
      "loss": 0.82360345,
      "num_input_tokens_seen": 130177885,
      "step": 6055,
      "time_per_iteration": 2.451216459274292
    },
    {
      "auxiliary_loss_clip": 0.01157912,
      "auxiliary_loss_mlp": 0.01022527,
      "balance_loss_clip": 1.0509063,
      "balance_loss_mlp": 1.01553559,
      "epoch": 0.7281909457103348,
      "flos": 37196308293120.0,
      "grad_norm": 1.636765370791178,
      "language_loss": 0.57367694,
      "learning_rate": 7.260645599718045e-07,
      "loss": 0.59548134,
      "num_input_tokens_seen": 130204240,
      "step": 6056,
      "time_per_iteration": 2.5949149131774902
    },
    {
      "auxiliary_loss_clip": 0.01139184,
      "auxiliary_loss_mlp": 0.01029046,
      "balance_loss_clip": 1.04522789,
      "balance_loss_mlp": 1.02120817,
      "epoch": 0.728311188600974,
      "flos": 20667094087680.0,
      "grad_norm": 7.417084390698381,
      "language_loss": 0.674986,
      "learning_rate": 7.254641547916767e-07,
      "loss": 0.69666833,
      "num_input_tokens_seen": 130221735,
      "step": 6057,
      "time_per_iteration": 2.465744733810425
    },
    {
      "auxiliary_loss_clip": 0.01168454,
      "auxiliary_loss_mlp": 0.01030964,
      "balance_loss_clip": 1.05055022,
      "balance_loss_mlp": 1.02382374,
      "epoch": 0.728431431491613,
      "flos": 28840685616000.0,
      "grad_norm": 2.034867113792075,
      "language_loss": 0.69522333,
      "learning_rate": 7.248639429576226e-07,
      "loss": 0.7172175,
      "num_input_tokens_seen": 130241190,
      "step": 6058,
      "time_per_iteration": 2.4541234970092773
    },
    {
      "auxiliary_loss_clip": 0.01153669,
      "auxiliary_loss_mlp": 0.0102727,
      "balance_loss_clip": 1.04661965,
      "balance_loss_mlp": 1.02001333,
      "epoch": 0.7285516743822521,
      "flos": 25991856092160.0,
      "grad_norm": 1.601315687084765,
      "language_loss": 0.71790141,
      "learning_rate": 7.242639245606959e-07,
      "loss": 0.73971081,
      "num_input_tokens_seen": 130260980,
      "step": 6059,
      "time_per_iteration": 2.473053455352783
    },
    {
      "auxiliary_loss_clip": 0.01145635,
      "auxiliary_loss_mlp": 0.01026477,
      "balance_loss_clip": 1.04517055,
      "balance_loss_mlp": 1.01922345,
      "epoch": 0.7286719172728913,
      "flos": 16399721675520.0,
      "grad_norm": 1.8144714658929,
      "language_loss": 0.82548076,
      "learning_rate": 7.236640996919168e-07,
      "loss": 0.84720182,
      "num_input_tokens_seen": 130280025,
      "step": 6060,
      "time_per_iteration": 2.4468722343444824
    },
    {
      "auxiliary_loss_clip": 0.01155341,
      "auxiliary_loss_mlp": 0.01025189,
      "balance_loss_clip": 1.04712296,
      "balance_loss_mlp": 1.01819754,
      "epoch": 0.7287921601635303,
      "flos": 22018161277440.0,
      "grad_norm": 1.5124226124954496,
      "language_loss": 0.7071898,
      "learning_rate": 7.230644684422782e-07,
      "loss": 0.72899508,
      "num_input_tokens_seen": 130300255,
      "step": 6061,
      "time_per_iteration": 2.454136610031128
    },
    {
      "auxiliary_loss_clip": 0.0112672,
      "auxiliary_loss_mlp": 0.01027259,
      "balance_loss_clip": 1.04550767,
      "balance_loss_mlp": 1.02050257,
      "epoch": 0.7289124030541694,
      "flos": 24600927784320.0,
      "grad_norm": 1.712803627920868,
      "language_loss": 0.8133564,
      "learning_rate": 7.224650309027451e-07,
      "loss": 0.83489615,
      "num_input_tokens_seen": 130320005,
      "step": 6062,
      "time_per_iteration": 2.5313501358032227
    },
    {
      "auxiliary_loss_clip": 0.01158991,
      "auxiliary_loss_mlp": 0.01022877,
      "balance_loss_clip": 1.05110061,
      "balance_loss_mlp": 1.01581049,
      "epoch": 0.7290326459448085,
      "flos": 21393638484480.0,
      "grad_norm": 2.899120479963915,
      "language_loss": 0.68734419,
      "learning_rate": 7.218657871642506e-07,
      "loss": 0.70916289,
      "num_input_tokens_seen": 130338810,
      "step": 6063,
      "time_per_iteration": 2.4475221633911133
    },
    {
      "auxiliary_loss_clip": 0.01170728,
      "auxiliary_loss_mlp": 0.01026677,
      "balance_loss_clip": 1.04975176,
      "balance_loss_mlp": 1.01953936,
      "epoch": 0.7291528888354476,
      "flos": 18587686821120.0,
      "grad_norm": 2.0182695687246297,
      "language_loss": 0.62324286,
      "learning_rate": 7.212667373177012e-07,
      "loss": 0.64521694,
      "num_input_tokens_seen": 130353805,
      "step": 6064,
      "time_per_iteration": 2.3907220363616943
    },
    {
      "auxiliary_loss_clip": 0.01125621,
      "auxiliary_loss_mlp": 0.01023874,
      "balance_loss_clip": 1.04461861,
      "balance_loss_mlp": 1.01669705,
      "epoch": 0.7292731317260867,
      "flos": 18951066760320.0,
      "grad_norm": 5.4491074619613,
      "language_loss": 0.75430483,
      "learning_rate": 7.206678814539704e-07,
      "loss": 0.77579975,
      "num_input_tokens_seen": 130372105,
      "step": 6065,
      "time_per_iteration": 2.501145124435425
    },
    {
      "auxiliary_loss_clip": 0.01120068,
      "auxiliary_loss_mlp": 0.0102462,
      "balance_loss_clip": 1.04428017,
      "balance_loss_mlp": 1.01791716,
      "epoch": 0.7293933746167258,
      "flos": 21067569797760.0,
      "grad_norm": 2.071844414826551,
      "language_loss": 0.72687268,
      "learning_rate": 7.20069219663904e-07,
      "loss": 0.74831951,
      "num_input_tokens_seen": 130391990,
      "step": 6066,
      "time_per_iteration": 2.5456364154815674
    },
    {
      "auxiliary_loss_clip": 0.01155773,
      "auxiliary_loss_mlp": 0.01023117,
      "balance_loss_clip": 1.04570138,
      "balance_loss_mlp": 1.01618218,
      "epoch": 0.7295136175073649,
      "flos": 22453326547200.0,
      "grad_norm": 1.775308638986487,
      "language_loss": 0.79362333,
      "learning_rate": 7.1947075203832e-07,
      "loss": 0.81541228,
      "num_input_tokens_seen": 130411970,
      "step": 6067,
      "time_per_iteration": 2.4571807384490967
    },
    {
      "auxiliary_loss_clip": 0.01069948,
      "auxiliary_loss_mlp": 0.01002267,
      "balance_loss_clip": 1.0148766,
      "balance_loss_mlp": 1.00118196,
      "epoch": 0.7296338603980039,
      "flos": 56125506648960.0,
      "grad_norm": 0.8625292430452522,
      "language_loss": 0.60186422,
      "learning_rate": 7.188724786680049e-07,
      "loss": 0.62258625,
      "num_input_tokens_seen": 130472440,
      "step": 6068,
      "time_per_iteration": 3.006516695022583
    },
    {
      "auxiliary_loss_clip": 0.0113927,
      "auxiliary_loss_mlp": 0.01024096,
      "balance_loss_clip": 1.04480898,
      "balance_loss_mlp": 1.0167284,
      "epoch": 0.7297541032886431,
      "flos": 25228287751680.0,
      "grad_norm": 1.7191336596005689,
      "language_loss": 0.75628817,
      "learning_rate": 7.182743996437162e-07,
      "loss": 0.7779218,
      "num_input_tokens_seen": 130491975,
      "step": 6069,
      "time_per_iteration": 2.518704652786255
    },
    {
      "auxiliary_loss_clip": 0.01132922,
      "auxiliary_loss_mlp": 0.01022314,
      "balance_loss_clip": 1.04339182,
      "balance_loss_mlp": 1.01458955,
      "epoch": 0.7298743461792822,
      "flos": 26467600752000.0,
      "grad_norm": 1.8681838726139628,
      "language_loss": 0.6862877,
      "learning_rate": 7.176765150561819e-07,
      "loss": 0.70784003,
      "num_input_tokens_seen": 130510580,
      "step": 6070,
      "time_per_iteration": 4.110658884048462
    },
    {
      "auxiliary_loss_clip": 0.0116921,
      "auxiliary_loss_mlp": 0.01023712,
      "balance_loss_clip": 1.0474968,
      "balance_loss_mlp": 1.0165863,
      "epoch": 0.7299945890699212,
      "flos": 19569053278080.0,
      "grad_norm": 2.196190424696349,
      "language_loss": 0.79589492,
      "learning_rate": 7.170788249961002e-07,
      "loss": 0.81782413,
      "num_input_tokens_seen": 130529090,
      "step": 6071,
      "time_per_iteration": 2.4754576683044434
    },
    {
      "auxiliary_loss_clip": 0.01164138,
      "auxiliary_loss_mlp": 0.01023538,
      "balance_loss_clip": 1.04709721,
      "balance_loss_mlp": 1.01658547,
      "epoch": 0.7301148319605604,
      "flos": 22928963466240.0,
      "grad_norm": 1.8685895341075822,
      "language_loss": 0.88565135,
      "learning_rate": 7.164813295541418e-07,
      "loss": 0.90752816,
      "num_input_tokens_seen": 130548655,
      "step": 6072,
      "time_per_iteration": 3.26410174369812
    },
    {
      "auxiliary_loss_clip": 0.0114143,
      "auxiliary_loss_mlp": 0.01025777,
      "balance_loss_clip": 1.04490077,
      "balance_loss_mlp": 1.01874959,
      "epoch": 0.7302350748511994,
      "flos": 25369703596800.0,
      "grad_norm": 1.607299385512349,
      "language_loss": 0.70380247,
      "learning_rate": 7.15884028820944e-07,
      "loss": 0.72547454,
      "num_input_tokens_seen": 130567710,
      "step": 6073,
      "time_per_iteration": 2.502211570739746
    },
    {
      "auxiliary_loss_clip": 0.01118953,
      "auxiliary_loss_mlp": 0.01021036,
      "balance_loss_clip": 1.0405674,
      "balance_loss_mlp": 1.01415753,
      "epoch": 0.7303553177418385,
      "flos": 27819170732160.0,
      "grad_norm": 2.0068940394597554,
      "language_loss": 0.60835981,
      "learning_rate": 7.152869228871185e-07,
      "loss": 0.62975967,
      "num_input_tokens_seen": 130590195,
      "step": 6074,
      "time_per_iteration": 2.5531303882598877
    },
    {
      "auxiliary_loss_clip": 0.01135023,
      "auxiliary_loss_mlp": 0.01028159,
      "balance_loss_clip": 1.04469085,
      "balance_loss_mlp": 1.02057123,
      "epoch": 0.7304755606324776,
      "flos": 24426510318720.0,
      "grad_norm": 2.0650139150635085,
      "language_loss": 0.72399163,
      "learning_rate": 7.146900118432457e-07,
      "loss": 0.74562347,
      "num_input_tokens_seen": 130609940,
      "step": 6075,
      "time_per_iteration": 3.325765371322632
    },
    {
      "auxiliary_loss_clip": 0.01081949,
      "auxiliary_loss_mlp": 0.01029659,
      "balance_loss_clip": 1.03608203,
      "balance_loss_mlp": 1.02258372,
      "epoch": 0.7305958035231167,
      "flos": 23840483927040.0,
      "grad_norm": 1.6288849391698481,
      "language_loss": 0.85704744,
      "learning_rate": 7.140932957798753e-07,
      "loss": 0.87816358,
      "num_input_tokens_seen": 130628380,
      "step": 6076,
      "time_per_iteration": 2.7142691612243652
    },
    {
      "auxiliary_loss_clip": 0.01142766,
      "auxiliary_loss_mlp": 0.01022122,
      "balance_loss_clip": 1.04408932,
      "balance_loss_mlp": 1.01482594,
      "epoch": 0.7307160464137558,
      "flos": 16726939597440.0,
      "grad_norm": 2.9542674738387116,
      "language_loss": 0.71352798,
      "learning_rate": 7.134967747875309e-07,
      "loss": 0.73517686,
      "num_input_tokens_seen": 130646590,
      "step": 6077,
      "time_per_iteration": 2.49676775932312
    },
    {
      "auxiliary_loss_clip": 0.01149278,
      "auxiliary_loss_mlp": 0.01025124,
      "balance_loss_clip": 1.04496181,
      "balance_loss_mlp": 1.01800728,
      "epoch": 0.7308362893043949,
      "flos": 21798280172160.0,
      "grad_norm": 2.110313043272194,
      "language_loss": 0.8193149,
      "learning_rate": 7.129004489567014e-07,
      "loss": 0.84105891,
      "num_input_tokens_seen": 130664070,
      "step": 6078,
      "time_per_iteration": 2.446838617324829
    },
    {
      "auxiliary_loss_clip": 0.01130748,
      "auxiliary_loss_mlp": 0.01022242,
      "balance_loss_clip": 1.04479504,
      "balance_loss_mlp": 1.01500595,
      "epoch": 0.730956532195034,
      "flos": 10707377840640.0,
      "grad_norm": 2.174305448171294,
      "language_loss": 0.78347278,
      "learning_rate": 7.123043183778512e-07,
      "loss": 0.80500269,
      "num_input_tokens_seen": 130681400,
      "step": 6079,
      "time_per_iteration": 2.5183959007263184
    },
    {
      "auxiliary_loss_clip": 0.01133809,
      "auxiliary_loss_mlp": 0.0103189,
      "balance_loss_clip": 1.04717827,
      "balance_loss_mlp": 1.02471328,
      "epoch": 0.731076775085673,
      "flos": 19791987039360.0,
      "grad_norm": 2.3414133347998134,
      "language_loss": 0.65093482,
      "learning_rate": 7.117083831414114e-07,
      "loss": 0.67259181,
      "num_input_tokens_seen": 130700675,
      "step": 6080,
      "time_per_iteration": 2.5625460147857666
    },
    {
      "auxiliary_loss_clip": 0.0116518,
      "auxiliary_loss_mlp": 0.01022285,
      "balance_loss_clip": 1.04853845,
      "balance_loss_mlp": 1.01529944,
      "epoch": 0.7311970179763122,
      "flos": 20447033414400.0,
      "grad_norm": 1.9020191974456258,
      "language_loss": 0.69872558,
      "learning_rate": 7.11112643337787e-07,
      "loss": 0.72060025,
      "num_input_tokens_seen": 130719720,
      "step": 6081,
      "time_per_iteration": 2.460615873336792
    },
    {
      "auxiliary_loss_clip": 0.01143326,
      "auxiliary_loss_mlp": 0.01030189,
      "balance_loss_clip": 1.0495975,
      "balance_loss_mlp": 1.02253628,
      "epoch": 0.7313172608669513,
      "flos": 18513818501760.0,
      "grad_norm": 2.0916858946149977,
      "language_loss": 0.76452732,
      "learning_rate": 7.10517099057349e-07,
      "loss": 0.78626251,
      "num_input_tokens_seen": 130736670,
      "step": 6082,
      "time_per_iteration": 2.538703203201294
    },
    {
      "auxiliary_loss_clip": 0.01138001,
      "auxiliary_loss_mlp": 0.01021984,
      "balance_loss_clip": 1.04422379,
      "balance_loss_mlp": 1.01442039,
      "epoch": 0.7314375037575903,
      "flos": 16180738410240.0,
      "grad_norm": 3.406099754138424,
      "language_loss": 0.61160922,
      "learning_rate": 7.099217503904411e-07,
      "loss": 0.63320905,
      "num_input_tokens_seen": 130754525,
      "step": 6083,
      "time_per_iteration": 2.470655918121338
    },
    {
      "auxiliary_loss_clip": 0.01142672,
      "auxiliary_loss_mlp": 0.01023461,
      "balance_loss_clip": 1.04683614,
      "balance_loss_mlp": 1.01662695,
      "epoch": 0.7315577466482295,
      "flos": 17967940536960.0,
      "grad_norm": 2.168818214662408,
      "language_loss": 0.90080929,
      "learning_rate": 7.093265974273788e-07,
      "loss": 0.92247063,
      "num_input_tokens_seen": 130772420,
      "step": 6084,
      "time_per_iteration": 2.451982259750366
    },
    {
      "auxiliary_loss_clip": 0.0115271,
      "auxiliary_loss_mlp": 0.01021605,
      "balance_loss_clip": 1.04469919,
      "balance_loss_mlp": 1.01465845,
      "epoch": 0.7316779895388685,
      "flos": 18405440190720.0,
      "grad_norm": 1.8290021345354643,
      "language_loss": 0.72191536,
      "learning_rate": 7.087316402584447e-07,
      "loss": 0.74365848,
      "num_input_tokens_seen": 130791245,
      "step": 6085,
      "time_per_iteration": 2.4172520637512207
    },
    {
      "auxiliary_loss_clip": 0.0116766,
      "auxiliary_loss_mlp": 0.01020727,
      "balance_loss_clip": 1.04932082,
      "balance_loss_mlp": 1.01332378,
      "epoch": 0.7317982324295076,
      "flos": 17928294900480.0,
      "grad_norm": 1.975844527751707,
      "language_loss": 0.863065,
      "learning_rate": 7.081368789738953e-07,
      "loss": 0.88494885,
      "num_input_tokens_seen": 130808445,
      "step": 6086,
      "time_per_iteration": 2.376862049102783
    },
    {
      "auxiliary_loss_clip": 0.01133227,
      "auxiliary_loss_mlp": 0.01023305,
      "balance_loss_clip": 1.0412426,
      "balance_loss_mlp": 1.01580071,
      "epoch": 0.7319184753201466,
      "flos": 27229840289280.0,
      "grad_norm": 2.4064825638920206,
      "language_loss": 0.77936935,
      "learning_rate": 7.075423136639537e-07,
      "loss": 0.80093479,
      "num_input_tokens_seen": 130827700,
      "step": 6087,
      "time_per_iteration": 2.518655300140381
    },
    {
      "auxiliary_loss_clip": 0.01117999,
      "auxiliary_loss_mlp": 0.01023328,
      "balance_loss_clip": 1.04141688,
      "balance_loss_mlp": 1.0159725,
      "epoch": 0.7320387182107858,
      "flos": 37448544574080.0,
      "grad_norm": 2.1011986921515855,
      "language_loss": 0.74276489,
      "learning_rate": 7.069479444188149e-07,
      "loss": 0.7641781,
      "num_input_tokens_seen": 130848290,
      "step": 6088,
      "time_per_iteration": 2.634871482849121
    },
    {
      "auxiliary_loss_clip": 0.01132309,
      "auxiliary_loss_mlp": 0.01023598,
      "balance_loss_clip": 1.04561925,
      "balance_loss_mlp": 1.01604056,
      "epoch": 0.7321589611014249,
      "flos": 17859023521920.0,
      "grad_norm": 1.822385432380353,
      "language_loss": 0.82249892,
      "learning_rate": 7.063537713286453e-07,
      "loss": 0.84405804,
      "num_input_tokens_seen": 130865970,
      "step": 6089,
      "time_per_iteration": 2.4458703994750977
    },
    {
      "auxiliary_loss_clip": 0.01145208,
      "auxiliary_loss_mlp": 0.01022444,
      "balance_loss_clip": 1.04608583,
      "balance_loss_mlp": 1.01515746,
      "epoch": 0.7322792039920639,
      "flos": 26100593539200.0,
      "grad_norm": 1.7472471266449063,
      "language_loss": 0.80879009,
      "learning_rate": 7.057597944835803e-07,
      "loss": 0.83046657,
      "num_input_tokens_seen": 130885245,
      "step": 6090,
      "time_per_iteration": 2.5119845867156982
    },
    {
      "auxiliary_loss_clip": 0.01131904,
      "auxiliary_loss_mlp": 0.01021081,
      "balance_loss_clip": 1.04381573,
      "balance_loss_mlp": 1.01420283,
      "epoch": 0.7323994468827031,
      "flos": 25369093065600.0,
      "grad_norm": 1.550271164377914,
      "language_loss": 0.745031,
      "learning_rate": 7.051660139737253e-07,
      "loss": 0.76656085,
      "num_input_tokens_seen": 130903465,
      "step": 6091,
      "time_per_iteration": 2.5374433994293213
    },
    {
      "auxiliary_loss_clip": 0.01151214,
      "auxiliary_loss_mlp": 0.00761542,
      "balance_loss_clip": 1.04873168,
      "balance_loss_mlp": 1.00032687,
      "epoch": 0.7325196897733421,
      "flos": 26907075653760.0,
      "grad_norm": 2.404974823497621,
      "language_loss": 0.7659322,
      "learning_rate": 7.045724298891565e-07,
      "loss": 0.78505969,
      "num_input_tokens_seen": 130922935,
      "step": 6092,
      "time_per_iteration": 2.483245849609375
    },
    {
      "auxiliary_loss_clip": 0.01152253,
      "auxiliary_loss_mlp": 0.01023174,
      "balance_loss_clip": 1.04722667,
      "balance_loss_mlp": 1.01623547,
      "epoch": 0.7326399326639812,
      "flos": 25775781828480.0,
      "grad_norm": 1.9320790573908304,
      "language_loss": 0.69105494,
      "learning_rate": 7.039790423199192e-07,
      "loss": 0.71280921,
      "num_input_tokens_seen": 130942575,
      "step": 6093,
      "time_per_iteration": 2.464826822280884
    },
    {
      "auxiliary_loss_clip": 0.01142848,
      "auxiliary_loss_mlp": 0.01024626,
      "balance_loss_clip": 1.04637527,
      "balance_loss_mlp": 1.01764011,
      "epoch": 0.7327601755546204,
      "flos": 21032269706880.0,
      "grad_norm": 1.9397062132468301,
      "language_loss": 0.77659988,
      "learning_rate": 7.033858513560322e-07,
      "loss": 0.79827464,
      "num_input_tokens_seen": 130958870,
      "step": 6094,
      "time_per_iteration": 2.446718454360962
    },
    {
      "auxiliary_loss_clip": 0.01153437,
      "auxiliary_loss_mlp": 0.01022726,
      "balance_loss_clip": 1.0477072,
      "balance_loss_mlp": 1.01576138,
      "epoch": 0.7328804184452594,
      "flos": 16289224462080.0,
      "grad_norm": 2.550591191607863,
      "language_loss": 0.76353878,
      "learning_rate": 7.027928570874794e-07,
      "loss": 0.78530037,
      "num_input_tokens_seen": 130977060,
      "step": 6095,
      "time_per_iteration": 2.411625623703003
    },
    {
      "auxiliary_loss_clip": 0.01165811,
      "auxiliary_loss_mlp": 0.01025871,
      "balance_loss_clip": 1.04815173,
      "balance_loss_mlp": 1.0187484,
      "epoch": 0.7330006613358985,
      "flos": 17858233422720.0,
      "grad_norm": 1.8203880795264558,
      "language_loss": 0.85420823,
      "learning_rate": 7.022000596042194e-07,
      "loss": 0.8761251,
      "num_input_tokens_seen": 130994160,
      "step": 6096,
      "time_per_iteration": 2.4382669925689697
    },
    {
      "auxiliary_loss_clip": 0.01126081,
      "auxiliary_loss_mlp": 0.0102531,
      "balance_loss_clip": 1.04124975,
      "balance_loss_mlp": 1.0181396,
      "epoch": 0.7331209042265376,
      "flos": 22492074343680.0,
      "grad_norm": 2.0782956590857826,
      "language_loss": 0.82074994,
      "learning_rate": 7.016074589961784e-07,
      "loss": 0.84226382,
      "num_input_tokens_seen": 131012725,
      "step": 6097,
      "time_per_iteration": 4.066557884216309
    },
    {
      "auxiliary_loss_clip": 0.01137424,
      "auxiliary_loss_mlp": 0.01024848,
      "balance_loss_clip": 1.0472331,
      "balance_loss_mlp": 1.0173645,
      "epoch": 0.7332411471171767,
      "flos": 33072757937280.0,
      "grad_norm": 3.373020572821285,
      "language_loss": 0.66768903,
      "learning_rate": 7.01015055353253e-07,
      "loss": 0.68931174,
      "num_input_tokens_seen": 131035150,
      "step": 6098,
      "time_per_iteration": 3.427626132965088
    },
    {
      "auxiliary_loss_clip": 0.01104349,
      "auxiliary_loss_mlp": 0.01025622,
      "balance_loss_clip": 1.04264235,
      "balance_loss_mlp": 1.01781642,
      "epoch": 0.7333613900078157,
      "flos": 22743017735040.0,
      "grad_norm": 1.6594426146958103,
      "language_loss": 0.77940929,
      "learning_rate": 7.004228487653123e-07,
      "loss": 0.80070901,
      "num_input_tokens_seen": 131055955,
      "step": 6099,
      "time_per_iteration": 2.600335121154785
    },
    {
      "auxiliary_loss_clip": 0.01119688,
      "auxiliary_loss_mlp": 0.01022542,
      "balance_loss_clip": 1.03903115,
      "balance_loss_mlp": 1.0155654,
      "epoch": 0.7334816328984549,
      "flos": 22346133384960.0,
      "grad_norm": 2.3250237476893187,
      "language_loss": 0.78377461,
      "learning_rate": 6.998308393221906e-07,
      "loss": 0.80519694,
      "num_input_tokens_seen": 131074360,
      "step": 6100,
      "time_per_iteration": 2.4972546100616455
    },
    {
      "auxiliary_loss_clip": 0.01126922,
      "auxiliary_loss_mlp": 0.01025261,
      "balance_loss_clip": 1.04453993,
      "balance_loss_mlp": 1.01818848,
      "epoch": 0.733601875789094,
      "flos": 20736149984640.0,
      "grad_norm": 2.0025848768325925,
      "language_loss": 0.71113706,
      "learning_rate": 6.992390271136977e-07,
      "loss": 0.73265886,
      "num_input_tokens_seen": 131090070,
      "step": 6101,
      "time_per_iteration": 2.498704433441162
    },
    {
      "auxiliary_loss_clip": 0.01146787,
      "auxiliary_loss_mlp": 0.01020907,
      "balance_loss_clip": 1.04429829,
      "balance_loss_mlp": 1.01414514,
      "epoch": 0.733722118679733,
      "flos": 22564362464640.0,
      "grad_norm": 1.8165193175406134,
      "language_loss": 0.85598075,
      "learning_rate": 6.986474122296094e-07,
      "loss": 0.87765771,
      "num_input_tokens_seen": 131109185,
      "step": 6102,
      "time_per_iteration": 3.1823179721832275
    },
    {
      "auxiliary_loss_clip": 0.01172218,
      "auxiliary_loss_mlp": 0.01020854,
      "balance_loss_clip": 1.05054927,
      "balance_loss_mlp": 1.01384151,
      "epoch": 0.7338423615703722,
      "flos": 20084192179200.0,
      "grad_norm": 3.075498197473535,
      "language_loss": 0.72113538,
      "learning_rate": 6.980559947596751e-07,
      "loss": 0.74306607,
      "num_input_tokens_seen": 131127725,
      "step": 6103,
      "time_per_iteration": 2.397486686706543
    },
    {
      "auxiliary_loss_clip": 0.01108667,
      "auxiliary_loss_mlp": 0.01022405,
      "balance_loss_clip": 1.04046476,
      "balance_loss_mlp": 1.01496959,
      "epoch": 0.7339626044610112,
      "flos": 21687675217920.0,
      "grad_norm": 2.1976276906123364,
      "language_loss": 0.75900066,
      "learning_rate": 6.974647747936109e-07,
      "loss": 0.78031135,
      "num_input_tokens_seen": 131146110,
      "step": 6104,
      "time_per_iteration": 2.5461323261260986
    },
    {
      "auxiliary_loss_clip": 0.01169647,
      "auxiliary_loss_mlp": 0.00761873,
      "balance_loss_clip": 1.05007541,
      "balance_loss_mlp": 1.00040174,
      "epoch": 0.7340828473516503,
      "flos": 15268248282240.0,
      "grad_norm": 2.5093309413384755,
      "language_loss": 0.82421112,
      "learning_rate": 6.968737524211039e-07,
      "loss": 0.84352636,
      "num_input_tokens_seen": 131162920,
      "step": 6105,
      "time_per_iteration": 2.3735501766204834
    },
    {
      "auxiliary_loss_clip": 0.01153453,
      "auxiliary_loss_mlp": 0.01024431,
      "balance_loss_clip": 1.04783189,
      "balance_loss_mlp": 1.01715636,
      "epoch": 0.7342030902422895,
      "flos": 22930112701440.0,
      "grad_norm": 1.9187462857665418,
      "language_loss": 0.80317259,
      "learning_rate": 6.962829277318132e-07,
      "loss": 0.82495141,
      "num_input_tokens_seen": 131182515,
      "step": 6106,
      "time_per_iteration": 2.444582223892212
    },
    {
      "auxiliary_loss_clip": 0.01156594,
      "auxiliary_loss_mlp": 0.01025203,
      "balance_loss_clip": 1.05060506,
      "balance_loss_mlp": 1.01857531,
      "epoch": 0.7343233331329285,
      "flos": 25847890381440.0,
      "grad_norm": 1.7944830951726432,
      "language_loss": 0.83404624,
      "learning_rate": 6.956923008153652e-07,
      "loss": 0.85586423,
      "num_input_tokens_seen": 131202280,
      "step": 6107,
      "time_per_iteration": 2.4748213291168213
    },
    {
      "auxiliary_loss_clip": 0.01153147,
      "auxiliary_loss_mlp": 0.0102744,
      "balance_loss_clip": 1.04541135,
      "balance_loss_mlp": 1.02101445,
      "epoch": 0.7344435760235676,
      "flos": 18478985287680.0,
      "grad_norm": 2.1985630007422965,
      "language_loss": 0.84163302,
      "learning_rate": 6.951018717613593e-07,
      "loss": 0.86343884,
      "num_input_tokens_seen": 131221295,
      "step": 6108,
      "time_per_iteration": 2.452054738998413
    },
    {
      "auxiliary_loss_clip": 0.01152709,
      "auxiliary_loss_mlp": 0.01024608,
      "balance_loss_clip": 1.0479449,
      "balance_loss_mlp": 1.01766443,
      "epoch": 0.7345638189142067,
      "flos": 17640040256640.0,
      "grad_norm": 2.111463295746286,
      "language_loss": 0.7824921,
      "learning_rate": 6.945116406593614e-07,
      "loss": 0.80426526,
      "num_input_tokens_seen": 131240150,
      "step": 6109,
      "time_per_iteration": 2.4198927879333496
    },
    {
      "auxiliary_loss_clip": 0.01114618,
      "auxiliary_loss_mlp": 0.01025403,
      "balance_loss_clip": 1.04510975,
      "balance_loss_mlp": 1.01817298,
      "epoch": 0.7346840618048458,
      "flos": 20260225756800.0,
      "grad_norm": 2.0335258083724725,
      "language_loss": 0.74789572,
      "learning_rate": 6.939216075989089e-07,
      "loss": 0.76929587,
      "num_input_tokens_seen": 131258080,
      "step": 6110,
      "time_per_iteration": 2.5281388759613037
    },
    {
      "auxiliary_loss_clip": 0.01136416,
      "auxiliary_loss_mlp": 0.0102096,
      "balance_loss_clip": 1.04391694,
      "balance_loss_mlp": 1.01373553,
      "epoch": 0.7348043046954849,
      "flos": 29023183641600.0,
      "grad_norm": 2.0789368476138144,
      "language_loss": 0.66281021,
      "learning_rate": 6.933317726695109e-07,
      "loss": 0.68438393,
      "num_input_tokens_seen": 131279310,
      "step": 6111,
      "time_per_iteration": 2.523442506790161
    },
    {
      "auxiliary_loss_clip": 0.0112566,
      "auxiliary_loss_mlp": 0.01025986,
      "balance_loss_clip": 1.04791582,
      "balance_loss_mlp": 1.01862824,
      "epoch": 0.734924547586124,
      "flos": 17931203902080.0,
      "grad_norm": 2.316041614826212,
      "language_loss": 0.79884422,
      "learning_rate": 6.92742135960644e-07,
      "loss": 0.82036066,
      "num_input_tokens_seen": 131297010,
      "step": 6112,
      "time_per_iteration": 2.471081495285034
    },
    {
      "auxiliary_loss_clip": 0.01061137,
      "auxiliary_loss_mlp": 0.01001304,
      "balance_loss_clip": 1.01565647,
      "balance_loss_mlp": 1.00023687,
      "epoch": 0.7350447904767631,
      "flos": 63588319850880.0,
      "grad_norm": 0.8204316855328927,
      "language_loss": 0.55722958,
      "learning_rate": 6.921526975617556e-07,
      "loss": 0.57785398,
      "num_input_tokens_seen": 131356470,
      "step": 6113,
      "time_per_iteration": 3.0800039768218994
    },
    {
      "auxiliary_loss_clip": 0.01141019,
      "auxiliary_loss_mlp": 0.01024701,
      "balance_loss_clip": 1.04361629,
      "balance_loss_mlp": 1.01739979,
      "epoch": 0.7351650333674021,
      "flos": 21580015178880.0,
      "grad_norm": 2.0189428341341573,
      "language_loss": 0.75163162,
      "learning_rate": 6.915634575622631e-07,
      "loss": 0.77328885,
      "num_input_tokens_seen": 131374985,
      "step": 6114,
      "time_per_iteration": 2.4784367084503174
    },
    {
      "auxiliary_loss_clip": 0.01166651,
      "auxiliary_loss_mlp": 0.01022242,
      "balance_loss_clip": 1.04894483,
      "balance_loss_mlp": 1.01523268,
      "epoch": 0.7352852762580413,
      "flos": 18186349184640.0,
      "grad_norm": 1.8347008491799346,
      "language_loss": 0.7092185,
      "learning_rate": 6.909744160515532e-07,
      "loss": 0.73110735,
      "num_input_tokens_seen": 131393125,
      "step": 6115,
      "time_per_iteration": 2.388014316558838
    },
    {
      "auxiliary_loss_clip": 0.01137416,
      "auxiliary_loss_mlp": 0.01024828,
      "balance_loss_clip": 1.0452559,
      "balance_loss_mlp": 1.01779723,
      "epoch": 0.7354055191486804,
      "flos": 38910073063680.0,
      "grad_norm": 1.995171613961482,
      "language_loss": 0.69654024,
      "learning_rate": 6.903855731189849e-07,
      "loss": 0.71816266,
      "num_input_tokens_seen": 131415760,
      "step": 6116,
      "time_per_iteration": 2.6465911865234375
    },
    {
      "auxiliary_loss_clip": 0.01147112,
      "auxiliary_loss_mlp": 0.01030061,
      "balance_loss_clip": 1.04764831,
      "balance_loss_mlp": 1.02231574,
      "epoch": 0.7355257620393194,
      "flos": 16289978647680.0,
      "grad_norm": 2.388241065246289,
      "language_loss": 0.82299089,
      "learning_rate": 6.897969288538825e-07,
      "loss": 0.84476262,
      "num_input_tokens_seen": 131433705,
      "step": 6117,
      "time_per_iteration": 2.4507827758789062
    },
    {
      "auxiliary_loss_clip": 0.01138816,
      "auxiliary_loss_mlp": 0.01024592,
      "balance_loss_clip": 1.0463208,
      "balance_loss_mlp": 1.01778245,
      "epoch": 0.7356460049299585,
      "flos": 18114240631680.0,
      "grad_norm": 2.3434490001092554,
      "language_loss": 0.81514186,
      "learning_rate": 6.892084833455452e-07,
      "loss": 0.8367759,
      "num_input_tokens_seen": 131453275,
      "step": 6118,
      "time_per_iteration": 2.453321695327759
    },
    {
      "auxiliary_loss_clip": 0.0115268,
      "auxiliary_loss_mlp": 0.01022325,
      "balance_loss_clip": 1.04944384,
      "balance_loss_mlp": 1.01560163,
      "epoch": 0.7357662478205976,
      "flos": 21325193118720.0,
      "grad_norm": 1.4434957043541996,
      "language_loss": 0.83912963,
      "learning_rate": 6.886202366832384e-07,
      "loss": 0.86087966,
      "num_input_tokens_seen": 131474960,
      "step": 6119,
      "time_per_iteration": 2.463392972946167
    },
    {
      "auxiliary_loss_clip": 0.01113138,
      "auxiliary_loss_mlp": 0.01023785,
      "balance_loss_clip": 1.04653382,
      "balance_loss_mlp": 1.01610804,
      "epoch": 0.7358864907112367,
      "flos": 14246841139200.0,
      "grad_norm": 9.36868885106388,
      "language_loss": 0.73476374,
      "learning_rate": 6.880321889561987e-07,
      "loss": 0.75613296,
      "num_input_tokens_seen": 131492935,
      "step": 6120,
      "time_per_iteration": 2.504666805267334
    },
    {
      "auxiliary_loss_clip": 0.01119984,
      "auxiliary_loss_mlp": 0.01027779,
      "balance_loss_clip": 1.04332554,
      "balance_loss_mlp": 1.01975584,
      "epoch": 0.7360067336018757,
      "flos": 22309684058880.0,
      "grad_norm": 2.763689757625933,
      "language_loss": 0.65157712,
      "learning_rate": 6.874443402536338e-07,
      "loss": 0.67305475,
      "num_input_tokens_seen": 131512025,
      "step": 6121,
      "time_per_iteration": 2.507908821105957
    },
    {
      "auxiliary_loss_clip": 0.01142145,
      "auxiliary_loss_mlp": 0.01020396,
      "balance_loss_clip": 1.04724443,
      "balance_loss_mlp": 1.01326132,
      "epoch": 0.7361269764925149,
      "flos": 25554607833600.0,
      "grad_norm": 2.8295515221300467,
      "language_loss": 0.80553359,
      "learning_rate": 6.868566906647177e-07,
      "loss": 0.82715905,
      "num_input_tokens_seen": 131532975,
      "step": 6122,
      "time_per_iteration": 2.5304811000823975
    },
    {
      "auxiliary_loss_clip": 0.01153464,
      "auxiliary_loss_mlp": 0.01028199,
      "balance_loss_clip": 1.04568267,
      "balance_loss_mlp": 1.02060509,
      "epoch": 0.736247219383154,
      "flos": 20376505059840.0,
      "grad_norm": 2.8301423462248514,
      "language_loss": 0.83408904,
      "learning_rate": 6.862692402785984e-07,
      "loss": 0.85590565,
      "num_input_tokens_seen": 131553225,
      "step": 6123,
      "time_per_iteration": 3.161473274230957
    },
    {
      "auxiliary_loss_clip": 0.01033688,
      "auxiliary_loss_mlp": 0.01004321,
      "balance_loss_clip": 1.01557016,
      "balance_loss_mlp": 1.00307488,
      "epoch": 0.736367462273793,
      "flos": 70339525735680.0,
      "grad_norm": 0.7948722943119283,
      "language_loss": 0.4958145,
      "learning_rate": 6.856819891843899e-07,
      "loss": 0.51619458,
      "num_input_tokens_seen": 131617930,
      "step": 6124,
      "time_per_iteration": 4.001164436340332
    },
    {
      "auxiliary_loss_clip": 0.01096068,
      "auxiliary_loss_mlp": 0.01027882,
      "balance_loss_clip": 1.04080462,
      "balance_loss_mlp": 1.02078855,
      "epoch": 0.7364877051644322,
      "flos": 22412711243520.0,
      "grad_norm": 1.977683330044527,
      "language_loss": 0.72185355,
      "learning_rate": 6.8509493747118e-07,
      "loss": 0.74309307,
      "num_input_tokens_seen": 131636740,
      "step": 6125,
      "time_per_iteration": 3.3232955932617188
    },
    {
      "auxiliary_loss_clip": 0.01167659,
      "auxiliary_loss_mlp": 0.01024772,
      "balance_loss_clip": 1.04850829,
      "balance_loss_mlp": 1.01767015,
      "epoch": 0.7366079480550712,
      "flos": 12130266274560.0,
      "grad_norm": 4.056533982848506,
      "language_loss": 0.87854242,
      "learning_rate": 6.845080852280221e-07,
      "loss": 0.90046674,
      "num_input_tokens_seen": 131653810,
      "step": 6126,
      "time_per_iteration": 2.367185115814209
    },
    {
      "auxiliary_loss_clip": 0.0112573,
      "auxiliary_loss_mlp": 0.01021132,
      "balance_loss_clip": 1.0436331,
      "balance_loss_mlp": 1.01472116,
      "epoch": 0.7367281909457103,
      "flos": 15049336844160.0,
      "grad_norm": 1.608921876373448,
      "language_loss": 0.742347,
      "learning_rate": 6.839214325439409e-07,
      "loss": 0.76381558,
      "num_input_tokens_seen": 131671505,
      "step": 6127,
      "time_per_iteration": 2.48690128326416
    },
    {
      "auxiliary_loss_clip": 0.01133006,
      "auxiliary_loss_mlp": 0.01024216,
      "balance_loss_clip": 1.04617584,
      "balance_loss_mlp": 1.0170753,
      "epoch": 0.7368484338363495,
      "flos": 23510752053120.0,
      "grad_norm": 2.39098191918476,
      "language_loss": 0.7174834,
      "learning_rate": 6.833349795079327e-07,
      "loss": 0.73905563,
      "num_input_tokens_seen": 131690615,
      "step": 6128,
      "time_per_iteration": 3.2437474727630615
    },
    {
      "auxiliary_loss_clip": 0.01125568,
      "auxiliary_loss_mlp": 0.01025287,
      "balance_loss_clip": 1.04596829,
      "balance_loss_mlp": 1.01815772,
      "epoch": 0.7369686767269885,
      "flos": 27417833095680.0,
      "grad_norm": 1.66657925085952,
      "language_loss": 0.68344021,
      "learning_rate": 6.827487262089613e-07,
      "loss": 0.70494872,
      "num_input_tokens_seen": 131711120,
      "step": 6129,
      "time_per_iteration": 2.5470545291900635
    },
    {
      "auxiliary_loss_clip": 0.01040807,
      "auxiliary_loss_mlp": 0.01001375,
      "balance_loss_clip": 1.01108766,
      "balance_loss_mlp": 1.00024879,
      "epoch": 0.7370889196176276,
      "flos": 70293343824000.0,
      "grad_norm": 0.9011213052345667,
      "language_loss": 0.56821793,
      "learning_rate": 6.821626727359606e-07,
      "loss": 0.58863974,
      "num_input_tokens_seen": 131776680,
      "step": 6130,
      "time_per_iteration": 3.124809741973877
    },
    {
      "auxiliary_loss_clip": 0.01143709,
      "auxiliary_loss_mlp": 0.01023798,
      "balance_loss_clip": 1.05106902,
      "balance_loss_mlp": 1.01607335,
      "epoch": 0.7372091625082667,
      "flos": 18040839189120.0,
      "grad_norm": 2.9872590313058676,
      "language_loss": 0.76932818,
      "learning_rate": 6.815768191778348e-07,
      "loss": 0.79100323,
      "num_input_tokens_seen": 131794760,
      "step": 6131,
      "time_per_iteration": 2.4574620723724365
    },
    {
      "auxiliary_loss_clip": 0.01150286,
      "auxiliary_loss_mlp": 0.01026746,
      "balance_loss_clip": 1.04574406,
      "balance_loss_mlp": 1.01939034,
      "epoch": 0.7373294053989058,
      "flos": 33726331854720.0,
      "grad_norm": 2.072196462936505,
      "language_loss": 0.7303524,
      "learning_rate": 6.809911656234569e-07,
      "loss": 0.7521227,
      "num_input_tokens_seen": 131816735,
      "step": 6132,
      "time_per_iteration": 2.5581698417663574
    },
    {
      "auxiliary_loss_clip": 0.01126694,
      "auxiliary_loss_mlp": 0.01021401,
      "balance_loss_clip": 1.04247594,
      "balance_loss_mlp": 1.01460648,
      "epoch": 0.7374496482895448,
      "flos": 21506326427520.0,
      "grad_norm": 2.0023466777862127,
      "language_loss": 0.78540516,
      "learning_rate": 6.804057121616707e-07,
      "loss": 0.80688608,
      "num_input_tokens_seen": 131834940,
      "step": 6133,
      "time_per_iteration": 2.5201940536499023
    },
    {
      "auxiliary_loss_clip": 0.01154842,
      "auxiliary_loss_mlp": 0.01024463,
      "balance_loss_clip": 1.04774356,
      "balance_loss_mlp": 1.0167737,
      "epoch": 0.737569891180184,
      "flos": 24936908624640.0,
      "grad_norm": 2.035292912528272,
      "language_loss": 0.72040772,
      "learning_rate": 6.798204588812888e-07,
      "loss": 0.74220073,
      "num_input_tokens_seen": 131854355,
      "step": 6134,
      "time_per_iteration": 2.508913040161133
    },
    {
      "auxiliary_loss_clip": 0.01084932,
      "auxiliary_loss_mlp": 0.00761731,
      "balance_loss_clip": 1.03742397,
      "balance_loss_mlp": 1.00034809,
      "epoch": 0.7376901340708231,
      "flos": 20664544222080.0,
      "grad_norm": 1.793526734011821,
      "language_loss": 0.75624692,
      "learning_rate": 6.792354058710937e-07,
      "loss": 0.77471364,
      "num_input_tokens_seen": 131871825,
      "step": 6135,
      "time_per_iteration": 2.5681512355804443
    },
    {
      "auxiliary_loss_clip": 0.01161144,
      "auxiliary_loss_mlp": 0.0102108,
      "balance_loss_clip": 1.04705095,
      "balance_loss_mlp": 1.01413596,
      "epoch": 0.7378103769614621,
      "flos": 23805794367360.0,
      "grad_norm": 1.7824262238039135,
      "language_loss": 0.65195465,
      "learning_rate": 6.786505532198374e-07,
      "loss": 0.67377687,
      "num_input_tokens_seen": 131890770,
      "step": 6136,
      "time_per_iteration": 2.4442238807678223
    },
    {
      "auxiliary_loss_clip": 0.01167465,
      "auxiliary_loss_mlp": 0.01026631,
      "balance_loss_clip": 1.04805684,
      "balance_loss_mlp": 1.01902533,
      "epoch": 0.7379306198521013,
      "flos": 22237216369920.0,
      "grad_norm": 1.6295217909728643,
      "language_loss": 0.8544395,
      "learning_rate": 6.780659010162411e-07,
      "loss": 0.87638044,
      "num_input_tokens_seen": 131909720,
      "step": 6137,
      "time_per_iteration": 2.4078242778778076
    },
    {
      "auxiliary_loss_clip": 0.01130993,
      "auxiliary_loss_mlp": 0.01023,
      "balance_loss_clip": 1.04657292,
      "balance_loss_mlp": 1.01618147,
      "epoch": 0.7380508627427403,
      "flos": 14903108576640.0,
      "grad_norm": 1.7315278849538593,
      "language_loss": 0.83306575,
      "learning_rate": 6.774814493489975e-07,
      "loss": 0.85460567,
      "num_input_tokens_seen": 131927395,
      "step": 6138,
      "time_per_iteration": 2.4646408557891846
    },
    {
      "auxiliary_loss_clip": 0.01150791,
      "auxiliary_loss_mlp": 0.01023401,
      "balance_loss_clip": 1.04736924,
      "balance_loss_mlp": 1.01627243,
      "epoch": 0.7381711056333794,
      "flos": 21685843624320.0,
      "grad_norm": 1.8481432317643938,
      "language_loss": 0.66169411,
      "learning_rate": 6.768971983067655e-07,
      "loss": 0.68343604,
      "num_input_tokens_seen": 131947725,
      "step": 6139,
      "time_per_iteration": 2.444254159927368
    },
    {
      "auxiliary_loss_clip": 0.01068921,
      "auxiliary_loss_mlp": 0.01000978,
      "balance_loss_clip": 1.01360869,
      "balance_loss_mlp": 0.99999475,
      "epoch": 0.7382913485240186,
      "flos": 52404263596800.0,
      "grad_norm": 1.008046804286874,
      "language_loss": 0.67838311,
      "learning_rate": 6.763131479781772e-07,
      "loss": 0.69908208,
      "num_input_tokens_seen": 131997485,
      "step": 6140,
      "time_per_iteration": 2.824784517288208
    },
    {
      "auxiliary_loss_clip": 0.01132157,
      "auxiliary_loss_mlp": 0.01023985,
      "balance_loss_clip": 1.04671896,
      "balance_loss_mlp": 1.01700807,
      "epoch": 0.7384115914146576,
      "flos": 21798818876160.0,
      "grad_norm": 2.146481648944633,
      "language_loss": 0.76783538,
      "learning_rate": 6.757292984518316e-07,
      "loss": 0.78939682,
      "num_input_tokens_seen": 132016885,
      "step": 6141,
      "time_per_iteration": 2.456026792526245
    },
    {
      "auxiliary_loss_clip": 0.01059514,
      "auxiliary_loss_mlp": 0.01001796,
      "balance_loss_clip": 1.01408243,
      "balance_loss_mlp": 1.00078857,
      "epoch": 0.7385318343052967,
      "flos": 61494331662720.0,
      "grad_norm": 0.7408605739698927,
      "language_loss": 0.56436586,
      "learning_rate": 6.751456498162981e-07,
      "loss": 0.58497894,
      "num_input_tokens_seen": 132075920,
      "step": 6142,
      "time_per_iteration": 2.9201762676239014
    },
    {
      "auxiliary_loss_clip": 0.01150295,
      "auxiliary_loss_mlp": 0.01021738,
      "balance_loss_clip": 1.04317379,
      "balance_loss_mlp": 1.01553011,
      "epoch": 0.7386520771959358,
      "flos": 17013757697280.0,
      "grad_norm": 2.040981969223243,
      "language_loss": 0.85481173,
      "learning_rate": 6.745622021601174e-07,
      "loss": 0.87653208,
      "num_input_tokens_seen": 132092945,
      "step": 6143,
      "time_per_iteration": 2.398787498474121
    },
    {
      "auxiliary_loss_clip": 0.01127585,
      "auxiliary_loss_mlp": 0.01022816,
      "balance_loss_clip": 1.0437746,
      "balance_loss_mlp": 1.01579201,
      "epoch": 0.7387723200865749,
      "flos": 18770759464320.0,
      "grad_norm": 1.9343672579084095,
      "language_loss": 0.69383079,
      "learning_rate": 6.739789555717954e-07,
      "loss": 0.71533477,
      "num_input_tokens_seen": 132109920,
      "step": 6144,
      "time_per_iteration": 2.471367597579956
    },
    {
      "auxiliary_loss_clip": 0.01166018,
      "auxiliary_loss_mlp": 0.01024849,
      "balance_loss_clip": 1.04785812,
      "balance_loss_mlp": 1.01795232,
      "epoch": 0.738892562977214,
      "flos": 22525542840960.0,
      "grad_norm": 2.299903502533772,
      "language_loss": 0.77365208,
      "learning_rate": 6.733959101398124e-07,
      "loss": 0.79556072,
      "num_input_tokens_seen": 132128050,
      "step": 6145,
      "time_per_iteration": 2.4074547290802
    },
    {
      "auxiliary_loss_clip": 0.01137132,
      "auxiliary_loss_mlp": 0.01025278,
      "balance_loss_clip": 1.04421115,
      "balance_loss_mlp": 1.01754189,
      "epoch": 0.7390128058678531,
      "flos": 21501478091520.0,
      "grad_norm": 1.6021813279585075,
      "language_loss": 0.81531954,
      "learning_rate": 6.728130659526143e-07,
      "loss": 0.83694357,
      "num_input_tokens_seen": 132145860,
      "step": 6146,
      "time_per_iteration": 2.473816156387329
    },
    {
      "auxiliary_loss_clip": 0.01142773,
      "auxiliary_loss_mlp": 0.01029601,
      "balance_loss_clip": 1.04756582,
      "balance_loss_mlp": 1.02216864,
      "epoch": 0.7391330487584922,
      "flos": 25776176878080.0,
      "grad_norm": 2.4484079217432675,
      "language_loss": 0.71100426,
      "learning_rate": 6.7223042309862e-07,
      "loss": 0.73272794,
      "num_input_tokens_seen": 132166060,
      "step": 6147,
      "time_per_iteration": 2.5129122734069824
    },
    {
      "auxiliary_loss_clip": 0.01149319,
      "auxiliary_loss_mlp": 0.0102609,
      "balance_loss_clip": 1.04482603,
      "balance_loss_mlp": 1.01949739,
      "epoch": 0.7392532916491312,
      "flos": 28366736636160.0,
      "grad_norm": 4.769843803939903,
      "language_loss": 0.73553824,
      "learning_rate": 6.716479816662144e-07,
      "loss": 0.75729239,
      "num_input_tokens_seen": 132187790,
      "step": 6148,
      "time_per_iteration": 2.5065290927886963
    },
    {
      "auxiliary_loss_clip": 0.01142973,
      "auxiliary_loss_mlp": 0.01022549,
      "balance_loss_clip": 1.04529178,
      "balance_loss_mlp": 1.01551604,
      "epoch": 0.7393735345397703,
      "flos": 23585877348480.0,
      "grad_norm": 1.9293338262648343,
      "language_loss": 0.73069763,
      "learning_rate": 6.710657417437531e-07,
      "loss": 0.75235289,
      "num_input_tokens_seen": 132207495,
      "step": 6149,
      "time_per_iteration": 2.544508934020996
    },
    {
      "auxiliary_loss_clip": 0.01137384,
      "auxiliary_loss_mlp": 0.01024611,
      "balance_loss_clip": 1.04498255,
      "balance_loss_mlp": 1.0176549,
      "epoch": 0.7394937774304094,
      "flos": 19974772373760.0,
      "grad_norm": 2.4202571793140564,
      "language_loss": 0.79693204,
      "learning_rate": 6.704837034195628e-07,
      "loss": 0.81855202,
      "num_input_tokens_seen": 132225960,
      "step": 6150,
      "time_per_iteration": 4.103356122970581
    },
    {
      "auxiliary_loss_clip": 0.01144622,
      "auxiliary_loss_mlp": 0.0103129,
      "balance_loss_clip": 1.04429317,
      "balance_loss_mlp": 1.02390468,
      "epoch": 0.7396140203210485,
      "flos": 23478037741440.0,
      "grad_norm": 1.6533003076769768,
      "language_loss": 0.84895766,
      "learning_rate": 6.699018667819376e-07,
      "loss": 0.87071669,
      "num_input_tokens_seen": 132245360,
      "step": 6151,
      "time_per_iteration": 2.4704983234405518
    },
    {
      "auxiliary_loss_clip": 0.01150598,
      "auxiliary_loss_mlp": 0.0102859,
      "balance_loss_clip": 1.04499865,
      "balance_loss_mlp": 1.02094853,
      "epoch": 0.7397342632116876,
      "flos": 25555433846400.0,
      "grad_norm": 1.573566404652876,
      "language_loss": 0.72694886,
      "learning_rate": 6.693202319191415e-07,
      "loss": 0.74874073,
      "num_input_tokens_seen": 132267095,
      "step": 6152,
      "time_per_iteration": 3.3766157627105713
    },
    {
      "auxiliary_loss_clip": 0.0116696,
      "auxiliary_loss_mlp": 0.01026223,
      "balance_loss_clip": 1.05151534,
      "balance_loss_mlp": 1.01887691,
      "epoch": 0.7398545061023267,
      "flos": 24755021130240.0,
      "grad_norm": 1.8394961097270544,
      "language_loss": 0.7519151,
      "learning_rate": 6.687387989194084e-07,
      "loss": 0.77384698,
      "num_input_tokens_seen": 132286610,
      "step": 6153,
      "time_per_iteration": 2.5108325481414795
    },
    {
      "auxiliary_loss_clip": 0.01136614,
      "auxiliary_loss_mlp": 0.01025994,
      "balance_loss_clip": 1.04856825,
      "balance_loss_mlp": 1.01886511,
      "epoch": 0.7399747489929658,
      "flos": 16508602776960.0,
      "grad_norm": 1.7675194688513232,
      "language_loss": 0.79115498,
      "learning_rate": 6.681575678709404e-07,
      "loss": 0.8127811,
      "num_input_tokens_seen": 132305300,
      "step": 6154,
      "time_per_iteration": 2.4753262996673584
    },
    {
      "auxiliary_loss_clip": 0.01152331,
      "auxiliary_loss_mlp": 0.01022402,
      "balance_loss_clip": 1.04705024,
      "balance_loss_mlp": 1.01550889,
      "epoch": 0.7400949918836048,
      "flos": 24097065753600.0,
      "grad_norm": 1.9513790591154117,
      "language_loss": 0.70933914,
      "learning_rate": 6.67576538861911e-07,
      "loss": 0.73108649,
      "num_input_tokens_seen": 132323875,
      "step": 6155,
      "time_per_iteration": 3.265145778656006
    },
    {
      "auxiliary_loss_clip": 0.01134586,
      "auxiliary_loss_mlp": 0.01022995,
      "balance_loss_clip": 1.0445075,
      "balance_loss_mlp": 1.01632559,
      "epoch": 0.740215234774244,
      "flos": 21802517976960.0,
      "grad_norm": 1.914649125924512,
      "language_loss": 0.82184827,
      "learning_rate": 6.669957119804612e-07,
      "loss": 0.84342408,
      "num_input_tokens_seen": 132345510,
      "step": 6156,
      "time_per_iteration": 2.5210795402526855
    },
    {
      "auxiliary_loss_clip": 0.01147576,
      "auxiliary_loss_mlp": 0.01025451,
      "balance_loss_clip": 1.04784179,
      "balance_loss_mlp": 1.01861393,
      "epoch": 0.7403354776648831,
      "flos": 18733196816640.0,
      "grad_norm": 3.274396289426646,
      "language_loss": 0.71954364,
      "learning_rate": 6.66415087314702e-07,
      "loss": 0.74127388,
      "num_input_tokens_seen": 132360465,
      "step": 6157,
      "time_per_iteration": 2.4516172409057617
    },
    {
      "auxiliary_loss_clip": 0.01139076,
      "auxiliary_loss_mlp": 0.01019088,
      "balance_loss_clip": 1.04441166,
      "balance_loss_mlp": 1.01216197,
      "epoch": 0.7404557205555221,
      "flos": 16909581277440.0,
      "grad_norm": 1.9876777460500796,
      "language_loss": 0.73317605,
      "learning_rate": 6.65834664952714e-07,
      "loss": 0.75475764,
      "num_input_tokens_seen": 132377915,
      "step": 6158,
      "time_per_iteration": 2.4498064517974854
    },
    {
      "auxiliary_loss_clip": 0.01124776,
      "auxiliary_loss_mlp": 0.01022515,
      "balance_loss_clip": 1.0433557,
      "balance_loss_mlp": 1.01587176,
      "epoch": 0.7405759634461613,
      "flos": 21214408596480.0,
      "grad_norm": 5.215755168968883,
      "language_loss": 0.76198125,
      "learning_rate": 6.652544449825457e-07,
      "loss": 0.78345418,
      "num_input_tokens_seen": 132398170,
      "step": 6159,
      "time_per_iteration": 2.5090584754943848
    },
    {
      "auxiliary_loss_clip": 0.01146281,
      "auxiliary_loss_mlp": 0.01029306,
      "balance_loss_clip": 1.04439187,
      "balance_loss_mlp": 1.02203774,
      "epoch": 0.7406962063368003,
      "flos": 20480106862080.0,
      "grad_norm": 2.281255998077098,
      "language_loss": 0.76381552,
      "learning_rate": 6.646744274922182e-07,
      "loss": 0.7855714,
      "num_input_tokens_seen": 132416615,
      "step": 6160,
      "time_per_iteration": 2.4607415199279785
    },
    {
      "auxiliary_loss_clip": 0.01141031,
      "auxiliary_loss_mlp": 0.01019453,
      "balance_loss_clip": 1.0450536,
      "balance_loss_mlp": 1.0123508,
      "epoch": 0.7408164492274394,
      "flos": 19791915212160.0,
      "grad_norm": 2.6430930941410273,
      "language_loss": 0.75554162,
      "learning_rate": 6.640946125697171e-07,
      "loss": 0.77714646,
      "num_input_tokens_seen": 132434145,
      "step": 6161,
      "time_per_iteration": 2.4523236751556396
    },
    {
      "auxiliary_loss_clip": 0.01153634,
      "auxiliary_loss_mlp": 0.01021035,
      "balance_loss_clip": 1.04576206,
      "balance_loss_mlp": 1.01379323,
      "epoch": 0.7409366921180786,
      "flos": 29204855654400.0,
      "grad_norm": 1.7400778372577494,
      "language_loss": 0.75125176,
      "learning_rate": 6.635150003030017e-07,
      "loss": 0.77299845,
      "num_input_tokens_seen": 132452670,
      "step": 6162,
      "time_per_iteration": 2.5072247982025146
    },
    {
      "auxiliary_loss_clip": 0.01109275,
      "auxiliary_loss_mlp": 0.01022135,
      "balance_loss_clip": 1.03884137,
      "balance_loss_mlp": 1.0152154,
      "epoch": 0.7410569350087176,
      "flos": 22930004960640.0,
      "grad_norm": 2.5277331856093337,
      "language_loss": 0.85946649,
      "learning_rate": 6.629355907799981e-07,
      "loss": 0.88078064,
      "num_input_tokens_seen": 132472475,
      "step": 6163,
      "time_per_iteration": 2.524101495742798
    },
    {
      "auxiliary_loss_clip": 0.01152612,
      "auxiliary_loss_mlp": 0.01022343,
      "balance_loss_clip": 1.04496896,
      "balance_loss_mlp": 1.01539588,
      "epoch": 0.7411771778993567,
      "flos": 30440397726720.0,
      "grad_norm": 1.745145059194402,
      "language_loss": 0.69244128,
      "learning_rate": 6.623563840886015e-07,
      "loss": 0.71419084,
      "num_input_tokens_seen": 132493400,
      "step": 6164,
      "time_per_iteration": 2.5163278579711914
    },
    {
      "auxiliary_loss_clip": 0.01148683,
      "auxiliary_loss_mlp": 0.0102045,
      "balance_loss_clip": 1.04535162,
      "balance_loss_mlp": 1.01350021,
      "epoch": 0.7412974207899958,
      "flos": 20522050968960.0,
      "grad_norm": 1.5921267757564497,
      "language_loss": 0.69588256,
      "learning_rate": 6.617773803166795e-07,
      "loss": 0.71757388,
      "num_input_tokens_seen": 132511725,
      "step": 6165,
      "time_per_iteration": 2.429363489151001
    },
    {
      "auxiliary_loss_clip": 0.01144432,
      "auxiliary_loss_mlp": 0.00762124,
      "balance_loss_clip": 1.04588127,
      "balance_loss_mlp": 1.00039768,
      "epoch": 0.7414176636806349,
      "flos": 22090700793600.0,
      "grad_norm": 2.160269003135224,
      "language_loss": 0.82313693,
      "learning_rate": 6.611985795520634e-07,
      "loss": 0.84220243,
      "num_input_tokens_seen": 132530270,
      "step": 6166,
      "time_per_iteration": 2.475053071975708
    },
    {
      "auxiliary_loss_clip": 0.01134589,
      "auxiliary_loss_mlp": 0.01025548,
      "balance_loss_clip": 1.04705679,
      "balance_loss_mlp": 1.01823461,
      "epoch": 0.7415379065712739,
      "flos": 25155245445120.0,
      "grad_norm": 1.9970311768991247,
      "language_loss": 0.77234519,
      "learning_rate": 6.606199818825588e-07,
      "loss": 0.79394656,
      "num_input_tokens_seen": 132550725,
      "step": 6167,
      "time_per_iteration": 2.542524814605713
    },
    {
      "auxiliary_loss_clip": 0.01143904,
      "auxiliary_loss_mlp": 0.01023678,
      "balance_loss_clip": 1.04378426,
      "balance_loss_mlp": 1.01652575,
      "epoch": 0.7416581494619131,
      "flos": 16871731320960.0,
      "grad_norm": 3.0072285731994546,
      "language_loss": 0.81650358,
      "learning_rate": 6.600415873959377e-07,
      "loss": 0.83817947,
      "num_input_tokens_seen": 132568600,
      "step": 6168,
      "time_per_iteration": 2.438075065612793
    },
    {
      "auxiliary_loss_clip": 0.01095927,
      "auxiliary_loss_mlp": 0.00761202,
      "balance_loss_clip": 1.04002655,
      "balance_loss_mlp": 1.00032854,
      "epoch": 0.7417783923525522,
      "flos": 28438881102720.0,
      "grad_norm": 1.9182251782080653,
      "language_loss": 0.64667565,
      "learning_rate": 6.594633961799437e-07,
      "loss": 0.6652469,
      "num_input_tokens_seen": 132587640,
      "step": 6169,
      "time_per_iteration": 2.639529228210449
    },
    {
      "auxiliary_loss_clip": 0.01133024,
      "auxiliary_loss_mlp": 0.01022993,
      "balance_loss_clip": 1.04518294,
      "balance_loss_mlp": 1.01583755,
      "epoch": 0.7418986352431912,
      "flos": 20084299920000.0,
      "grad_norm": 2.1666729190003124,
      "language_loss": 0.81358981,
      "learning_rate": 6.588854083222857e-07,
      "loss": 0.83514994,
      "num_input_tokens_seen": 132607075,
      "step": 6170,
      "time_per_iteration": 2.5021615028381348
    },
    {
      "auxiliary_loss_clip": 0.01144455,
      "auxiliary_loss_mlp": 0.01026511,
      "balance_loss_clip": 1.0483315,
      "balance_loss_mlp": 1.01875067,
      "epoch": 0.7420188781338304,
      "flos": 18259571059200.0,
      "grad_norm": 5.59677325555764,
      "language_loss": 0.81034243,
      "learning_rate": 6.583076239106444e-07,
      "loss": 0.83205211,
      "num_input_tokens_seen": 132625580,
      "step": 6171,
      "time_per_iteration": 2.461170196533203
    },
    {
      "auxiliary_loss_clip": 0.01144108,
      "auxiliary_loss_mlp": 0.0102294,
      "balance_loss_clip": 1.04674983,
      "balance_loss_mlp": 1.01540005,
      "epoch": 0.7421391210244694,
      "flos": 13771994319360.0,
      "grad_norm": 2.001980496848609,
      "language_loss": 0.75233805,
      "learning_rate": 6.577300430326707e-07,
      "loss": 0.77400851,
      "num_input_tokens_seen": 132640525,
      "step": 6172,
      "time_per_iteration": 2.4620823860168457
    },
    {
      "auxiliary_loss_clip": 0.01124835,
      "auxiliary_loss_mlp": 0.01022405,
      "balance_loss_clip": 1.04649031,
      "balance_loss_mlp": 1.01547635,
      "epoch": 0.7422593639151085,
      "flos": 15961683317760.0,
      "grad_norm": 2.3653042260689228,
      "language_loss": 0.72111821,
      "learning_rate": 6.571526657759821e-07,
      "loss": 0.74259055,
      "num_input_tokens_seen": 132656265,
      "step": 6173,
      "time_per_iteration": 2.4795925617218018
    },
    {
      "auxiliary_loss_clip": 0.01146936,
      "auxiliary_loss_mlp": 0.01020344,
      "balance_loss_clip": 1.04446375,
      "balance_loss_mlp": 1.01336098,
      "epoch": 0.7423796068057477,
      "flos": 30114400867200.0,
      "grad_norm": 1.734758258463048,
      "language_loss": 0.70693886,
      "learning_rate": 6.565754922281663e-07,
      "loss": 0.72861171,
      "num_input_tokens_seen": 132678510,
      "step": 6174,
      "time_per_iteration": 2.5282599925994873
    },
    {
      "auxiliary_loss_clip": 0.01137581,
      "auxiliary_loss_mlp": 0.01027362,
      "balance_loss_clip": 1.04447603,
      "balance_loss_mlp": 1.02048039,
      "epoch": 0.7424998496963867,
      "flos": 20521907314560.0,
      "grad_norm": 1.8061434100116054,
      "language_loss": 0.78390324,
      "learning_rate": 6.559985224767801e-07,
      "loss": 0.8055526,
      "num_input_tokens_seen": 132696385,
      "step": 6175,
      "time_per_iteration": 2.5167171955108643
    },
    {
      "auxiliary_loss_clip": 0.01130282,
      "auxiliary_loss_mlp": 0.01026147,
      "balance_loss_clip": 1.04436326,
      "balance_loss_mlp": 1.01907206,
      "epoch": 0.7426200925870258,
      "flos": 21871573873920.0,
      "grad_norm": 3.420312848934449,
      "language_loss": 0.75403482,
      "learning_rate": 6.55421756609349e-07,
      "loss": 0.77559912,
      "num_input_tokens_seen": 132714640,
      "step": 6176,
      "time_per_iteration": 2.4969983100891113
    },
    {
      "auxiliary_loss_clip": 0.01152111,
      "auxiliary_loss_mlp": 0.01029325,
      "balance_loss_clip": 1.04969072,
      "balance_loss_mlp": 1.02165723,
      "epoch": 0.7427403354776649,
      "flos": 26432049265920.0,
      "grad_norm": 1.7760128107619337,
      "language_loss": 0.79036349,
      "learning_rate": 6.54845194713369e-07,
      "loss": 0.8121779,
      "num_input_tokens_seen": 132735590,
      "step": 6177,
      "time_per_iteration": 4.072268486022949
    },
    {
      "auxiliary_loss_clip": 0.01147979,
      "auxiliary_loss_mlp": 0.0102862,
      "balance_loss_clip": 1.04616809,
      "balance_loss_mlp": 1.02205801,
      "epoch": 0.742860578368304,
      "flos": 19898390102400.0,
      "grad_norm": 1.8916684867949831,
      "language_loss": 0.79498696,
      "learning_rate": 6.542688368763034e-07,
      "loss": 0.81675303,
      "num_input_tokens_seen": 132753995,
      "step": 6178,
      "time_per_iteration": 3.271073579788208
    },
    {
      "auxiliary_loss_clip": 0.01152102,
      "auxiliary_loss_mlp": 0.01027465,
      "balance_loss_clip": 1.0482794,
      "balance_loss_mlp": 1.02083421,
      "epoch": 0.742980821258943,
      "flos": 24827201510400.0,
      "grad_norm": 2.4925975872946586,
      "language_loss": 0.76932037,
      "learning_rate": 6.536926831855854e-07,
      "loss": 0.791116,
      "num_input_tokens_seen": 132773160,
      "step": 6179,
      "time_per_iteration": 2.45601749420166
    },
    {
      "auxiliary_loss_clip": 0.01135752,
      "auxiliary_loss_mlp": 0.01022823,
      "balance_loss_clip": 1.0452354,
      "balance_loss_mlp": 1.01607895,
      "epoch": 0.7431010641495821,
      "flos": 25228646887680.0,
      "grad_norm": 2.9300099208392836,
      "language_loss": 0.72909594,
      "learning_rate": 6.531167337286165e-07,
      "loss": 0.7506817,
      "num_input_tokens_seen": 132793180,
      "step": 6180,
      "time_per_iteration": 2.4815125465393066
    },
    {
      "auxiliary_loss_clip": 0.01138423,
      "auxiliary_loss_mlp": 0.0102296,
      "balance_loss_clip": 1.0474757,
      "balance_loss_mlp": 1.01621222,
      "epoch": 0.7432213070402213,
      "flos": 21762369550080.0,
      "grad_norm": 1.403660437381547,
      "language_loss": 0.7972523,
      "learning_rate": 6.52540988592768e-07,
      "loss": 0.81886613,
      "num_input_tokens_seen": 132814200,
      "step": 6181,
      "time_per_iteration": 2.4895172119140625
    },
    {
      "auxiliary_loss_clip": 0.01140588,
      "auxiliary_loss_mlp": 0.01020485,
      "balance_loss_clip": 1.04636407,
      "balance_loss_mlp": 1.01364517,
      "epoch": 0.7433415499308603,
      "flos": 14793832425600.0,
      "grad_norm": 1.9555727524976907,
      "language_loss": 0.83165836,
      "learning_rate": 6.519654478653814e-07,
      "loss": 0.85326904,
      "num_input_tokens_seen": 132832565,
      "step": 6182,
      "time_per_iteration": 3.2044496536254883
    },
    {
      "auxiliary_loss_clip": 0.01052846,
      "auxiliary_loss_mlp": 0.01002711,
      "balance_loss_clip": 1.01496482,
      "balance_loss_mlp": 1.00160837,
      "epoch": 0.7434617928214994,
      "flos": 67155577297920.0,
      "grad_norm": 0.7421561489254233,
      "language_loss": 0.5614903,
      "learning_rate": 6.51390111633763e-07,
      "loss": 0.58204579,
      "num_input_tokens_seen": 132897840,
      "step": 6183,
      "time_per_iteration": 3.105877637863159
    },
    {
      "auxiliary_loss_clip": 0.01094522,
      "auxiliary_loss_mlp": 0.01020144,
      "balance_loss_clip": 1.03875446,
      "balance_loss_mlp": 1.01344442,
      "epoch": 0.7435820357121385,
      "flos": 27377576928000.0,
      "grad_norm": 1.6357073275961822,
      "language_loss": 0.76207614,
      "learning_rate": 6.508149799851932e-07,
      "loss": 0.78322279,
      "num_input_tokens_seen": 132919505,
      "step": 6184,
      "time_per_iteration": 2.60357928276062
    },
    {
      "auxiliary_loss_clip": 0.01136877,
      "auxiliary_loss_mlp": 0.01021035,
      "balance_loss_clip": 1.04546118,
      "balance_loss_mlp": 1.01407325,
      "epoch": 0.7437022786027776,
      "flos": 23987645948160.0,
      "grad_norm": 2.062518575948394,
      "language_loss": 0.61426228,
      "learning_rate": 6.502400530069183e-07,
      "loss": 0.63584143,
      "num_input_tokens_seen": 132939390,
      "step": 6185,
      "time_per_iteration": 2.455746650695801
    },
    {
      "auxiliary_loss_clip": 0.01126645,
      "auxiliary_loss_mlp": 0.01030541,
      "balance_loss_clip": 1.04526186,
      "balance_loss_mlp": 1.0229888,
      "epoch": 0.7438225214934167,
      "flos": 21866761451520.0,
      "grad_norm": 1.6500706286797915,
      "language_loss": 0.68501008,
      "learning_rate": 6.496653307861535e-07,
      "loss": 0.70658195,
      "num_input_tokens_seen": 132960060,
      "step": 6186,
      "time_per_iteration": 2.522322416305542
    },
    {
      "auxiliary_loss_clip": 0.01157895,
      "auxiliary_loss_mlp": 0.01024997,
      "balance_loss_clip": 1.04782844,
      "balance_loss_mlp": 1.01790082,
      "epoch": 0.7439427643840558,
      "flos": 20230097224320.0,
      "grad_norm": 3.6776352129083842,
      "language_loss": 0.65812725,
      "learning_rate": 6.490908134100857e-07,
      "loss": 0.6799562,
      "num_input_tokens_seen": 132978525,
      "step": 6187,
      "time_per_iteration": 2.436244487762451
    },
    {
      "auxiliary_loss_clip": 0.01158552,
      "auxiliary_loss_mlp": 0.01021989,
      "balance_loss_clip": 1.04821706,
      "balance_loss_mlp": 1.01454389,
      "epoch": 0.7440630072746949,
      "flos": 20849915335680.0,
      "grad_norm": 2.2181268072344094,
      "language_loss": 0.69327259,
      "learning_rate": 6.48516500965866e-07,
      "loss": 0.71507794,
      "num_input_tokens_seen": 132998460,
      "step": 6188,
      "time_per_iteration": 2.444633722305298
    },
    {
      "auxiliary_loss_clip": 0.01154872,
      "auxiliary_loss_mlp": 0.01021909,
      "balance_loss_clip": 1.04431081,
      "balance_loss_mlp": 1.01470256,
      "epoch": 0.7441832501653339,
      "flos": 26503762769280.0,
      "grad_norm": 1.8008470146056523,
      "language_loss": 0.81646919,
      "learning_rate": 6.479423935406192e-07,
      "loss": 0.83823705,
      "num_input_tokens_seen": 133018445,
      "step": 6189,
      "time_per_iteration": 2.476483106613159
    },
    {
      "auxiliary_loss_clip": 0.01044096,
      "auxiliary_loss_mlp": 0.01001698,
      "balance_loss_clip": 1.01650774,
      "balance_loss_mlp": 1.00073802,
      "epoch": 0.7443034930559731,
      "flos": 68602848088320.0,
      "grad_norm": 0.8120585173931032,
      "language_loss": 0.61998612,
      "learning_rate": 6.473684912214357e-07,
      "loss": 0.64044404,
      "num_input_tokens_seen": 133082005,
      "step": 6190,
      "time_per_iteration": 3.1991283893585205
    },
    {
      "auxiliary_loss_clip": 0.01153062,
      "auxiliary_loss_mlp": 0.01021295,
      "balance_loss_clip": 1.04863572,
      "balance_loss_mlp": 1.01427984,
      "epoch": 0.7444237359466122,
      "flos": 18654982951680.0,
      "grad_norm": 5.5890299380191895,
      "language_loss": 0.69773829,
      "learning_rate": 6.467947940953778e-07,
      "loss": 0.71948195,
      "num_input_tokens_seen": 133100530,
      "step": 6191,
      "time_per_iteration": 2.4406604766845703
    },
    {
      "auxiliary_loss_clip": 0.01138434,
      "auxiliary_loss_mlp": 0.0102792,
      "balance_loss_clip": 1.04459417,
      "balance_loss_mlp": 1.02111959,
      "epoch": 0.7445439788372512,
      "flos": 22817604326400.0,
      "grad_norm": 1.7888926965732181,
      "language_loss": 0.72490942,
      "learning_rate": 6.462213022494732e-07,
      "loss": 0.74657297,
      "num_input_tokens_seen": 133119775,
      "step": 6192,
      "time_per_iteration": 2.484675645828247
    },
    {
      "auxiliary_loss_clip": 0.0106198,
      "auxiliary_loss_mlp": 0.01001569,
      "balance_loss_clip": 1.0152657,
      "balance_loss_mlp": 1.00056195,
      "epoch": 0.7446642217278904,
      "flos": 67045690615680.0,
      "grad_norm": 0.7677672766641063,
      "language_loss": 0.61051834,
      "learning_rate": 6.456480157707201e-07,
      "loss": 0.63115382,
      "num_input_tokens_seen": 133184550,
      "step": 6193,
      "time_per_iteration": 2.9859704971313477
    },
    {
      "auxiliary_loss_clip": 0.01118272,
      "auxiliary_loss_mlp": 0.01027632,
      "balance_loss_clip": 1.04254794,
      "balance_loss_mlp": 1.0199039,
      "epoch": 0.7447844646185294,
      "flos": 17417465631360.0,
      "grad_norm": 1.819256562295751,
      "language_loss": 0.84713626,
      "learning_rate": 6.450749347460866e-07,
      "loss": 0.86859524,
      "num_input_tokens_seen": 133201525,
      "step": 6194,
      "time_per_iteration": 2.4781603813171387
    },
    {
      "auxiliary_loss_clip": 0.01167648,
      "auxiliary_loss_mlp": 0.0102575,
      "balance_loss_clip": 1.04753089,
      "balance_loss_mlp": 1.01884818,
      "epoch": 0.7449047075091685,
      "flos": 26615876094720.0,
      "grad_norm": 1.8097698630998473,
      "language_loss": 0.78789788,
      "learning_rate": 6.445020592625083e-07,
      "loss": 0.80983186,
      "num_input_tokens_seen": 133222175,
      "step": 6195,
      "time_per_iteration": 2.4590728282928467
    },
    {
      "auxiliary_loss_clip": 0.01166421,
      "auxiliary_loss_mlp": 0.01026383,
      "balance_loss_clip": 1.04703259,
      "balance_loss_mlp": 1.01885176,
      "epoch": 0.7450249503998077,
      "flos": 14170458867840.0,
      "grad_norm": 2.310880236555306,
      "language_loss": 0.80482769,
      "learning_rate": 6.4392938940689e-07,
      "loss": 0.8267557,
      "num_input_tokens_seen": 133237590,
      "step": 6196,
      "time_per_iteration": 2.3711137771606445
    },
    {
      "auxiliary_loss_clip": 0.01109084,
      "auxiliary_loss_mlp": 0.00761268,
      "balance_loss_clip": 1.04393423,
      "balance_loss_mlp": 1.0003289,
      "epoch": 0.7451451932904467,
      "flos": 19606687752960.0,
      "grad_norm": 2.3412638046916,
      "language_loss": 0.72118419,
      "learning_rate": 6.433569252661049e-07,
      "loss": 0.73988771,
      "num_input_tokens_seen": 133255590,
      "step": 6197,
      "time_per_iteration": 2.5289671421051025
    },
    {
      "auxiliary_loss_clip": 0.01118567,
      "auxiliary_loss_mlp": 0.01023067,
      "balance_loss_clip": 1.04270315,
      "balance_loss_mlp": 1.01637971,
      "epoch": 0.7452654361810858,
      "flos": 12495405980160.0,
      "grad_norm": 1.7419545742608133,
      "language_loss": 0.71128464,
      "learning_rate": 6.427846669269952e-07,
      "loss": 0.73270094,
      "num_input_tokens_seen": 133273210,
      "step": 6198,
      "time_per_iteration": 2.473322629928589
    },
    {
      "auxiliary_loss_clip": 0.01170042,
      "auxiliary_loss_mlp": 0.01027548,
      "balance_loss_clip": 1.05222011,
      "balance_loss_mlp": 1.02088404,
      "epoch": 0.7453856790717249,
      "flos": 22127329687680.0,
      "grad_norm": 2.2693060423334908,
      "language_loss": 0.82625848,
      "learning_rate": 6.422126144763729e-07,
      "loss": 0.84823436,
      "num_input_tokens_seen": 133292600,
      "step": 6199,
      "time_per_iteration": 2.4111764430999756
    },
    {
      "auxiliary_loss_clip": 0.01123233,
      "auxiliary_loss_mlp": 0.00761984,
      "balance_loss_clip": 1.0410471,
      "balance_loss_mlp": 1.00040412,
      "epoch": 0.745505921962364,
      "flos": 20010682995840.0,
      "grad_norm": 3.7885517949085235,
      "language_loss": 0.76948673,
      "learning_rate": 6.416407680010174e-07,
      "loss": 0.7883389,
      "num_input_tokens_seen": 133306960,
      "step": 6200,
      "time_per_iteration": 2.4838271141052246
    },
    {
      "auxiliary_loss_clip": 0.01126198,
      "auxiliary_loss_mlp": 0.01027572,
      "balance_loss_clip": 1.04662418,
      "balance_loss_mlp": 1.01968622,
      "epoch": 0.745626164853003,
      "flos": 24677884673280.0,
      "grad_norm": 1.9412734105472431,
      "language_loss": 0.81055725,
      "learning_rate": 6.410691275876774e-07,
      "loss": 0.83209497,
      "num_input_tokens_seen": 133326380,
      "step": 6201,
      "time_per_iteration": 2.561248540878296
    },
    {
      "auxiliary_loss_clip": 0.01146038,
      "auxiliary_loss_mlp": 0.01023811,
      "balance_loss_clip": 1.04729605,
      "balance_loss_mlp": 1.0167954,
      "epoch": 0.7457464077436422,
      "flos": 14538830797440.0,
      "grad_norm": 3.983802267878781,
      "language_loss": 0.76758128,
      "learning_rate": 6.404976933230704e-07,
      "loss": 0.78927982,
      "num_input_tokens_seen": 133342900,
      "step": 6202,
      "time_per_iteration": 2.43813157081604
    },
    {
      "auxiliary_loss_clip": 0.01144489,
      "auxiliary_loss_mlp": 0.01025861,
      "balance_loss_clip": 1.04676485,
      "balance_loss_mlp": 1.01875329,
      "epoch": 0.7458666506342813,
      "flos": 34021194600960.0,
      "grad_norm": 2.05260282849613,
      "language_loss": 0.72575796,
      "learning_rate": 6.399264652938813e-07,
      "loss": 0.7474615,
      "num_input_tokens_seen": 133363805,
      "step": 6203,
      "time_per_iteration": 2.574030876159668
    },
    {
      "auxiliary_loss_clip": 0.01136499,
      "auxiliary_loss_mlp": 0.01022459,
      "balance_loss_clip": 1.04431498,
      "balance_loss_mlp": 1.01539612,
      "epoch": 0.7459868935249203,
      "flos": 24279025075200.0,
      "grad_norm": 1.9374598061480668,
      "language_loss": 0.74500263,
      "learning_rate": 6.393554435867679e-07,
      "loss": 0.76659214,
      "num_input_tokens_seen": 133384655,
      "step": 6204,
      "time_per_iteration": 4.084227561950684
    },
    {
      "auxiliary_loss_clip": 0.01121001,
      "auxiliary_loss_mlp": 0.01023674,
      "balance_loss_clip": 1.04250634,
      "balance_loss_mlp": 1.01597285,
      "epoch": 0.7461071364155595,
      "flos": 21908777385600.0,
      "grad_norm": 1.9800080127206716,
      "language_loss": 0.83316207,
      "learning_rate": 6.387846282883502e-07,
      "loss": 0.85460889,
      "num_input_tokens_seen": 133401185,
      "step": 6205,
      "time_per_iteration": 3.3325726985931396
    },
    {
      "auxiliary_loss_clip": 0.01164902,
      "auxiliary_loss_mlp": 0.01021893,
      "balance_loss_clip": 1.04754555,
      "balance_loss_mlp": 1.0149188,
      "epoch": 0.7462273793061985,
      "flos": 22889712879360.0,
      "grad_norm": 2.9273996422156014,
      "language_loss": 0.76747751,
      "learning_rate": 6.38214019485223e-07,
      "loss": 0.78934544,
      "num_input_tokens_seen": 133420010,
      "step": 6206,
      "time_per_iteration": 2.4077541828155518
    },
    {
      "auxiliary_loss_clip": 0.01093658,
      "auxiliary_loss_mlp": 0.01024832,
      "balance_loss_clip": 1.04005182,
      "balance_loss_mlp": 1.01764965,
      "epoch": 0.7463476221968376,
      "flos": 19968451580160.0,
      "grad_norm": 1.859492413114982,
      "language_loss": 0.71791303,
      "learning_rate": 6.376436172639461e-07,
      "loss": 0.73909795,
      "num_input_tokens_seen": 133437855,
      "step": 6207,
      "time_per_iteration": 2.5536484718322754
    },
    {
      "auxiliary_loss_clip": 0.01088513,
      "auxiliary_loss_mlp": 0.01030115,
      "balance_loss_clip": 1.04204917,
      "balance_loss_mlp": 1.0219785,
      "epoch": 0.7464678650874768,
      "flos": 16836610798080.0,
      "grad_norm": 2.3585745994790352,
      "language_loss": 0.64730155,
      "learning_rate": 6.370734217110487e-07,
      "loss": 0.66848779,
      "num_input_tokens_seen": 133456600,
      "step": 6208,
      "time_per_iteration": 3.3388936519622803
    },
    {
      "auxiliary_loss_clip": 0.01142146,
      "auxiliary_loss_mlp": 0.01027459,
      "balance_loss_clip": 1.04774106,
      "balance_loss_mlp": 1.01996636,
      "epoch": 0.7465881079781158,
      "flos": 48100869843840.0,
      "grad_norm": 1.4505603185848304,
      "language_loss": 0.64256763,
      "learning_rate": 6.36503432913031e-07,
      "loss": 0.66426361,
      "num_input_tokens_seen": 133479745,
      "step": 6209,
      "time_per_iteration": 2.729149103164673
    },
    {
      "auxiliary_loss_clip": 0.01150543,
      "auxiliary_loss_mlp": 0.01024486,
      "balance_loss_clip": 1.04698348,
      "balance_loss_mlp": 1.01703,
      "epoch": 0.7467083508687549,
      "flos": 19677359761920.0,
      "grad_norm": 2.0386818034073877,
      "language_loss": 0.68967742,
      "learning_rate": 6.359336509563569e-07,
      "loss": 0.71142769,
      "num_input_tokens_seen": 133495765,
      "step": 6210,
      "time_per_iteration": 2.407444953918457
    },
    {
      "auxiliary_loss_clip": 0.01111907,
      "auxiliary_loss_mlp": 0.01030829,
      "balance_loss_clip": 1.04248261,
      "balance_loss_mlp": 1.02367055,
      "epoch": 0.7468285937593939,
      "flos": 17895436934400.0,
      "grad_norm": 1.7850169020944548,
      "language_loss": 0.80585963,
      "learning_rate": 6.353640759274641e-07,
      "loss": 0.82728696,
      "num_input_tokens_seen": 133514655,
      "step": 6211,
      "time_per_iteration": 2.466000556945801
    },
    {
      "auxiliary_loss_clip": 0.01148716,
      "auxiliary_loss_mlp": 0.01021498,
      "balance_loss_clip": 1.0438211,
      "balance_loss_mlp": 1.01427948,
      "epoch": 0.7469488366500331,
      "flos": 23141446369920.0,
      "grad_norm": 3.815208700603355,
      "language_loss": 0.73478317,
      "learning_rate": 6.347947079127556e-07,
      "loss": 0.75648534,
      "num_input_tokens_seen": 133532555,
      "step": 6212,
      "time_per_iteration": 2.448587656021118
    },
    {
      "auxiliary_loss_clip": 0.01135783,
      "auxiliary_loss_mlp": 0.0102314,
      "balance_loss_clip": 1.04566979,
      "balance_loss_mlp": 1.01576042,
      "epoch": 0.7470690795406721,
      "flos": 16690849407360.0,
      "grad_norm": 2.2187525466721603,
      "language_loss": 0.77419782,
      "learning_rate": 6.342255469986053e-07,
      "loss": 0.79578704,
      "num_input_tokens_seen": 133551300,
      "step": 6213,
      "time_per_iteration": 2.444871187210083
    },
    {
      "auxiliary_loss_clip": 0.0116441,
      "auxiliary_loss_mlp": 0.01025067,
      "balance_loss_clip": 1.04717588,
      "balance_loss_mlp": 1.01775301,
      "epoch": 0.7471893224313112,
      "flos": 25192700352000.0,
      "grad_norm": 1.8812834755729595,
      "language_loss": 0.76202929,
      "learning_rate": 6.336565932713533e-07,
      "loss": 0.7839241,
      "num_input_tokens_seen": 133570725,
      "step": 6214,
      "time_per_iteration": 2.43023419380188
    },
    {
      "auxiliary_loss_clip": 0.0113915,
      "auxiliary_loss_mlp": 0.01025334,
      "balance_loss_clip": 1.0497694,
      "balance_loss_mlp": 1.01788592,
      "epoch": 0.7473095653219504,
      "flos": 22526225199360.0,
      "grad_norm": 1.9257933638632283,
      "language_loss": 0.77874303,
      "learning_rate": 6.330878468173088e-07,
      "loss": 0.80038786,
      "num_input_tokens_seen": 133590790,
      "step": 6215,
      "time_per_iteration": 2.4914753437042236
    },
    {
      "auxiliary_loss_clip": 0.01144302,
      "auxiliary_loss_mlp": 0.01021666,
      "balance_loss_clip": 1.044276,
      "balance_loss_mlp": 1.01465321,
      "epoch": 0.7474298082125894,
      "flos": 18113989236480.0,
      "grad_norm": 1.833161394688964,
      "language_loss": 0.73129159,
      "learning_rate": 6.32519307722752e-07,
      "loss": 0.75295126,
      "num_input_tokens_seen": 133608685,
      "step": 6216,
      "time_per_iteration": 2.413458824157715
    },
    {
      "auxiliary_loss_clip": 0.01040271,
      "auxiliary_loss_mlp": 0.01002351,
      "balance_loss_clip": 1.02107894,
      "balance_loss_mlp": 1.00124824,
      "epoch": 0.7475500511032285,
      "flos": 62086535193600.0,
      "grad_norm": 0.9147310792187742,
      "language_loss": 0.54990089,
      "learning_rate": 6.31950976073929e-07,
      "loss": 0.5703271,
      "num_input_tokens_seen": 133662775,
      "step": 6217,
      "time_per_iteration": 3.065585136413574
    },
    {
      "auxiliary_loss_clip": 0.0111098,
      "auxiliary_loss_mlp": 0.01023694,
      "balance_loss_clip": 1.04553318,
      "balance_loss_mlp": 1.01650882,
      "epoch": 0.7476702939938676,
      "flos": 17785586165760.0,
      "grad_norm": 2.8981021199372297,
      "language_loss": 0.8071233,
      "learning_rate": 6.31382851957055e-07,
      "loss": 0.82847011,
      "num_input_tokens_seen": 133679595,
      "step": 6218,
      "time_per_iteration": 2.5134427547454834
    },
    {
      "auxiliary_loss_clip": 0.01121556,
      "auxiliary_loss_mlp": 0.00761805,
      "balance_loss_clip": 1.04384255,
      "balance_loss_mlp": 1.00042236,
      "epoch": 0.7477905368845067,
      "flos": 27927944092800.0,
      "grad_norm": 1.9047534806383175,
      "language_loss": 0.7137059,
      "learning_rate": 6.308149354583143e-07,
      "loss": 0.73253953,
      "num_input_tokens_seen": 133699000,
      "step": 6219,
      "time_per_iteration": 2.542675256729126
    },
    {
      "auxiliary_loss_clip": 0.01158034,
      "auxiliary_loss_mlp": 0.01027928,
      "balance_loss_clip": 1.04858708,
      "balance_loss_mlp": 1.02016139,
      "epoch": 0.7479107797751458,
      "flos": 26870374932480.0,
      "grad_norm": 2.067737699056454,
      "language_loss": 0.82009405,
      "learning_rate": 6.302472266638586e-07,
      "loss": 0.84195369,
      "num_input_tokens_seen": 133719540,
      "step": 6220,
      "time_per_iteration": 2.4710311889648438
    },
    {
      "auxiliary_loss_clip": 0.01174272,
      "auxiliary_loss_mlp": 0.01027114,
      "balance_loss_clip": 1.05049396,
      "balance_loss_mlp": 1.01959229,
      "epoch": 0.7480310226657849,
      "flos": 33943375785600.0,
      "grad_norm": 2.3184549032649464,
      "language_loss": 0.70081878,
      "learning_rate": 6.296797256598101e-07,
      "loss": 0.72283262,
      "num_input_tokens_seen": 133741020,
      "step": 6221,
      "time_per_iteration": 2.5132362842559814
    },
    {
      "auxiliary_loss_clip": 0.01115356,
      "auxiliary_loss_mlp": 0.01020992,
      "balance_loss_clip": 1.04266918,
      "balance_loss_mlp": 1.01424766,
      "epoch": 0.748151265556424,
      "flos": 24826555065600.0,
      "grad_norm": 1.6507157849663419,
      "language_loss": 0.81287891,
      "learning_rate": 6.291124325322576e-07,
      "loss": 0.83424234,
      "num_input_tokens_seen": 133761145,
      "step": 6222,
      "time_per_iteration": 2.529958486557007
    },
    {
      "auxiliary_loss_clip": 0.01144576,
      "auxiliary_loss_mlp": 0.01021906,
      "balance_loss_clip": 1.04648924,
      "balance_loss_mlp": 1.01461053,
      "epoch": 0.748271508447063,
      "flos": 38399351535360.0,
      "grad_norm": 1.5315019259196678,
      "language_loss": 0.62317777,
      "learning_rate": 6.285453473672595e-07,
      "loss": 0.64484257,
      "num_input_tokens_seen": 133783715,
      "step": 6223,
      "time_per_iteration": 2.622898817062378
    },
    {
      "auxiliary_loss_clip": 0.01162977,
      "auxiliary_loss_mlp": 0.01022608,
      "balance_loss_clip": 1.04584289,
      "balance_loss_mlp": 1.01568222,
      "epoch": 0.7483917513377022,
      "flos": 21541842000000.0,
      "grad_norm": 1.8350760105936486,
      "language_loss": 0.75425625,
      "learning_rate": 6.279784702508415e-07,
      "loss": 0.77611214,
      "num_input_tokens_seen": 133804465,
      "step": 6224,
      "time_per_iteration": 2.430347442626953
    },
    {
      "auxiliary_loss_clip": 0.01042139,
      "auxiliary_loss_mlp": 0.01001089,
      "balance_loss_clip": 1.01416397,
      "balance_loss_mlp": 1.00000405,
      "epoch": 0.7485119942283412,
      "flos": 62314532772480.0,
      "grad_norm": 0.7883586487722969,
      "language_loss": 0.5861026,
      "learning_rate": 6.274118012689979e-07,
      "loss": 0.6065349,
      "num_input_tokens_seen": 133866365,
      "step": 6225,
      "time_per_iteration": 3.1924705505371094
    },
    {
      "auxiliary_loss_clip": 0.01132486,
      "auxiliary_loss_mlp": 0.01020595,
      "balance_loss_clip": 1.04461217,
      "balance_loss_mlp": 1.01350808,
      "epoch": 0.7486322371189803,
      "flos": 29937613104000.0,
      "grad_norm": 1.4412620309893303,
      "language_loss": 0.67983925,
      "learning_rate": 6.268453405076943e-07,
      "loss": 0.70137006,
      "num_input_tokens_seen": 133888760,
      "step": 6226,
      "time_per_iteration": 2.5535829067230225
    },
    {
      "auxiliary_loss_clip": 0.01137645,
      "auxiliary_loss_mlp": 0.01020275,
      "balance_loss_clip": 1.04396915,
      "balance_loss_mlp": 1.01374507,
      "epoch": 0.7487524800096195,
      "flos": 18949414734720.0,
      "grad_norm": 1.8144143569730453,
      "language_loss": 0.82340175,
      "learning_rate": 6.262790880528592e-07,
      "loss": 0.84498096,
      "num_input_tokens_seen": 133906380,
      "step": 6227,
      "time_per_iteration": 2.4477949142456055
    },
    {
      "auxiliary_loss_clip": 0.01134435,
      "auxiliary_loss_mlp": 0.01030438,
      "balance_loss_clip": 1.04232669,
      "balance_loss_mlp": 1.02282381,
      "epoch": 0.7488727229002585,
      "flos": 18697393935360.0,
      "grad_norm": 2.344175243345799,
      "language_loss": 0.79441434,
      "learning_rate": 6.257130439903951e-07,
      "loss": 0.81606305,
      "num_input_tokens_seen": 133922875,
      "step": 6228,
      "time_per_iteration": 2.4687416553497314
    },
    {
      "auxiliary_loss_clip": 0.01170136,
      "auxiliary_loss_mlp": 0.01024206,
      "balance_loss_clip": 1.05075121,
      "balance_loss_mlp": 1.0170629,
      "epoch": 0.7489929657908976,
      "flos": 23623368168960.0,
      "grad_norm": 1.8729090487936177,
      "language_loss": 0.81070983,
      "learning_rate": 6.251472084061695e-07,
      "loss": 0.83265328,
      "num_input_tokens_seen": 133941795,
      "step": 6229,
      "time_per_iteration": 2.4162254333496094
    },
    {
      "auxiliary_loss_clip": 0.01153283,
      "auxiliary_loss_mlp": 0.01023761,
      "balance_loss_clip": 1.04921317,
      "balance_loss_mlp": 1.01695156,
      "epoch": 0.7491132086815367,
      "flos": 20551533056640.0,
      "grad_norm": 1.876590929168162,
      "language_loss": 0.88928324,
      "learning_rate": 6.245815813860191e-07,
      "loss": 0.91105366,
      "num_input_tokens_seen": 133957305,
      "step": 6230,
      "time_per_iteration": 3.991645336151123
    },
    {
      "auxiliary_loss_clip": 0.01169023,
      "auxiliary_loss_mlp": 0.01023028,
      "balance_loss_clip": 1.04729509,
      "balance_loss_mlp": 1.01581001,
      "epoch": 0.7492334515721758,
      "flos": 23003011353600.0,
      "grad_norm": 1.9645323292033885,
      "language_loss": 0.70029616,
      "learning_rate": 6.240161630157495e-07,
      "loss": 0.72221667,
      "num_input_tokens_seen": 133976660,
      "step": 6231,
      "time_per_iteration": 2.4113821983337402
    },
    {
      "auxiliary_loss_clip": 0.01169446,
      "auxiliary_loss_mlp": 0.01023414,
      "balance_loss_clip": 1.04816628,
      "balance_loss_mlp": 1.01628804,
      "epoch": 0.7493536944628149,
      "flos": 16398823835520.0,
      "grad_norm": 1.84444668515726,
      "language_loss": 0.70071959,
      "learning_rate": 6.23450953381133e-07,
      "loss": 0.72264814,
      "num_input_tokens_seen": 133994750,
      "step": 6232,
      "time_per_iteration": 3.219994306564331
    },
    {
      "auxiliary_loss_clip": 0.0113269,
      "auxiliary_loss_mlp": 0.01023618,
      "balance_loss_clip": 1.04464459,
      "balance_loss_mlp": 1.01659024,
      "epoch": 0.749473937353454,
      "flos": 15338561155200.0,
      "grad_norm": 1.9465253736845634,
      "language_loss": 0.68101096,
      "learning_rate": 6.228859525679131e-07,
      "loss": 0.70257407,
      "num_input_tokens_seen": 134009165,
      "step": 6233,
      "time_per_iteration": 2.4291491508483887
    },
    {
      "auxiliary_loss_clip": 0.01151357,
      "auxiliary_loss_mlp": 0.01023636,
      "balance_loss_clip": 1.04587197,
      "balance_loss_mlp": 1.01654911,
      "epoch": 0.7495941802440931,
      "flos": 18951138587520.0,
      "grad_norm": 2.370861030985141,
      "language_loss": 0.79979414,
      "learning_rate": 6.223211606617986e-07,
      "loss": 0.82154411,
      "num_input_tokens_seen": 134027585,
      "step": 6234,
      "time_per_iteration": 2.4236137866973877
    },
    {
      "auxiliary_loss_clip": 0.01151537,
      "auxiliary_loss_mlp": 0.01023931,
      "balance_loss_clip": 1.05012703,
      "balance_loss_mlp": 1.01771128,
      "epoch": 0.7497144231347321,
      "flos": 22492469393280.0,
      "grad_norm": 1.8198836772299694,
      "language_loss": 0.83996075,
      "learning_rate": 6.217565777484701e-07,
      "loss": 0.86171544,
      "num_input_tokens_seen": 134046680,
      "step": 6235,
      "time_per_iteration": 3.1726624965667725
    },
    {
      "auxiliary_loss_clip": 0.01134361,
      "auxiliary_loss_mlp": 0.00761634,
      "balance_loss_clip": 1.04438329,
      "balance_loss_mlp": 1.00038457,
      "epoch": 0.7498346660253713,
      "flos": 24243509502720.0,
      "grad_norm": 1.7571962626905717,
      "language_loss": 0.80194199,
      "learning_rate": 6.211922039135722e-07,
      "loss": 0.82090193,
      "num_input_tokens_seen": 134066825,
      "step": 6236,
      "time_per_iteration": 2.5021450519561768
    },
    {
      "auxiliary_loss_clip": 0.01168505,
      "auxiliary_loss_mlp": 0.01025935,
      "balance_loss_clip": 1.049384,
      "balance_loss_mlp": 1.01885366,
      "epoch": 0.7499549089160104,
      "flos": 24387080163840.0,
      "grad_norm": 2.9861287230376203,
      "language_loss": 0.80835712,
      "learning_rate": 6.206280392427201e-07,
      "loss": 0.83030152,
      "num_input_tokens_seen": 134086410,
      "step": 6237,
      "time_per_iteration": 2.4432663917541504
    },
    {
      "auxiliary_loss_clip": 0.01146797,
      "auxiliary_loss_mlp": 0.01021113,
      "balance_loss_clip": 1.04471016,
      "balance_loss_mlp": 1.01388574,
      "epoch": 0.7500751518066494,
      "flos": 34057320704640.0,
      "grad_norm": 1.568529236453423,
      "language_loss": 0.73682833,
      "learning_rate": 6.200640838214983e-07,
      "loss": 0.75850737,
      "num_input_tokens_seen": 134109185,
      "step": 6238,
      "time_per_iteration": 2.5562078952789307
    },
    {
      "auxiliary_loss_clip": 0.01166918,
      "auxiliary_loss_mlp": 0.01025921,
      "balance_loss_clip": 1.04849482,
      "balance_loss_mlp": 1.01878011,
      "epoch": 0.7501953946972886,
      "flos": 18843586289280.0,
      "grad_norm": 1.8420751057366733,
      "language_loss": 0.66660023,
      "learning_rate": 6.195003377354578e-07,
      "loss": 0.68852854,
      "num_input_tokens_seen": 134128455,
      "step": 6239,
      "time_per_iteration": 2.402932643890381
    },
    {
      "auxiliary_loss_clip": 0.01150001,
      "auxiliary_loss_mlp": 0.01029442,
      "balance_loss_clip": 1.04455245,
      "balance_loss_mlp": 1.02169895,
      "epoch": 0.7503156375879276,
      "flos": 20257675891200.0,
      "grad_norm": 2.626204985629956,
      "language_loss": 0.72696888,
      "learning_rate": 6.189368010701183e-07,
      "loss": 0.74876332,
      "num_input_tokens_seen": 134145515,
      "step": 6240,
      "time_per_iteration": 2.4027717113494873
    },
    {
      "auxiliary_loss_clip": 0.01157392,
      "auxiliary_loss_mlp": 0.01025955,
      "balance_loss_clip": 1.04523158,
      "balance_loss_mlp": 1.01860261,
      "epoch": 0.7504358804785667,
      "flos": 13480040574720.0,
      "grad_norm": 1.9453612773620705,
      "language_loss": 0.76529676,
      "learning_rate": 6.183734739109683e-07,
      "loss": 0.78713024,
      "num_input_tokens_seen": 134163335,
      "step": 6241,
      "time_per_iteration": 2.4118711948394775
    },
    {
      "auxiliary_loss_clip": 0.01160642,
      "auxiliary_loss_mlp": 0.01025354,
      "balance_loss_clip": 1.04854369,
      "balance_loss_mlp": 1.01773643,
      "epoch": 0.7505561233692057,
      "flos": 29461042431360.0,
      "grad_norm": 2.2504640541617373,
      "language_loss": 0.68508184,
      "learning_rate": 6.178103563434629e-07,
      "loss": 0.70694178,
      "num_input_tokens_seen": 134182335,
      "step": 6242,
      "time_per_iteration": 2.481496572494507
    },
    {
      "auxiliary_loss_clip": 0.01166838,
      "auxiliary_loss_mlp": 0.01029457,
      "balance_loss_clip": 1.04749024,
      "balance_loss_mlp": 1.02251315,
      "epoch": 0.7506763662598449,
      "flos": 20302457172480.0,
      "grad_norm": 1.7734571656123108,
      "language_loss": 0.83663851,
      "learning_rate": 6.172474484530283e-07,
      "loss": 0.85860145,
      "num_input_tokens_seen": 134201070,
      "step": 6243,
      "time_per_iteration": 2.3832345008850098
    },
    {
      "auxiliary_loss_clip": 0.01129106,
      "auxiliary_loss_mlp": 0.01025663,
      "balance_loss_clip": 1.0410111,
      "balance_loss_mlp": 1.01842117,
      "epoch": 0.750796609150484,
      "flos": 37230961939200.0,
      "grad_norm": 2.171412603974266,
      "language_loss": 0.75914609,
      "learning_rate": 6.166847503250563e-07,
      "loss": 0.78069377,
      "num_input_tokens_seen": 134223310,
      "step": 6244,
      "time_per_iteration": 2.6111512184143066
    },
    {
      "auxiliary_loss_clip": 0.01141789,
      "auxiliary_loss_mlp": 0.01023911,
      "balance_loss_clip": 1.04663026,
      "balance_loss_mlp": 1.01670218,
      "epoch": 0.750916852041123,
      "flos": 19609417186560.0,
      "grad_norm": 2.3001814639975025,
      "language_loss": 0.79256475,
      "learning_rate": 6.161222620449078e-07,
      "loss": 0.81422174,
      "num_input_tokens_seen": 134242085,
      "step": 6245,
      "time_per_iteration": 2.5093770027160645
    },
    {
      "auxiliary_loss_clip": 0.01126965,
      "auxiliary_loss_mlp": 0.01028541,
      "balance_loss_clip": 1.04454315,
      "balance_loss_mlp": 1.0212009,
      "epoch": 0.7510370949317622,
      "flos": 25112690807040.0,
      "grad_norm": 6.03203468351411,
      "language_loss": 0.79935658,
      "learning_rate": 6.155599836979117e-07,
      "loss": 0.82091165,
      "num_input_tokens_seen": 134260770,
      "step": 6246,
      "time_per_iteration": 2.533597230911255
    },
    {
      "auxiliary_loss_clip": 0.01110671,
      "auxiliary_loss_mlp": 0.01028378,
      "balance_loss_clip": 1.04131794,
      "balance_loss_mlp": 1.02083826,
      "epoch": 0.7511573378224012,
      "flos": 19062282245760.0,
      "grad_norm": 2.476737835506978,
      "language_loss": 0.81687248,
      "learning_rate": 6.149979153693649e-07,
      "loss": 0.83826292,
      "num_input_tokens_seen": 134278025,
      "step": 6247,
      "time_per_iteration": 2.517467498779297
    },
    {
      "auxiliary_loss_clip": 0.01150882,
      "auxiliary_loss_mlp": 0.01021136,
      "balance_loss_clip": 1.04562283,
      "balance_loss_mlp": 1.01387036,
      "epoch": 0.7512775807130403,
      "flos": 19937676602880.0,
      "grad_norm": 1.9242567808821216,
      "language_loss": 0.76984453,
      "learning_rate": 6.144360571445343e-07,
      "loss": 0.7915647,
      "num_input_tokens_seen": 134297170,
      "step": 6248,
      "time_per_iteration": 2.4166879653930664
    },
    {
      "auxiliary_loss_clip": 0.01153255,
      "auxiliary_loss_mlp": 0.01024226,
      "balance_loss_clip": 1.04967964,
      "balance_loss_mlp": 1.01670694,
      "epoch": 0.7513978236036795,
      "flos": 20739920912640.0,
      "grad_norm": 1.677912911373662,
      "language_loss": 0.80133832,
      "learning_rate": 6.138744091086509e-07,
      "loss": 0.82311308,
      "num_input_tokens_seen": 134316755,
      "step": 6249,
      "time_per_iteration": 2.4671390056610107
    },
    {
      "auxiliary_loss_clip": 0.01129486,
      "auxiliary_loss_mlp": 0.01023699,
      "balance_loss_clip": 1.04612064,
      "balance_loss_mlp": 1.01657593,
      "epoch": 0.7515180664943185,
      "flos": 27563163523200.0,
      "grad_norm": 2.9477022416916894,
      "language_loss": 0.72612911,
      "learning_rate": 6.133129713469183e-07,
      "loss": 0.74766099,
      "num_input_tokens_seen": 134335960,
      "step": 6250,
      "time_per_iteration": 2.5370397567749023
    },
    {
      "auxiliary_loss_clip": 0.0113391,
      "auxiliary_loss_mlp": 0.01024471,
      "balance_loss_clip": 1.04283834,
      "balance_loss_mlp": 1.0167762,
      "epoch": 0.7516383093849576,
      "flos": 33803181002880.0,
      "grad_norm": 1.642104574358395,
      "language_loss": 0.63948023,
      "learning_rate": 6.127517439445053e-07,
      "loss": 0.66106409,
      "num_input_tokens_seen": 134356805,
      "step": 6251,
      "time_per_iteration": 2.640925645828247
    },
    {
      "auxiliary_loss_clip": 0.0110339,
      "auxiliary_loss_mlp": 0.01025942,
      "balance_loss_clip": 1.04175401,
      "balance_loss_mlp": 1.01900971,
      "epoch": 0.7517585522755967,
      "flos": 29746172592000.0,
      "grad_norm": 1.7587436804556467,
      "language_loss": 0.81388497,
      "learning_rate": 6.121907269865498e-07,
      "loss": 0.83517826,
      "num_input_tokens_seen": 134376295,
      "step": 6252,
      "time_per_iteration": 2.587433099746704
    },
    {
      "auxiliary_loss_clip": 0.01033395,
      "auxiliary_loss_mlp": 0.01001507,
      "balance_loss_clip": 1.01297998,
      "balance_loss_mlp": 1.00034511,
      "epoch": 0.7518787951662358,
      "flos": 69807974319360.0,
      "grad_norm": 0.9252908462086723,
      "language_loss": 0.67270464,
      "learning_rate": 6.116299205581577e-07,
      "loss": 0.69305366,
      "num_input_tokens_seen": 134431125,
      "step": 6253,
      "time_per_iteration": 3.0550522804260254
    },
    {
      "auxiliary_loss_clip": 0.01172567,
      "auxiliary_loss_mlp": 0.01029086,
      "balance_loss_clip": 1.05047369,
      "balance_loss_mlp": 1.02153063,
      "epoch": 0.7519990380568748,
      "flos": 34203225749760.0,
      "grad_norm": 1.9753732488135523,
      "language_loss": 0.684681,
      "learning_rate": 6.110693247444018e-07,
      "loss": 0.70669746,
      "num_input_tokens_seen": 134452960,
      "step": 6254,
      "time_per_iteration": 2.520676374435425
    },
    {
      "auxiliary_loss_clip": 0.0111267,
      "auxiliary_loss_mlp": 0.01021024,
      "balance_loss_clip": 1.04187322,
      "balance_loss_mlp": 1.01428902,
      "epoch": 0.752119280947514,
      "flos": 21725704742400.0,
      "grad_norm": 2.126531339211925,
      "language_loss": 0.82428157,
      "learning_rate": 6.105089396303258e-07,
      "loss": 0.84561855,
      "num_input_tokens_seen": 134471350,
      "step": 6255,
      "time_per_iteration": 2.487708568572998
    },
    {
      "auxiliary_loss_clip": 0.01138127,
      "auxiliary_loss_mlp": 0.01028442,
      "balance_loss_clip": 1.04393554,
      "balance_loss_mlp": 1.02078891,
      "epoch": 0.7522395238381531,
      "flos": 32742774668160.0,
      "grad_norm": 2.0933952787177934,
      "language_loss": 0.75474834,
      "learning_rate": 6.099487653009383e-07,
      "loss": 0.77641404,
      "num_input_tokens_seen": 134490695,
      "step": 6256,
      "time_per_iteration": 2.5547077655792236
    },
    {
      "auxiliary_loss_clip": 0.01150694,
      "auxiliary_loss_mlp": 0.01021254,
      "balance_loss_clip": 1.04513693,
      "balance_loss_mlp": 1.01500177,
      "epoch": 0.7523597667287921,
      "flos": 23476026579840.0,
      "grad_norm": 1.981453671770173,
      "language_loss": 0.82870805,
      "learning_rate": 6.093888018412192e-07,
      "loss": 0.85042763,
      "num_input_tokens_seen": 134506885,
      "step": 6257,
      "time_per_iteration": 4.064921140670776
    },
    {
      "auxiliary_loss_clip": 0.01060228,
      "auxiliary_loss_mlp": 0.01001605,
      "balance_loss_clip": 1.01348948,
      "balance_loss_mlp": 1.00044858,
      "epoch": 0.7524800096194313,
      "flos": 67346730501120.0,
      "grad_norm": 0.706416242686214,
      "language_loss": 0.54675937,
      "learning_rate": 6.088290493361125e-07,
      "loss": 0.56737769,
      "num_input_tokens_seen": 134571770,
      "step": 6258,
      "time_per_iteration": 3.1746506690979004
    },
    {
      "auxiliary_loss_clip": 0.01103374,
      "auxiliary_loss_mlp": 0.01025143,
      "balance_loss_clip": 1.04152441,
      "balance_loss_mlp": 1.01802027,
      "epoch": 0.7526002525100703,
      "flos": 13006055681280.0,
      "grad_norm": 2.1684632685671823,
      "language_loss": 0.71786094,
      "learning_rate": 6.082695078705322e-07,
      "loss": 0.73914611,
      "num_input_tokens_seen": 134589250,
      "step": 6259,
      "time_per_iteration": 3.35184907913208
    },
    {
      "auxiliary_loss_clip": 0.01146666,
      "auxiliary_loss_mlp": 0.01025885,
      "balance_loss_clip": 1.04649723,
      "balance_loss_mlp": 1.01827335,
      "epoch": 0.7527204954007094,
      "flos": 21397229844480.0,
      "grad_norm": 2.0456011277912367,
      "language_loss": 0.69125962,
      "learning_rate": 6.077101775293618e-07,
      "loss": 0.71298516,
      "num_input_tokens_seen": 134608075,
      "step": 6260,
      "time_per_iteration": 2.4235687255859375
    },
    {
      "auxiliary_loss_clip": 0.01155083,
      "auxiliary_loss_mlp": 0.0102883,
      "balance_loss_clip": 1.04685998,
      "balance_loss_mlp": 1.02077448,
      "epoch": 0.7528407382913486,
      "flos": 18947188091520.0,
      "grad_norm": 2.4849077527889643,
      "language_loss": 0.82339543,
      "learning_rate": 6.071510583974504e-07,
      "loss": 0.84523457,
      "num_input_tokens_seen": 134623260,
      "step": 6261,
      "time_per_iteration": 2.3993842601776123
    },
    {
      "auxiliary_loss_clip": 0.01170101,
      "auxiliary_loss_mlp": 0.01029477,
      "balance_loss_clip": 1.04961562,
      "balance_loss_mlp": 1.02209187,
      "epoch": 0.7529609811819876,
      "flos": 15231798956160.0,
      "grad_norm": 1.8486718868682808,
      "language_loss": 0.71922195,
      "learning_rate": 6.065921505596161e-07,
      "loss": 0.74121767,
      "num_input_tokens_seen": 134641540,
      "step": 6262,
      "time_per_iteration": 3.09307599067688
    },
    {
      "auxiliary_loss_clip": 0.01125493,
      "auxiliary_loss_mlp": 0.01023624,
      "balance_loss_clip": 1.04510975,
      "balance_loss_mlp": 1.01632512,
      "epoch": 0.7530812240726267,
      "flos": 19354487385600.0,
      "grad_norm": 1.6740837587178967,
      "language_loss": 0.77016187,
      "learning_rate": 6.060334541006445e-07,
      "loss": 0.79165304,
      "num_input_tokens_seen": 134660035,
      "step": 6263,
      "time_per_iteration": 2.4801292419433594
    },
    {
      "auxiliary_loss_clip": 0.01127439,
      "auxiliary_loss_mlp": 0.0102579,
      "balance_loss_clip": 1.04247904,
      "balance_loss_mlp": 1.01879549,
      "epoch": 0.7532014669632658,
      "flos": 27748247328000.0,
      "grad_norm": 1.4816375557735484,
      "language_loss": 0.69224095,
      "learning_rate": 6.05474969105289e-07,
      "loss": 0.71377319,
      "num_input_tokens_seen": 134683025,
      "step": 6264,
      "time_per_iteration": 2.579890727996826
    },
    {
      "auxiliary_loss_clip": 0.01154516,
      "auxiliary_loss_mlp": 0.01022254,
      "balance_loss_clip": 1.04752994,
      "balance_loss_mlp": 1.01479459,
      "epoch": 0.7533217098539049,
      "flos": 14137421333760.0,
      "grad_norm": 2.1586156920426607,
      "language_loss": 0.73210609,
      "learning_rate": 6.049166956582725e-07,
      "loss": 0.75387383,
      "num_input_tokens_seen": 134701290,
      "step": 6265,
      "time_per_iteration": 2.4073972702026367
    },
    {
      "auxiliary_loss_clip": 0.01148341,
      "auxiliary_loss_mlp": 0.01019485,
      "balance_loss_clip": 1.04510522,
      "balance_loss_mlp": 1.01263928,
      "epoch": 0.753441952744544,
      "flos": 26429068437120.0,
      "grad_norm": 2.938928415162882,
      "language_loss": 0.87530792,
      "learning_rate": 6.043586338442841e-07,
      "loss": 0.89698619,
      "num_input_tokens_seen": 134720345,
      "step": 6266,
      "time_per_iteration": 2.4748902320861816
    },
    {
      "auxiliary_loss_clip": 0.01164837,
      "auxiliary_loss_mlp": 0.01021315,
      "balance_loss_clip": 1.04902983,
      "balance_loss_mlp": 1.01508367,
      "epoch": 0.7535621956351831,
      "flos": 23878621192320.0,
      "grad_norm": 1.7953251826041832,
      "language_loss": 0.73136187,
      "learning_rate": 6.038007837479815e-07,
      "loss": 0.75322342,
      "num_input_tokens_seen": 134741450,
      "step": 6267,
      "time_per_iteration": 2.4239704608917236
    },
    {
      "auxiliary_loss_clip": 0.01150586,
      "auxiliary_loss_mlp": 0.01026156,
      "balance_loss_clip": 1.04704654,
      "balance_loss_mlp": 1.0188092,
      "epoch": 0.7536824385258222,
      "flos": 21795873960960.0,
      "grad_norm": 1.9848182025672334,
      "language_loss": 0.64212251,
      "learning_rate": 6.032431454539897e-07,
      "loss": 0.66389,
      "num_input_tokens_seen": 134760295,
      "step": 6268,
      "time_per_iteration": 2.4287822246551514
    },
    {
      "auxiliary_loss_clip": 0.01127872,
      "auxiliary_loss_mlp": 0.01028251,
      "balance_loss_clip": 1.04533052,
      "balance_loss_mlp": 1.02164364,
      "epoch": 0.7538026814164612,
      "flos": 28911644933760.0,
      "grad_norm": 1.8428271433944978,
      "language_loss": 0.81465751,
      "learning_rate": 6.026857190469014e-07,
      "loss": 0.83621871,
      "num_input_tokens_seen": 134782050,
      "step": 6269,
      "time_per_iteration": 2.5652706623077393
    },
    {
      "auxiliary_loss_clip": 0.01140396,
      "auxiliary_loss_mlp": 0.01021166,
      "balance_loss_clip": 1.04469585,
      "balance_loss_mlp": 1.01370335,
      "epoch": 0.7539229243071004,
      "flos": 21104701482240.0,
      "grad_norm": 6.019051247493286,
      "language_loss": 0.74187309,
      "learning_rate": 6.0212850461128e-07,
      "loss": 0.76348871,
      "num_input_tokens_seen": 134801170,
      "step": 6270,
      "time_per_iteration": 2.4639132022857666
    },
    {
      "auxiliary_loss_clip": 0.01142583,
      "auxiliary_loss_mlp": 0.01025749,
      "balance_loss_clip": 1.04521811,
      "balance_loss_mlp": 1.01805985,
      "epoch": 0.7540431671977395,
      "flos": 15158469340800.0,
      "grad_norm": 2.074958772768113,
      "language_loss": 0.74484587,
      "learning_rate": 6.015715022316516e-07,
      "loss": 0.76652914,
      "num_input_tokens_seen": 134819150,
      "step": 6271,
      "time_per_iteration": 2.4526193141937256
    },
    {
      "auxiliary_loss_clip": 0.01111515,
      "auxiliary_loss_mlp": 0.01021022,
      "balance_loss_clip": 1.04073691,
      "balance_loss_mlp": 1.01323223,
      "epoch": 0.7541634100883785,
      "flos": 18770579896320.0,
      "grad_norm": 2.8255669662731067,
      "language_loss": 0.78141332,
      "learning_rate": 6.010147119925154e-07,
      "loss": 0.80273867,
      "num_input_tokens_seen": 134836905,
      "step": 6272,
      "time_per_iteration": 2.5063862800598145
    },
    {
      "auxiliary_loss_clip": 0.01117236,
      "auxiliary_loss_mlp": 0.01023636,
      "balance_loss_clip": 1.04290366,
      "balance_loss_mlp": 1.01656091,
      "epoch": 0.7542836529790176,
      "flos": 20594770053120.0,
      "grad_norm": 1.975469233250893,
      "language_loss": 0.66183078,
      "learning_rate": 6.004581339783348e-07,
      "loss": 0.68323946,
      "num_input_tokens_seen": 134855225,
      "step": 6273,
      "time_per_iteration": 2.518101930618286
    },
    {
      "auxiliary_loss_clip": 0.01157985,
      "auxiliary_loss_mlp": 0.01032267,
      "balance_loss_clip": 1.04805088,
      "balance_loss_mlp": 1.02433991,
      "epoch": 0.7544038958696567,
      "flos": 19095104298240.0,
      "grad_norm": 2.6312171637024746,
      "language_loss": 0.68493694,
      "learning_rate": 5.999017682735425e-07,
      "loss": 0.70683944,
      "num_input_tokens_seen": 134871615,
      "step": 6274,
      "time_per_iteration": 2.5629022121429443
    },
    {
      "auxiliary_loss_clip": 0.01104552,
      "auxiliary_loss_mlp": 0.01027091,
      "balance_loss_clip": 1.04253376,
      "balance_loss_mlp": 1.01959562,
      "epoch": 0.7545241387602958,
      "flos": 31723306859520.0,
      "grad_norm": 2.002320912570464,
      "language_loss": 0.66431183,
      "learning_rate": 5.993456149625387e-07,
      "loss": 0.68562829,
      "num_input_tokens_seen": 134892765,
      "step": 6275,
      "time_per_iteration": 2.713529348373413
    },
    {
      "auxiliary_loss_clip": 0.0111629,
      "auxiliary_loss_mlp": 0.01024479,
      "balance_loss_clip": 1.04333472,
      "balance_loss_mlp": 1.01787746,
      "epoch": 0.7546443816509348,
      "flos": 20296495514880.0,
      "grad_norm": 1.7324732139310497,
      "language_loss": 0.82305419,
      "learning_rate": 5.987896741296909e-07,
      "loss": 0.8444618,
      "num_input_tokens_seen": 134910505,
      "step": 6276,
      "time_per_iteration": 2.501146078109741
    },
    {
      "auxiliary_loss_clip": 0.01141642,
      "auxiliary_loss_mlp": 0.01028032,
      "balance_loss_clip": 1.04825306,
      "balance_loss_mlp": 1.02077234,
      "epoch": 0.754764624541574,
      "flos": 23696159080320.0,
      "grad_norm": 2.4922362202068573,
      "language_loss": 0.783961,
      "learning_rate": 5.982339458593361e-07,
      "loss": 0.80565774,
      "num_input_tokens_seen": 134930445,
      "step": 6277,
      "time_per_iteration": 2.5138630867004395
    },
    {
      "auxiliary_loss_clip": 0.011516,
      "auxiliary_loss_mlp": 0.00761755,
      "balance_loss_clip": 1.04850817,
      "balance_loss_mlp": 1.00035167,
      "epoch": 0.7548848674322131,
      "flos": 25337204766720.0,
      "grad_norm": 1.5599870295067508,
      "language_loss": 0.83855003,
      "learning_rate": 5.976784302357767e-07,
      "loss": 0.8576836,
      "num_input_tokens_seen": 134951010,
      "step": 6278,
      "time_per_iteration": 2.4912402629852295
    },
    {
      "auxiliary_loss_clip": 0.01156583,
      "auxiliary_loss_mlp": 0.0102687,
      "balance_loss_clip": 1.04810667,
      "balance_loss_mlp": 1.01972926,
      "epoch": 0.7550051103228521,
      "flos": 19573147428480.0,
      "grad_norm": 1.8807749057620473,
      "language_loss": 0.73131055,
      "learning_rate": 5.971231273432855e-07,
      "loss": 0.7531451,
      "num_input_tokens_seen": 134970495,
      "step": 6279,
      "time_per_iteration": 2.426175355911255
    },
    {
      "auxiliary_loss_clip": 0.01060137,
      "auxiliary_loss_mlp": 0.0100264,
      "balance_loss_clip": 1.01441216,
      "balance_loss_mlp": 1.0015732,
      "epoch": 0.7551253532134913,
      "flos": 64150068648960.0,
      "grad_norm": 0.8104748717816136,
      "language_loss": 0.54582214,
      "learning_rate": 5.965680372661e-07,
      "loss": 0.56644988,
      "num_input_tokens_seen": 135028060,
      "step": 6280,
      "time_per_iteration": 2.947946310043335
    },
    {
      "auxiliary_loss_clip": 0.01140063,
      "auxiliary_loss_mlp": 0.01022911,
      "balance_loss_clip": 1.04693937,
      "balance_loss_mlp": 1.01661992,
      "epoch": 0.7552455961041303,
      "flos": 26067986968320.0,
      "grad_norm": 1.8685220353776606,
      "language_loss": 0.5656051,
      "learning_rate": 5.960131600884266e-07,
      "loss": 0.58723485,
      "num_input_tokens_seen": 135047330,
      "step": 6281,
      "time_per_iteration": 2.5105249881744385
    },
    {
      "auxiliary_loss_clip": 0.01129489,
      "auxiliary_loss_mlp": 0.01022468,
      "balance_loss_clip": 1.04580986,
      "balance_loss_mlp": 1.01599193,
      "epoch": 0.7553658389947694,
      "flos": 24498223822080.0,
      "grad_norm": 1.7390483990100314,
      "language_loss": 0.7589674,
      "learning_rate": 5.954584958944413e-07,
      "loss": 0.78048694,
      "num_input_tokens_seen": 135065995,
      "step": 6282,
      "time_per_iteration": 2.52677059173584
    },
    {
      "auxiliary_loss_clip": 0.01127613,
      "auxiliary_loss_mlp": 0.00761876,
      "balance_loss_clip": 1.04241645,
      "balance_loss_mlp": 1.00032043,
      "epoch": 0.7554860818854086,
      "flos": 21799465320960.0,
      "grad_norm": 1.8134519638142823,
      "language_loss": 0.81914151,
      "learning_rate": 5.949040447682854e-07,
      "loss": 0.83803642,
      "num_input_tokens_seen": 135085820,
      "step": 6283,
      "time_per_iteration": 3.2417562007904053
    },
    {
      "auxiliary_loss_clip": 0.0114508,
      "auxiliary_loss_mlp": 0.01023518,
      "balance_loss_clip": 1.0461179,
      "balance_loss_mlp": 1.01664615,
      "epoch": 0.7556063247760476,
      "flos": 16362123114240.0,
      "grad_norm": 3.0804674350465837,
      "language_loss": 0.68603671,
      "learning_rate": 5.943498067940686e-07,
      "loss": 0.70772272,
      "num_input_tokens_seen": 135102845,
      "step": 6284,
      "time_per_iteration": 3.283679723739624
    },
    {
      "auxiliary_loss_clip": 0.01140149,
      "auxiliary_loss_mlp": 0.01027324,
      "balance_loss_clip": 1.05275655,
      "balance_loss_mlp": 1.01999843,
      "epoch": 0.7557265676666867,
      "flos": 27235155502080.0,
      "grad_norm": 1.6548234805581539,
      "language_loss": 0.81631684,
      "learning_rate": 5.937957820558686e-07,
      "loss": 0.8379916,
      "num_input_tokens_seen": 135122190,
      "step": 6285,
      "time_per_iteration": 3.3709068298339844
    },
    {
      "auxiliary_loss_clip": 0.01049981,
      "auxiliary_loss_mlp": 0.01001505,
      "balance_loss_clip": 1.01219296,
      "balance_loss_mlp": 1.00046813,
      "epoch": 0.7558468105573258,
      "flos": 62189131415040.0,
      "grad_norm": 0.8473369133802555,
      "language_loss": 0.65465021,
      "learning_rate": 5.932419706377296e-07,
      "loss": 0.67516506,
      "num_input_tokens_seen": 135180495,
      "step": 6286,
      "time_per_iteration": 3.029397487640381
    },
    {
      "auxiliary_loss_clip": 0.01125351,
      "auxiliary_loss_mlp": 0.01022959,
      "balance_loss_clip": 1.04825294,
      "balance_loss_mlp": 1.01568711,
      "epoch": 0.7559670534479649,
      "flos": 33249078823680.0,
      "grad_norm": 1.8219643399996757,
      "language_loss": 0.74012923,
      "learning_rate": 5.92688372623666e-07,
      "loss": 0.76161242,
      "num_input_tokens_seen": 135199200,
      "step": 6287,
      "time_per_iteration": 2.5882673263549805
    },
    {
      "auxiliary_loss_clip": 0.01153921,
      "auxiliary_loss_mlp": 0.01022768,
      "balance_loss_clip": 1.04545832,
      "balance_loss_mlp": 1.01556754,
      "epoch": 0.7560872963386039,
      "flos": 14064379027200.0,
      "grad_norm": 2.0510412323897054,
      "language_loss": 0.73849952,
      "learning_rate": 5.921349880976574e-07,
      "loss": 0.76026642,
      "num_input_tokens_seen": 135217035,
      "step": 6288,
      "time_per_iteration": 3.205052137374878
    },
    {
      "auxiliary_loss_clip": 0.01142175,
      "auxiliary_loss_mlp": 0.00762325,
      "balance_loss_clip": 1.04425597,
      "balance_loss_mlp": 1.00038171,
      "epoch": 0.7562075392292431,
      "flos": 20412307941120.0,
      "grad_norm": 1.8696031497189385,
      "language_loss": 0.82365584,
      "learning_rate": 5.915818171436515e-07,
      "loss": 0.84270084,
      "num_input_tokens_seen": 135236370,
      "step": 6289,
      "time_per_iteration": 2.487180233001709
    },
    {
      "auxiliary_loss_clip": 0.01138726,
      "auxiliary_loss_mlp": 0.01025319,
      "balance_loss_clip": 1.04262042,
      "balance_loss_mlp": 1.01832187,
      "epoch": 0.7563277821198822,
      "flos": 20376792368640.0,
      "grad_norm": 2.262014802609539,
      "language_loss": 0.74437463,
      "learning_rate": 5.910288598455642e-07,
      "loss": 0.76601505,
      "num_input_tokens_seen": 135255720,
      "step": 6290,
      "time_per_iteration": 2.4932262897491455
    },
    {
      "auxiliary_loss_clip": 0.01159518,
      "auxiliary_loss_mlp": 0.01030382,
      "balance_loss_clip": 1.04856634,
      "balance_loss_mlp": 1.02306581,
      "epoch": 0.7564480250105212,
      "flos": 18588261438720.0,
      "grad_norm": 2.1522170690851654,
      "language_loss": 0.74586409,
      "learning_rate": 5.90476116287278e-07,
      "loss": 0.76776308,
      "num_input_tokens_seen": 135273320,
      "step": 6291,
      "time_per_iteration": 2.4486305713653564
    },
    {
      "auxiliary_loss_clip": 0.01140761,
      "auxiliary_loss_mlp": 0.0102789,
      "balance_loss_clip": 1.04886472,
      "balance_loss_mlp": 1.02071905,
      "epoch": 0.7565682679011604,
      "flos": 21215521918080.0,
      "grad_norm": 1.8188353913665978,
      "language_loss": 0.68033367,
      "learning_rate": 5.899235865526456e-07,
      "loss": 0.70202017,
      "num_input_tokens_seen": 135292615,
      "step": 6292,
      "time_per_iteration": 2.476522922515869
    },
    {
      "auxiliary_loss_clip": 0.01116689,
      "auxiliary_loss_mlp": 0.0102574,
      "balance_loss_clip": 1.04239488,
      "balance_loss_mlp": 1.01904321,
      "epoch": 0.7566885107917994,
      "flos": 20449008662400.0,
      "grad_norm": 1.7866296552006307,
      "language_loss": 0.82062238,
      "learning_rate": 5.893712707254825e-07,
      "loss": 0.84204674,
      "num_input_tokens_seen": 135310075,
      "step": 6293,
      "time_per_iteration": 2.4998016357421875
    },
    {
      "auxiliary_loss_clip": 0.01106037,
      "auxiliary_loss_mlp": 0.01023374,
      "balance_loss_clip": 1.04016626,
      "balance_loss_mlp": 1.01555347,
      "epoch": 0.7568087536824385,
      "flos": 19025832919680.0,
      "grad_norm": 2.5138350037075416,
      "language_loss": 0.65726644,
      "learning_rate": 5.888191688895769e-07,
      "loss": 0.67856061,
      "num_input_tokens_seen": 135327335,
      "step": 6294,
      "time_per_iteration": 2.5095326900482178
    },
    {
      "auxiliary_loss_clip": 0.01167241,
      "auxiliary_loss_mlp": 0.01027279,
      "balance_loss_clip": 1.04636717,
      "balance_loss_mlp": 1.0193038,
      "epoch": 0.7569289965730777,
      "flos": 15225442248960.0,
      "grad_norm": 2.2802058498177815,
      "language_loss": 0.61875337,
      "learning_rate": 5.882672811286813e-07,
      "loss": 0.64069855,
      "num_input_tokens_seen": 135343615,
      "step": 6295,
      "time_per_iteration": 2.3773233890533447
    },
    {
      "auxiliary_loss_clip": 0.01169594,
      "auxiliary_loss_mlp": 0.01026128,
      "balance_loss_clip": 1.04825592,
      "balance_loss_mlp": 1.01882625,
      "epoch": 0.7570492394637167,
      "flos": 20769367086720.0,
      "grad_norm": 2.0122279612698626,
      "language_loss": 0.69431567,
      "learning_rate": 5.877156075265166e-07,
      "loss": 0.71627289,
      "num_input_tokens_seen": 135359880,
      "step": 6296,
      "time_per_iteration": 2.3994789123535156
    },
    {
      "auxiliary_loss_clip": 0.01139063,
      "auxiliary_loss_mlp": 0.01023036,
      "balance_loss_clip": 1.04427767,
      "balance_loss_mlp": 1.01538897,
      "epoch": 0.7571694823543558,
      "flos": 15664091137920.0,
      "grad_norm": 2.568869135287148,
      "language_loss": 0.69711959,
      "learning_rate": 5.871641481667715e-07,
      "loss": 0.71874058,
      "num_input_tokens_seen": 135374325,
      "step": 6297,
      "time_per_iteration": 2.4964351654052734
    },
    {
      "auxiliary_loss_clip": 0.01114522,
      "auxiliary_loss_mlp": 0.01030237,
      "balance_loss_clip": 1.04375172,
      "balance_loss_mlp": 1.02252758,
      "epoch": 0.7572897252449949,
      "flos": 25409241492480.0,
      "grad_norm": 1.8676425103516465,
      "language_loss": 0.84341598,
      "learning_rate": 5.866129031331011e-07,
      "loss": 0.86486351,
      "num_input_tokens_seen": 135393980,
      "step": 6298,
      "time_per_iteration": 2.567317008972168
    },
    {
      "auxiliary_loss_clip": 0.01141324,
      "auxiliary_loss_mlp": 0.01025239,
      "balance_loss_clip": 1.04521346,
      "balance_loss_mlp": 1.01779151,
      "epoch": 0.757409968135634,
      "flos": 24279348297600.0,
      "grad_norm": 1.9529690391343197,
      "language_loss": 0.83697963,
      "learning_rate": 5.8606187250913e-07,
      "loss": 0.85864532,
      "num_input_tokens_seen": 135412030,
      "step": 6299,
      "time_per_iteration": 2.510033130645752
    },
    {
      "auxiliary_loss_clip": 0.01154667,
      "auxiliary_loss_mlp": 0.00761668,
      "balance_loss_clip": 1.04941511,
      "balance_loss_mlp": 1.0003686,
      "epoch": 0.757530211026273,
      "flos": 24133766474880.0,
      "grad_norm": 2.0252841485884705,
      "language_loss": 0.84210235,
      "learning_rate": 5.855110563784482e-07,
      "loss": 0.86126566,
      "num_input_tokens_seen": 135430565,
      "step": 6300,
      "time_per_iteration": 2.453111171722412
    },
    {
      "auxiliary_loss_clip": 0.01147931,
      "auxiliary_loss_mlp": 0.00761874,
      "balance_loss_clip": 1.04491329,
      "balance_loss_mlp": 1.00032973,
      "epoch": 0.7576504539169122,
      "flos": 23951807153280.0,
      "grad_norm": 2.119686534995756,
      "language_loss": 0.64331168,
      "learning_rate": 5.849604548246156e-07,
      "loss": 0.66240972,
      "num_input_tokens_seen": 135451675,
      "step": 6301,
      "time_per_iteration": 2.470508337020874
    },
    {
      "auxiliary_loss_clip": 0.01145463,
      "auxiliary_loss_mlp": 0.00761639,
      "balance_loss_clip": 1.04802024,
      "balance_loss_mlp": 1.00039577,
      "epoch": 0.7577706968075513,
      "flos": 21251360712960.0,
      "grad_norm": 2.0044358807027254,
      "language_loss": 0.80423015,
      "learning_rate": 5.844100679311565e-07,
      "loss": 0.8233012,
      "num_input_tokens_seen": 135470635,
      "step": 6302,
      "time_per_iteration": 2.495260238647461
    },
    {
      "auxiliary_loss_clip": 0.01143786,
      "auxiliary_loss_mlp": 0.01022826,
      "balance_loss_clip": 1.04951477,
      "balance_loss_mlp": 1.01506507,
      "epoch": 0.7578909396981903,
      "flos": 18296595002880.0,
      "grad_norm": 2.768617825031151,
      "language_loss": 0.76597059,
      "learning_rate": 5.838598957815637e-07,
      "loss": 0.78763676,
      "num_input_tokens_seen": 135487865,
      "step": 6303,
      "time_per_iteration": 2.4927802085876465
    },
    {
      "auxiliary_loss_clip": 0.01132554,
      "auxiliary_loss_mlp": 0.01023391,
      "balance_loss_clip": 1.04364729,
      "balance_loss_mlp": 1.01637888,
      "epoch": 0.7580111825888295,
      "flos": 25373869574400.0,
      "grad_norm": 1.5876508215880005,
      "language_loss": 0.85302413,
      "learning_rate": 5.833099384592996e-07,
      "loss": 0.8745836,
      "num_input_tokens_seen": 135508440,
      "step": 6304,
      "time_per_iteration": 2.5008881092071533
    },
    {
      "auxiliary_loss_clip": 0.01134277,
      "auxiliary_loss_mlp": 0.0102561,
      "balance_loss_clip": 1.04370856,
      "balance_loss_mlp": 1.01822841,
      "epoch": 0.7581314254794685,
      "flos": 23768662682880.0,
      "grad_norm": 2.2896850810795932,
      "language_loss": 0.7142691,
      "learning_rate": 5.827601960477913e-07,
      "loss": 0.73586798,
      "num_input_tokens_seen": 135526365,
      "step": 6305,
      "time_per_iteration": 2.475066900253296
    },
    {
      "auxiliary_loss_clip": 0.01148639,
      "auxiliary_loss_mlp": 0.01026794,
      "balance_loss_clip": 1.04398727,
      "balance_loss_mlp": 1.02036834,
      "epoch": 0.7582516683701076,
      "flos": 22054610603520.0,
      "grad_norm": 2.4584644419153996,
      "language_loss": 0.70667589,
      "learning_rate": 5.822106686304344e-07,
      "loss": 0.72843021,
      "num_input_tokens_seen": 135545655,
      "step": 6306,
      "time_per_iteration": 2.4346423149108887
    },
    {
      "auxiliary_loss_clip": 0.01133031,
      "auxiliary_loss_mlp": 0.01027149,
      "balance_loss_clip": 1.04505849,
      "balance_loss_mlp": 1.01981211,
      "epoch": 0.7583719112607467,
      "flos": 31649725848960.0,
      "grad_norm": 1.6930919223942935,
      "language_loss": 0.57783252,
      "learning_rate": 5.816613562905919e-07,
      "loss": 0.59943438,
      "num_input_tokens_seen": 135566840,
      "step": 6307,
      "time_per_iteration": 2.568549156188965
    },
    {
      "auxiliary_loss_clip": 0.01125657,
      "auxiliary_loss_mlp": 0.01025938,
      "balance_loss_clip": 1.04938924,
      "balance_loss_mlp": 1.0185889,
      "epoch": 0.7584921541513858,
      "flos": 33068376478080.0,
      "grad_norm": 1.5333518695777595,
      "language_loss": 0.70068789,
      "learning_rate": 5.811122591115933e-07,
      "loss": 0.72220385,
      "num_input_tokens_seen": 135587825,
      "step": 6308,
      "time_per_iteration": 2.584228038787842
    },
    {
      "auxiliary_loss_clip": 0.01128449,
      "auxiliary_loss_mlp": 0.010271,
      "balance_loss_clip": 1.05008614,
      "balance_loss_mlp": 1.01992035,
      "epoch": 0.7586123970420249,
      "flos": 23326350606720.0,
      "grad_norm": 2.238996947319549,
      "language_loss": 0.71755725,
      "learning_rate": 5.805633771767376e-07,
      "loss": 0.73911279,
      "num_input_tokens_seen": 135605220,
      "step": 6309,
      "time_per_iteration": 2.510951042175293
    },
    {
      "auxiliary_loss_clip": 0.0113747,
      "auxiliary_loss_mlp": 0.01023247,
      "balance_loss_clip": 1.04668093,
      "balance_loss_mlp": 1.01530159,
      "epoch": 0.7587326399326639,
      "flos": 18334229477760.0,
      "grad_norm": 1.6933524051913462,
      "language_loss": 0.77881581,
      "learning_rate": 5.800147105692888e-07,
      "loss": 0.80042303,
      "num_input_tokens_seen": 135624795,
      "step": 6310,
      "time_per_iteration": 3.3414499759674072
    },
    {
      "auxiliary_loss_clip": 0.01153978,
      "auxiliary_loss_mlp": 0.0102343,
      "balance_loss_clip": 1.04512882,
      "balance_loss_mlp": 1.01655173,
      "epoch": 0.7588528828233031,
      "flos": 17275080119040.0,
      "grad_norm": 1.9194437294654993,
      "language_loss": 0.79095036,
      "learning_rate": 5.794662593724795e-07,
      "loss": 0.81272447,
      "num_input_tokens_seen": 135643800,
      "step": 6311,
      "time_per_iteration": 3.282911777496338
    },
    {
      "auxiliary_loss_clip": 0.01171036,
      "auxiliary_loss_mlp": 0.01029126,
      "balance_loss_clip": 1.05131805,
      "balance_loss_mlp": 1.02155638,
      "epoch": 0.7589731257139422,
      "flos": 17713621267200.0,
      "grad_norm": 1.8620825579727243,
      "language_loss": 0.74696571,
      "learning_rate": 5.789180236695091e-07,
      "loss": 0.76896739,
      "num_input_tokens_seen": 135660655,
      "step": 6312,
      "time_per_iteration": 2.3927524089813232
    },
    {
      "auxiliary_loss_clip": 0.01148521,
      "auxiliary_loss_mlp": 0.01026573,
      "balance_loss_clip": 1.04728293,
      "balance_loss_mlp": 1.02004337,
      "epoch": 0.7590933686045812,
      "flos": 15961072786560.0,
      "grad_norm": 1.9442211615831841,
      "language_loss": 0.85097373,
      "learning_rate": 5.78370003543544e-07,
      "loss": 0.87272465,
      "num_input_tokens_seen": 135679410,
      "step": 6313,
      "time_per_iteration": 2.412609338760376
    },
    {
      "auxiliary_loss_clip": 0.01154447,
      "auxiliary_loss_mlp": 0.00761981,
      "balance_loss_clip": 1.04832172,
      "balance_loss_mlp": 1.00038409,
      "epoch": 0.7592136114952204,
      "flos": 21068072588160.0,
      "grad_norm": 2.083486064330563,
      "language_loss": 0.8371017,
      "learning_rate": 5.778221990777203e-07,
      "loss": 0.85626602,
      "num_input_tokens_seen": 135697150,
      "step": 6314,
      "time_per_iteration": 2.4422402381896973
    },
    {
      "auxiliary_loss_clip": 0.0114386,
      "auxiliary_loss_mlp": 0.01026918,
      "balance_loss_clip": 1.04937983,
      "balance_loss_mlp": 1.01949739,
      "epoch": 0.7593338543858594,
      "flos": 25297666871040.0,
      "grad_norm": 2.5559611022323057,
      "language_loss": 0.82623255,
      "learning_rate": 5.772746103551372e-07,
      "loss": 0.84794039,
      "num_input_tokens_seen": 135712545,
      "step": 6315,
      "time_per_iteration": 3.2547085285186768
    },
    {
      "auxiliary_loss_clip": 0.01138076,
      "auxiliary_loss_mlp": 0.01021115,
      "balance_loss_clip": 1.04687536,
      "balance_loss_mlp": 1.01356304,
      "epoch": 0.7594540972764985,
      "flos": 31832367528960.0,
      "grad_norm": 1.7597377282269855,
      "language_loss": 0.72190166,
      "learning_rate": 5.767272374588648e-07,
      "loss": 0.74349356,
      "num_input_tokens_seen": 135733950,
      "step": 6316,
      "time_per_iteration": 2.556236743927002
    },
    {
      "auxiliary_loss_clip": 0.0115302,
      "auxiliary_loss_mlp": 0.01024452,
      "balance_loss_clip": 1.04926753,
      "balance_loss_mlp": 1.01697111,
      "epoch": 0.7595743401671377,
      "flos": 37597250880000.0,
      "grad_norm": 1.6275790280984483,
      "language_loss": 0.77904081,
      "learning_rate": 5.76180080471939e-07,
      "loss": 0.80081546,
      "num_input_tokens_seen": 135757120,
      "step": 6317,
      "time_per_iteration": 2.581899404525757
    },
    {
      "auxiliary_loss_clip": 0.01172257,
      "auxiliary_loss_mlp": 0.0102548,
      "balance_loss_clip": 1.04931593,
      "balance_loss_mlp": 1.01769018,
      "epoch": 0.7596945830577767,
      "flos": 18287724343680.0,
      "grad_norm": 2.3262235525151467,
      "language_loss": 0.72180593,
      "learning_rate": 5.756331394773631e-07,
      "loss": 0.74378324,
      "num_input_tokens_seen": 135773335,
      "step": 6318,
      "time_per_iteration": 2.3827598094940186
    },
    {
      "auxiliary_loss_clip": 0.01099971,
      "auxiliary_loss_mlp": 0.00762549,
      "balance_loss_clip": 1.04296005,
      "balance_loss_mlp": 1.00030613,
      "epoch": 0.7598148259484158,
      "flos": 22233122219520.0,
      "grad_norm": 1.696122889528536,
      "language_loss": 0.76261896,
      "learning_rate": 5.750864145581071e-07,
      "loss": 0.78124416,
      "num_input_tokens_seen": 135792555,
      "step": 6319,
      "time_per_iteration": 2.5931317806243896
    },
    {
      "auxiliary_loss_clip": 0.01168761,
      "auxiliary_loss_mlp": 0.0102676,
      "balance_loss_clip": 1.0505681,
      "balance_loss_mlp": 1.01969683,
      "epoch": 0.7599350688390549,
      "flos": 27161718145920.0,
      "grad_norm": 1.9137831335682258,
      "language_loss": 0.86044788,
      "learning_rate": 5.745399057971085e-07,
      "loss": 0.88240302,
      "num_input_tokens_seen": 135813690,
      "step": 6320,
      "time_per_iteration": 2.442300796508789
    },
    {
      "auxiliary_loss_clip": 0.01158099,
      "auxiliary_loss_mlp": 0.01026452,
      "balance_loss_clip": 1.0477246,
      "balance_loss_mlp": 1.01948154,
      "epoch": 0.760055311729694,
      "flos": 15560704817280.0,
      "grad_norm": 2.6722157325804483,
      "language_loss": 0.75298023,
      "learning_rate": 5.739936132772738e-07,
      "loss": 0.77482575,
      "num_input_tokens_seen": 135832255,
      "step": 6321,
      "time_per_iteration": 2.4137682914733887
    },
    {
      "auxiliary_loss_clip": 0.01166094,
      "auxiliary_loss_mlp": 0.01025137,
      "balance_loss_clip": 1.04670417,
      "balance_loss_mlp": 1.01763618,
      "epoch": 0.760175554620333,
      "flos": 25155496840320.0,
      "grad_norm": 1.9331697096220553,
      "language_loss": 0.74442792,
      "learning_rate": 5.734475370814733e-07,
      "loss": 0.76634014,
      "num_input_tokens_seen": 135851935,
      "step": 6322,
      "time_per_iteration": 2.423246145248413
    },
    {
      "auxiliary_loss_clip": 0.01154586,
      "auxiliary_loss_mlp": 0.01022408,
      "balance_loss_clip": 1.04521108,
      "balance_loss_mlp": 1.01519561,
      "epoch": 0.7602957975109722,
      "flos": 24353791234560.0,
      "grad_norm": 1.6153738544491134,
      "language_loss": 0.78686702,
      "learning_rate": 5.729016772925483e-07,
      "loss": 0.80863696,
      "num_input_tokens_seen": 135873510,
      "step": 6323,
      "time_per_iteration": 2.457078456878662
    },
    {
      "auxiliary_loss_clip": 0.01110636,
      "auxiliary_loss_mlp": 0.01023412,
      "balance_loss_clip": 1.04593015,
      "balance_loss_mlp": 1.01565695,
      "epoch": 0.7604160404016113,
      "flos": 25192664438400.0,
      "grad_norm": 1.9692730370201807,
      "language_loss": 0.70632821,
      "learning_rate": 5.723560339933038e-07,
      "loss": 0.7276687,
      "num_input_tokens_seen": 135893845,
      "step": 6324,
      "time_per_iteration": 2.5802526473999023
    },
    {
      "auxiliary_loss_clip": 0.01151902,
      "auxiliary_loss_mlp": 0.00761791,
      "balance_loss_clip": 1.04662216,
      "balance_loss_mlp": 1.00032759,
      "epoch": 0.7605362832922503,
      "flos": 29861841363840.0,
      "grad_norm": 2.3544129384779113,
      "language_loss": 0.65566218,
      "learning_rate": 5.71810607266513e-07,
      "loss": 0.67479908,
      "num_input_tokens_seen": 135912430,
      "step": 6325,
      "time_per_iteration": 2.4846885204315186
    },
    {
      "auxiliary_loss_clip": 0.01153364,
      "auxiliary_loss_mlp": 0.01024451,
      "balance_loss_clip": 1.04577374,
      "balance_loss_mlp": 1.01741421,
      "epoch": 0.7606565261828895,
      "flos": 13917935278080.0,
      "grad_norm": 2.1290613247396397,
      "language_loss": 0.60401756,
      "learning_rate": 5.712653971949184e-07,
      "loss": 0.62579566,
      "num_input_tokens_seen": 135930550,
      "step": 6326,
      "time_per_iteration": 2.4105725288391113
    },
    {
      "auxiliary_loss_clip": 0.01148444,
      "auxiliary_loss_mlp": 0.01022567,
      "balance_loss_clip": 1.04608309,
      "balance_loss_mlp": 1.01508641,
      "epoch": 0.7607767690735285,
      "flos": 18551273408640.0,
      "grad_norm": 2.197977500654092,
      "language_loss": 0.7530911,
      "learning_rate": 5.707204038612268e-07,
      "loss": 0.77480125,
      "num_input_tokens_seen": 135947980,
      "step": 6327,
      "time_per_iteration": 2.3939802646636963
    },
    {
      "auxiliary_loss_clip": 0.01151838,
      "auxiliary_loss_mlp": 0.01027807,
      "balance_loss_clip": 1.05369282,
      "balance_loss_mlp": 1.02002311,
      "epoch": 0.7608970119641676,
      "flos": 20922993555840.0,
      "grad_norm": 4.1618849003005325,
      "language_loss": 0.74070084,
      "learning_rate": 5.701756273481138e-07,
      "loss": 0.76249731,
      "num_input_tokens_seen": 135965400,
      "step": 6328,
      "time_per_iteration": 2.4617631435394287
    },
    {
      "auxiliary_loss_clip": 0.01143758,
      "auxiliary_loss_mlp": 0.01023547,
      "balance_loss_clip": 1.04528272,
      "balance_loss_mlp": 1.01641846,
      "epoch": 0.7610172548548068,
      "flos": 23807302738560.0,
      "grad_norm": 1.4458777093984885,
      "language_loss": 0.73968965,
      "learning_rate": 5.696310677382212e-07,
      "loss": 0.76136267,
      "num_input_tokens_seen": 135986795,
      "step": 6329,
      "time_per_iteration": 2.492581367492676
    },
    {
      "auxiliary_loss_clip": 0.01030973,
      "auxiliary_loss_mlp": 0.01001464,
      "balance_loss_clip": 1.01165676,
      "balance_loss_mlp": 1.00058734,
      "epoch": 0.7611374977454458,
      "flos": 66496580426880.0,
      "grad_norm": 0.8720645697198282,
      "language_loss": 0.61830842,
      "learning_rate": 5.690867251141576e-07,
      "loss": 0.63863277,
      "num_input_tokens_seen": 136053450,
      "step": 6330,
      "time_per_iteration": 3.2151288986206055
    },
    {
      "auxiliary_loss_clip": 0.0116113,
      "auxiliary_loss_mlp": 0.01025574,
      "balance_loss_clip": 1.04748785,
      "balance_loss_mlp": 1.01826048,
      "epoch": 0.7612577406360849,
      "flos": 15633136592640.0,
      "grad_norm": 2.993612463154022,
      "language_loss": 0.9221698,
      "learning_rate": 5.685425995585013e-07,
      "loss": 0.94403684,
      "num_input_tokens_seen": 136071375,
      "step": 6331,
      "time_per_iteration": 2.42741322517395
    },
    {
      "auxiliary_loss_clip": 0.01048877,
      "auxiliary_loss_mlp": 0.01004664,
      "balance_loss_clip": 1.01247931,
      "balance_loss_mlp": 1.0036273,
      "epoch": 0.761377983526724,
      "flos": 60526253237760.0,
      "grad_norm": 3.8826306507364126,
      "language_loss": 0.59045696,
      "learning_rate": 5.679986911537935e-07,
      "loss": 0.61099243,
      "num_input_tokens_seen": 136138905,
      "step": 6332,
      "time_per_iteration": 3.1932616233825684
    },
    {
      "auxiliary_loss_clip": 0.0110315,
      "auxiliary_loss_mlp": 0.01021479,
      "balance_loss_clip": 1.04544961,
      "balance_loss_mlp": 1.01418352,
      "epoch": 0.7614982264173631,
      "flos": 35772522019200.0,
      "grad_norm": 2.1024744526939783,
      "language_loss": 0.67546231,
      "learning_rate": 5.674549999825462e-07,
      "loss": 0.69670856,
      "num_input_tokens_seen": 136161720,
      "step": 6333,
      "time_per_iteration": 2.655038833618164
    },
    {
      "auxiliary_loss_clip": 0.01058547,
      "auxiliary_loss_mlp": 0.010017,
      "balance_loss_clip": 1.01239836,
      "balance_loss_mlp": 1.00067508,
      "epoch": 0.7616184693080021,
      "flos": 67925502345600.0,
      "grad_norm": 0.9183383795375871,
      "language_loss": 0.71333563,
      "learning_rate": 5.669115261272363e-07,
      "loss": 0.7339381,
      "num_input_tokens_seen": 136222040,
      "step": 6334,
      "time_per_iteration": 3.023500442504883
    },
    {
      "auxiliary_loss_clip": 0.01155372,
      "auxiliary_loss_mlp": 0.010263,
      "balance_loss_clip": 1.04781616,
      "balance_loss_mlp": 1.01885581,
      "epoch": 0.7617387121986413,
      "flos": 20521979141760.0,
      "grad_norm": 2.3209834273778966,
      "language_loss": 0.72619081,
      "learning_rate": 5.663682696703081e-07,
      "loss": 0.74800754,
      "num_input_tokens_seen": 136240305,
      "step": 6335,
      "time_per_iteration": 2.424441337585449
    },
    {
      "auxiliary_loss_clip": 0.01166859,
      "auxiliary_loss_mlp": 0.01022455,
      "balance_loss_clip": 1.04935789,
      "balance_loss_mlp": 1.01562095,
      "epoch": 0.7618589550892804,
      "flos": 18624495283200.0,
      "grad_norm": 2.13907609455361,
      "language_loss": 0.82004869,
      "learning_rate": 5.658252306941746e-07,
      "loss": 0.84194183,
      "num_input_tokens_seen": 136259625,
      "step": 6336,
      "time_per_iteration": 2.392655372619629
    },
    {
      "auxiliary_loss_clip": 0.01113659,
      "auxiliary_loss_mlp": 0.0102801,
      "balance_loss_clip": 1.04392028,
      "balance_loss_mlp": 1.02003527,
      "epoch": 0.7619791979799194,
      "flos": 17453735389440.0,
      "grad_norm": 2.0472294080051188,
      "language_loss": 0.75158238,
      "learning_rate": 5.65282409281212e-07,
      "loss": 0.77299917,
      "num_input_tokens_seen": 136277090,
      "step": 6337,
      "time_per_iteration": 4.117333173751831
    },
    {
      "auxiliary_loss_clip": 0.01136397,
      "auxiliary_loss_mlp": 0.01024597,
      "balance_loss_clip": 1.0446856,
      "balance_loss_mlp": 1.01742387,
      "epoch": 0.7620994408705585,
      "flos": 14137421333760.0,
      "grad_norm": 2.817679875138117,
      "language_loss": 0.70092213,
      "learning_rate": 5.64739805513768e-07,
      "loss": 0.72253203,
      "num_input_tokens_seen": 136294635,
      "step": 6338,
      "time_per_iteration": 3.2620275020599365
    },
    {
      "auxiliary_loss_clip": 0.01053926,
      "auxiliary_loss_mlp": 0.00753067,
      "balance_loss_clip": 1.01274848,
      "balance_loss_mlp": 1.00004756,
      "epoch": 0.7622196837611976,
      "flos": 70708792527360.0,
      "grad_norm": 0.7868444945715333,
      "language_loss": 0.55714929,
      "learning_rate": 5.641974194741541e-07,
      "loss": 0.57521927,
      "num_input_tokens_seen": 136350320,
      "step": 6339,
      "time_per_iteration": 2.9368066787719727
    },
    {
      "auxiliary_loss_clip": 0.0104348,
      "auxiliary_loss_mlp": 0.01002475,
      "balance_loss_clip": 1.02246153,
      "balance_loss_mlp": 1.00114536,
      "epoch": 0.7623399266518367,
      "flos": 60684150447360.0,
      "grad_norm": 0.7965721282224352,
      "language_loss": 0.63755411,
      "learning_rate": 5.636552512446502e-07,
      "loss": 0.65801364,
      "num_input_tokens_seen": 136411375,
      "step": 6340,
      "time_per_iteration": 2.9715940952301025
    },
    {
      "auxiliary_loss_clip": 0.01147749,
      "auxiliary_loss_mlp": 0.0102425,
      "balance_loss_clip": 1.04569745,
      "balance_loss_mlp": 1.01706481,
      "epoch": 0.7624601695424758,
      "flos": 26468893641600.0,
      "grad_norm": 1.9332538891919064,
      "language_loss": 0.77745032,
      "learning_rate": 5.631133009075027e-07,
      "loss": 0.79917026,
      "num_input_tokens_seen": 136430560,
      "step": 6341,
      "time_per_iteration": 2.4662132263183594
    },
    {
      "auxiliary_loss_clip": 0.01154421,
      "auxiliary_loss_mlp": 0.00761352,
      "balance_loss_clip": 1.04762554,
      "balance_loss_mlp": 1.00031853,
      "epoch": 0.7625804124331149,
      "flos": 19135755515520.0,
      "grad_norm": 1.7510602788399283,
      "language_loss": 0.68527842,
      "learning_rate": 5.625715685449242e-07,
      "loss": 0.70443618,
      "num_input_tokens_seen": 136448665,
      "step": 6342,
      "time_per_iteration": 3.2139532566070557
    },
    {
      "auxiliary_loss_clip": 0.01128327,
      "auxiliary_loss_mlp": 0.01025456,
      "balance_loss_clip": 1.05097675,
      "balance_loss_mlp": 1.01882803,
      "epoch": 0.762700655323754,
      "flos": 26213101914240.0,
      "grad_norm": 1.629358366918023,
      "language_loss": 0.71679193,
      "learning_rate": 5.620300542390966e-07,
      "loss": 0.73832971,
      "num_input_tokens_seen": 136469710,
      "step": 6343,
      "time_per_iteration": 2.5413007736206055
    },
    {
      "auxiliary_loss_clip": 0.01135849,
      "auxiliary_loss_mlp": 0.01024264,
      "balance_loss_clip": 1.0427084,
      "balance_loss_mlp": 1.01771009,
      "epoch": 0.762820898214393,
      "flos": 22382582711040.0,
      "grad_norm": 1.9192033933468766,
      "language_loss": 0.849204,
      "learning_rate": 5.614887580721659e-07,
      "loss": 0.87080508,
      "num_input_tokens_seen": 136489855,
      "step": 6344,
      "time_per_iteration": 2.4717347621917725
    },
    {
      "auxiliary_loss_clip": 0.01117947,
      "auxiliary_loss_mlp": 0.0103225,
      "balance_loss_clip": 1.04618454,
      "balance_loss_mlp": 1.02439153,
      "epoch": 0.7629411411050322,
      "flos": 15700504550400.0,
      "grad_norm": 1.9564479352594817,
      "language_loss": 0.73763824,
      "learning_rate": 5.609476801262481e-07,
      "loss": 0.75914013,
      "num_input_tokens_seen": 136504715,
      "step": 6345,
      "time_per_iteration": 2.42869234085083
    },
    {
      "auxiliary_loss_clip": 0.01126714,
      "auxiliary_loss_mlp": 0.01027097,
      "balance_loss_clip": 1.04847682,
      "balance_loss_mlp": 1.01982534,
      "epoch": 0.7630613839956712,
      "flos": 13770342293760.0,
      "grad_norm": 2.292100234323005,
      "language_loss": 0.63951814,
      "learning_rate": 5.604068204834223e-07,
      "loss": 0.66105622,
      "num_input_tokens_seen": 136521610,
      "step": 6346,
      "time_per_iteration": 2.4694647789001465
    },
    {
      "auxiliary_loss_clip": 0.01111715,
      "auxiliary_loss_mlp": 0.00762354,
      "balance_loss_clip": 1.04542756,
      "balance_loss_mlp": 1.000283,
      "epoch": 0.7631816268863103,
      "flos": 14569569861120.0,
      "grad_norm": 2.215049450923147,
      "language_loss": 0.76720798,
      "learning_rate": 5.598661792257367e-07,
      "loss": 0.78594863,
      "num_input_tokens_seen": 136538655,
      "step": 6347,
      "time_per_iteration": 2.509542942047119
    },
    {
      "auxiliary_loss_clip": 0.01150963,
      "auxiliary_loss_mlp": 0.01024132,
      "balance_loss_clip": 1.04513907,
      "balance_loss_mlp": 1.01718175,
      "epoch": 0.7633018697769495,
      "flos": 19062210418560.0,
      "grad_norm": 1.9012677857481546,
      "language_loss": 0.75694811,
      "learning_rate": 5.593257564352071e-07,
      "loss": 0.77869904,
      "num_input_tokens_seen": 136557095,
      "step": 6348,
      "time_per_iteration": 2.4221768379211426
    },
    {
      "auxiliary_loss_clip": 0.01151656,
      "auxiliary_loss_mlp": 0.0102128,
      "balance_loss_clip": 1.047261,
      "balance_loss_mlp": 1.01425278,
      "epoch": 0.7634221126675885,
      "flos": 22052958577920.0,
      "grad_norm": 1.546631299463628,
      "language_loss": 0.75615633,
      "learning_rate": 5.58785552193815e-07,
      "loss": 0.77788568,
      "num_input_tokens_seen": 136577340,
      "step": 6349,
      "time_per_iteration": 2.4384586811065674
    },
    {
      "auxiliary_loss_clip": 0.01168528,
      "auxiliary_loss_mlp": 0.01022205,
      "balance_loss_clip": 1.04902315,
      "balance_loss_mlp": 1.01530015,
      "epoch": 0.7635423555582276,
      "flos": 29382720825600.0,
      "grad_norm": 2.1326665327485106,
      "language_loss": 0.75637043,
      "learning_rate": 5.582455665835086e-07,
      "loss": 0.77827775,
      "num_input_tokens_seen": 136597635,
      "step": 6350,
      "time_per_iteration": 2.4667088985443115
    },
    {
      "auxiliary_loss_clip": 0.01150137,
      "auxiliary_loss_mlp": 0.01030328,
      "balance_loss_clip": 1.04547262,
      "balance_loss_mlp": 1.02212048,
      "epoch": 0.7636625984488667,
      "flos": 17784903807360.0,
      "grad_norm": 2.596099468780283,
      "language_loss": 0.72555166,
      "learning_rate": 5.577057996862036e-07,
      "loss": 0.7473563,
      "num_input_tokens_seen": 136615260,
      "step": 6351,
      "time_per_iteration": 2.4519476890563965
    },
    {
      "auxiliary_loss_clip": 0.01163613,
      "auxiliary_loss_mlp": 0.01024034,
      "balance_loss_clip": 1.04783523,
      "balance_loss_mlp": 1.01721549,
      "epoch": 0.7637828413395058,
      "flos": 23734583654400.0,
      "grad_norm": 1.6057167603519746,
      "language_loss": 0.76335788,
      "learning_rate": 5.571662515837814e-07,
      "loss": 0.78523433,
      "num_input_tokens_seen": 136637220,
      "step": 6352,
      "time_per_iteration": 2.4916059970855713
    },
    {
      "auxiliary_loss_clip": 0.01138141,
      "auxiliary_loss_mlp": 0.01024031,
      "balance_loss_clip": 1.04600549,
      "balance_loss_mlp": 1.01711392,
      "epoch": 0.7639030842301449,
      "flos": 36283279461120.0,
      "grad_norm": 1.873407667167574,
      "language_loss": 0.83926988,
      "learning_rate": 5.566269223580926e-07,
      "loss": 0.86089158,
      "num_input_tokens_seen": 136658930,
      "step": 6353,
      "time_per_iteration": 2.601808786392212
    },
    {
      "auxiliary_loss_clip": 0.01156984,
      "auxiliary_loss_mlp": 0.01023895,
      "balance_loss_clip": 1.04849374,
      "balance_loss_mlp": 1.01662576,
      "epoch": 0.764023327120784,
      "flos": 28878104609280.0,
      "grad_norm": 1.8511146631553665,
      "language_loss": 0.75109136,
      "learning_rate": 5.560878120909511e-07,
      "loss": 0.77290016,
      "num_input_tokens_seen": 136681530,
      "step": 6354,
      "time_per_iteration": 2.5004525184631348
    },
    {
      "auxiliary_loss_clip": 0.01059698,
      "auxiliary_loss_mlp": 0.01002478,
      "balance_loss_clip": 1.01318908,
      "balance_loss_mlp": 1.0014708,
      "epoch": 0.7641435700114231,
      "flos": 64789711067520.0,
      "grad_norm": 0.8441767138699058,
      "language_loss": 0.58611226,
      "learning_rate": 5.55548920864141e-07,
      "loss": 0.60673404,
      "num_input_tokens_seen": 136742185,
      "step": 6355,
      "time_per_iteration": 3.04837703704834
    },
    {
      "auxiliary_loss_clip": 0.01155418,
      "auxiliary_loss_mlp": 0.01020966,
      "balance_loss_clip": 1.05129814,
      "balance_loss_mlp": 1.01443887,
      "epoch": 0.7642638129020621,
      "flos": 16835784785280.0,
      "grad_norm": 1.8017246285669568,
      "language_loss": 0.77789485,
      "learning_rate": 5.550102487594113e-07,
      "loss": 0.79965872,
      "num_input_tokens_seen": 136760855,
      "step": 6356,
      "time_per_iteration": 2.4332199096679688
    },
    {
      "auxiliary_loss_clip": 0.01114587,
      "auxiliary_loss_mlp": 0.00761259,
      "balance_loss_clip": 1.04167366,
      "balance_loss_mlp": 1.0003233,
      "epoch": 0.7643840557927013,
      "flos": 30408940391040.0,
      "grad_norm": 1.8794261144617133,
      "language_loss": 0.71704465,
      "learning_rate": 5.54471795858477e-07,
      "loss": 0.73580307,
      "num_input_tokens_seen": 136780925,
      "step": 6357,
      "time_per_iteration": 2.609037160873413
    },
    {
      "auxiliary_loss_clip": 0.01125005,
      "auxiliary_loss_mlp": 0.01028119,
      "balance_loss_clip": 1.04139495,
      "balance_loss_mlp": 1.02074564,
      "epoch": 0.7645042986833404,
      "flos": 16983234115200.0,
      "grad_norm": 1.9632775658355854,
      "language_loss": 0.82754779,
      "learning_rate": 5.539335622430235e-07,
      "loss": 0.84907901,
      "num_input_tokens_seen": 136799545,
      "step": 6358,
      "time_per_iteration": 2.483186960220337
    },
    {
      "auxiliary_loss_clip": 0.01146309,
      "auxiliary_loss_mlp": 0.01026911,
      "balance_loss_clip": 1.04414415,
      "balance_loss_mlp": 1.01919866,
      "epoch": 0.7646245415739794,
      "flos": 17311493531520.0,
      "grad_norm": 2.0609308329142393,
      "language_loss": 0.74821275,
      "learning_rate": 5.533955479946975e-07,
      "loss": 0.76994491,
      "num_input_tokens_seen": 136818325,
      "step": 6359,
      "time_per_iteration": 2.4492714405059814
    },
    {
      "auxiliary_loss_clip": 0.01034095,
      "auxiliary_loss_mlp": 0.00753077,
      "balance_loss_clip": 1.02201557,
      "balance_loss_mlp": 0.99990511,
      "epoch": 0.7647447844646186,
      "flos": 70402332666240.0,
      "grad_norm": 0.8732725267242816,
      "language_loss": 0.65782553,
      "learning_rate": 5.528577531951173e-07,
      "loss": 0.67569721,
      "num_input_tokens_seen": 136878730,
      "step": 6360,
      "time_per_iteration": 3.057543992996216
    },
    {
      "auxiliary_loss_clip": 0.01144509,
      "auxiliary_loss_mlp": 0.01022368,
      "balance_loss_clip": 1.04717374,
      "balance_loss_mlp": 1.01574552,
      "epoch": 0.7648650273552576,
      "flos": 17675914965120.0,
      "grad_norm": 2.466310627002327,
      "language_loss": 0.73924428,
      "learning_rate": 5.523201779258653e-07,
      "loss": 0.76091301,
      "num_input_tokens_seen": 136897705,
      "step": 6361,
      "time_per_iteration": 2.45358943939209
    },
    {
      "auxiliary_loss_clip": 0.01165745,
      "auxiliary_loss_mlp": 0.01023562,
      "balance_loss_clip": 1.04690981,
      "balance_loss_mlp": 1.0161891,
      "epoch": 0.7649852702458967,
      "flos": 22162019247360.0,
      "grad_norm": 1.8628793001611974,
      "language_loss": 0.84109479,
      "learning_rate": 5.517828222684912e-07,
      "loss": 0.86298788,
      "num_input_tokens_seen": 136918360,
      "step": 6362,
      "time_per_iteration": 2.4129037857055664
    },
    {
      "auxiliary_loss_clip": 0.0104627,
      "auxiliary_loss_mlp": 0.01001362,
      "balance_loss_clip": 1.01423407,
      "balance_loss_mlp": 1.0002774,
      "epoch": 0.7651055131365359,
      "flos": 69848338227840.0,
      "grad_norm": 0.765834522646588,
      "language_loss": 0.59067565,
      "learning_rate": 5.512456863045117e-07,
      "loss": 0.61115199,
      "num_input_tokens_seen": 136979050,
      "step": 6363,
      "time_per_iteration": 3.871473789215088
    },
    {
      "auxiliary_loss_clip": 0.01166817,
      "auxiliary_loss_mlp": 0.01026442,
      "balance_loss_clip": 1.04681683,
      "balance_loss_mlp": 1.01928055,
      "epoch": 0.7652257560271749,
      "flos": 19464014931840.0,
      "grad_norm": 2.3135936498083147,
      "language_loss": 0.74049056,
      "learning_rate": 5.507087701154089e-07,
      "loss": 0.76242316,
      "num_input_tokens_seen": 136998970,
      "step": 6364,
      "time_per_iteration": 3.1764132976531982
    },
    {
      "auxiliary_loss_clip": 0.01112698,
      "auxiliary_loss_mlp": 0.01026497,
      "balance_loss_clip": 1.04280519,
      "balance_loss_mlp": 1.01966918,
      "epoch": 0.765345998917814,
      "flos": 15961108700160.0,
      "grad_norm": 1.9229534114550089,
      "language_loss": 0.75533187,
      "learning_rate": 5.50172073782634e-07,
      "loss": 0.7767238,
      "num_input_tokens_seen": 137016950,
      "step": 6365,
      "time_per_iteration": 3.3357675075531006
    },
    {
      "auxiliary_loss_clip": 0.01123193,
      "auxiliary_loss_mlp": 0.01027664,
      "balance_loss_clip": 1.04585505,
      "balance_loss_mlp": 1.02054691,
      "epoch": 0.7654662418084531,
      "flos": 23659853408640.0,
      "grad_norm": 4.170047747101967,
      "language_loss": 0.87785631,
      "learning_rate": 5.496355973876023e-07,
      "loss": 0.89936483,
      "num_input_tokens_seen": 137036205,
      "step": 6366,
      "time_per_iteration": 2.5032167434692383
    },
    {
      "auxiliary_loss_clip": 0.01121971,
      "auxiliary_loss_mlp": 0.00762334,
      "balance_loss_clip": 1.043329,
      "balance_loss_mlp": 1.0003792,
      "epoch": 0.7655864846990922,
      "flos": 41463608878080.0,
      "grad_norm": 21.05141699631102,
      "language_loss": 0.70696187,
      "learning_rate": 5.490993410116984e-07,
      "loss": 0.72580487,
      "num_input_tokens_seen": 137059195,
      "step": 6367,
      "time_per_iteration": 2.689823627471924
    },
    {
      "auxiliary_loss_clip": 0.01123308,
      "auxiliary_loss_mlp": 0.01027936,
      "balance_loss_clip": 1.04692626,
      "balance_loss_mlp": 1.02109945,
      "epoch": 0.7657067275897312,
      "flos": 43142684088960.0,
      "grad_norm": 1.731143379743734,
      "language_loss": 0.69647586,
      "learning_rate": 5.485633047362704e-07,
      "loss": 0.71798825,
      "num_input_tokens_seen": 137081200,
      "step": 6368,
      "time_per_iteration": 2.697878837585449
    },
    {
      "auxiliary_loss_clip": 0.01174076,
      "auxiliary_loss_mlp": 0.01029737,
      "balance_loss_clip": 1.05254412,
      "balance_loss_mlp": 1.02226305,
      "epoch": 0.7658269704803703,
      "flos": 17311780840320.0,
      "grad_norm": 5.347963610685303,
      "language_loss": 0.78490973,
      "learning_rate": 5.480274886426341e-07,
      "loss": 0.80694783,
      "num_input_tokens_seen": 137097840,
      "step": 6369,
      "time_per_iteration": 3.1596245765686035
    },
    {
      "auxiliary_loss_clip": 0.01151102,
      "auxiliary_loss_mlp": 0.01022818,
      "balance_loss_clip": 1.04919469,
      "balance_loss_mlp": 1.01610088,
      "epoch": 0.7659472133710095,
      "flos": 12568160977920.0,
      "grad_norm": 1.9345798010231363,
      "language_loss": 0.77938432,
      "learning_rate": 5.474918928120744e-07,
      "loss": 0.8011235,
      "num_input_tokens_seen": 137114335,
      "step": 6370,
      "time_per_iteration": 2.4045112133026123
    },
    {
      "auxiliary_loss_clip": 0.0115023,
      "auxiliary_loss_mlp": 0.01021179,
      "balance_loss_clip": 1.0463264,
      "balance_loss_mlp": 1.01470876,
      "epoch": 0.7660674562616485,
      "flos": 22707430335360.0,
      "grad_norm": 1.9443593274740107,
      "language_loss": 0.87434733,
      "learning_rate": 5.469565173258392e-07,
      "loss": 0.89606148,
      "num_input_tokens_seen": 137132850,
      "step": 6371,
      "time_per_iteration": 2.441770076751709
    },
    {
      "auxiliary_loss_clip": 0.01170854,
      "auxiliary_loss_mlp": 0.01028128,
      "balance_loss_clip": 1.04856074,
      "balance_loss_mlp": 1.02049899,
      "epoch": 0.7661876991522876,
      "flos": 17056455989760.0,
      "grad_norm": 1.7071123412324896,
      "language_loss": 0.63544959,
      "learning_rate": 5.464213622651454e-07,
      "loss": 0.65743947,
      "num_input_tokens_seen": 137150665,
      "step": 6372,
      "time_per_iteration": 2.415595531463623
    },
    {
      "auxiliary_loss_clip": 0.01131167,
      "auxiliary_loss_mlp": 0.01026077,
      "balance_loss_clip": 1.04567599,
      "balance_loss_mlp": 1.01888835,
      "epoch": 0.7663079420429267,
      "flos": 20084228092800.0,
      "grad_norm": 1.6241856462781512,
      "language_loss": 0.84267974,
      "learning_rate": 5.458864277111753e-07,
      "loss": 0.86425221,
      "num_input_tokens_seen": 137168500,
      "step": 6373,
      "time_per_iteration": 2.4927902221679688
    },
    {
      "auxiliary_loss_clip": 0.01131159,
      "auxiliary_loss_mlp": 0.0076095,
      "balance_loss_clip": 1.04256582,
      "balance_loss_mlp": 1.00027359,
      "epoch": 0.7664281849335658,
      "flos": 12677473042560.0,
      "grad_norm": 2.3540503747733594,
      "language_loss": 0.69292229,
      "learning_rate": 5.453517137450769e-07,
      "loss": 0.71184337,
      "num_input_tokens_seen": 137185075,
      "step": 6374,
      "time_per_iteration": 2.4372057914733887
    },
    {
      "auxiliary_loss_clip": 0.01152503,
      "auxiliary_loss_mlp": 0.01025327,
      "balance_loss_clip": 1.04839587,
      "balance_loss_mlp": 1.01791263,
      "epoch": 0.7665484278242048,
      "flos": 22345271458560.0,
      "grad_norm": 2.0147409502210394,
      "language_loss": 0.75881517,
      "learning_rate": 5.448172204479684e-07,
      "loss": 0.78059345,
      "num_input_tokens_seen": 137204355,
      "step": 6375,
      "time_per_iteration": 2.4697751998901367
    },
    {
      "auxiliary_loss_clip": 0.01164884,
      "auxiliary_loss_mlp": 0.01025059,
      "balance_loss_clip": 1.04755449,
      "balance_loss_mlp": 1.01797748,
      "epoch": 0.766668670714844,
      "flos": 23617909301760.0,
      "grad_norm": 1.9338781180831894,
      "language_loss": 0.7459054,
      "learning_rate": 5.442829479009294e-07,
      "loss": 0.76780474,
      "num_input_tokens_seen": 137223135,
      "step": 6376,
      "time_per_iteration": 2.409158945083618
    },
    {
      "auxiliary_loss_clip": 0.01159459,
      "auxiliary_loss_mlp": 0.01027574,
      "balance_loss_clip": 1.04668212,
      "balance_loss_mlp": 1.01988459,
      "epoch": 0.7667889136054831,
      "flos": 19427134642560.0,
      "grad_norm": 2.392654489155968,
      "language_loss": 0.71623802,
      "learning_rate": 5.437488961850103e-07,
      "loss": 0.73810834,
      "num_input_tokens_seen": 137242935,
      "step": 6377,
      "time_per_iteration": 2.4200100898742676
    },
    {
      "auxiliary_loss_clip": 0.01106178,
      "auxiliary_loss_mlp": 0.01024907,
      "balance_loss_clip": 1.04220891,
      "balance_loss_mlp": 1.01852059,
      "epoch": 0.7669091564961221,
      "flos": 26866352609280.0,
      "grad_norm": 1.7484663900688924,
      "language_loss": 0.75362182,
      "learning_rate": 5.432150653812258e-07,
      "loss": 0.77493268,
      "num_input_tokens_seen": 137262970,
      "step": 6378,
      "time_per_iteration": 2.573922872543335
    },
    {
      "auxiliary_loss_clip": 0.01150978,
      "auxiliary_loss_mlp": 0.01025203,
      "balance_loss_clip": 1.04763174,
      "balance_loss_mlp": 1.01786304,
      "epoch": 0.7670293993867613,
      "flos": 12385303816320.0,
      "grad_norm": 2.564257105109169,
      "language_loss": 0.82675385,
      "learning_rate": 5.42681455570557e-07,
      "loss": 0.84851563,
      "num_input_tokens_seen": 137279500,
      "step": 6379,
      "time_per_iteration": 2.4332940578460693
    },
    {
      "auxiliary_loss_clip": 0.01163439,
      "auxiliary_loss_mlp": 0.01022548,
      "balance_loss_clip": 1.04626977,
      "balance_loss_mlp": 1.01560998,
      "epoch": 0.7671496422774003,
      "flos": 21762944167680.0,
      "grad_norm": 1.991398931282379,
      "language_loss": 0.64638889,
      "learning_rate": 5.42148066833954e-07,
      "loss": 0.66824877,
      "num_input_tokens_seen": 137298745,
      "step": 6380,
      "time_per_iteration": 2.394430637359619
    },
    {
      "auxiliary_loss_clip": 0.01165104,
      "auxiliary_loss_mlp": 0.01024203,
      "balance_loss_clip": 1.04786932,
      "balance_loss_mlp": 1.01716423,
      "epoch": 0.7672698851680394,
      "flos": 21069221823360.0,
      "grad_norm": 2.2928881409693727,
      "language_loss": 0.75220078,
      "learning_rate": 5.416148992523289e-07,
      "loss": 0.77409387,
      "num_input_tokens_seen": 137317320,
      "step": 6381,
      "time_per_iteration": 2.3966023921966553
    },
    {
      "auxiliary_loss_clip": 0.01081798,
      "auxiliary_loss_mlp": 0.01025551,
      "balance_loss_clip": 1.0407145,
      "balance_loss_mlp": 1.01870561,
      "epoch": 0.7673901280586786,
      "flos": 16976697840000.0,
      "grad_norm": 2.3913270229965367,
      "language_loss": 0.784832,
      "learning_rate": 5.410819529065644e-07,
      "loss": 0.80590546,
      "num_input_tokens_seen": 137335275,
      "step": 6382,
      "time_per_iteration": 2.564643144607544
    },
    {
      "auxiliary_loss_clip": 0.01110222,
      "auxiliary_loss_mlp": 0.01022767,
      "balance_loss_clip": 1.04181111,
      "balance_loss_mlp": 1.0159874,
      "epoch": 0.7675103709493176,
      "flos": 29242669697280.0,
      "grad_norm": 2.0581881229278545,
      "language_loss": 0.65522915,
      "learning_rate": 5.405492278775079e-07,
      "loss": 0.67655903,
      "num_input_tokens_seen": 137355055,
      "step": 6383,
      "time_per_iteration": 2.597848415374756
    },
    {
      "auxiliary_loss_clip": 0.01139524,
      "auxiliary_loss_mlp": 0.01026378,
      "balance_loss_clip": 1.04418361,
      "balance_loss_mlp": 1.01903796,
      "epoch": 0.7676306138399567,
      "flos": 29023004073600.0,
      "grad_norm": 2.4440017688204447,
      "language_loss": 0.79515409,
      "learning_rate": 5.400167242459732e-07,
      "loss": 0.81681311,
      "num_input_tokens_seen": 137374015,
      "step": 6384,
      "time_per_iteration": 2.5313050746917725
    },
    {
      "auxiliary_loss_clip": 0.01150861,
      "auxiliary_loss_mlp": 0.01027705,
      "balance_loss_clip": 1.04624724,
      "balance_loss_mlp": 1.02072835,
      "epoch": 0.7677508567305958,
      "flos": 22565116650240.0,
      "grad_norm": 1.7362755035977755,
      "language_loss": 0.80614436,
      "learning_rate": 5.394844420927405e-07,
      "loss": 0.82793003,
      "num_input_tokens_seen": 137393625,
      "step": 6385,
      "time_per_iteration": 2.4353082180023193
    },
    {
      "auxiliary_loss_clip": 0.01165397,
      "auxiliary_loss_mlp": 0.01029275,
      "balance_loss_clip": 1.04734337,
      "balance_loss_mlp": 1.02205706,
      "epoch": 0.7678710996212349,
      "flos": 25411432222080.0,
      "grad_norm": 2.024455177025597,
      "language_loss": 0.73396903,
      "learning_rate": 5.389523814985562e-07,
      "loss": 0.75591576,
      "num_input_tokens_seen": 137413045,
      "step": 6386,
      "time_per_iteration": 2.427029848098755
    },
    {
      "auxiliary_loss_clip": 0.01111857,
      "auxiliary_loss_mlp": 0.01023286,
      "balance_loss_clip": 1.043805,
      "balance_loss_mlp": 1.01593113,
      "epoch": 0.767991342511874,
      "flos": 26756825063040.0,
      "grad_norm": 2.589964151125589,
      "language_loss": 0.76044774,
      "learning_rate": 5.384205425441344e-07,
      "loss": 0.78179914,
      "num_input_tokens_seen": 137433955,
      "step": 6387,
      "time_per_iteration": 2.579573631286621
    },
    {
      "auxiliary_loss_clip": 0.01139821,
      "auxiliary_loss_mlp": 0.01022253,
      "balance_loss_clip": 1.04385543,
      "balance_loss_mlp": 1.01521337,
      "epoch": 0.7681115854025131,
      "flos": 26359509749760.0,
      "grad_norm": 1.8126868902802877,
      "language_loss": 0.8405416,
      "learning_rate": 5.378889253101537e-07,
      "loss": 0.86216235,
      "num_input_tokens_seen": 137454510,
      "step": 6388,
      "time_per_iteration": 2.5149872303009033
    },
    {
      "auxiliary_loss_clip": 0.01153351,
      "auxiliary_loss_mlp": 0.01022075,
      "balance_loss_clip": 1.04650307,
      "balance_loss_mlp": 1.01528931,
      "epoch": 0.7682318282931522,
      "flos": 23257043314560.0,
      "grad_norm": 1.6333737764074625,
      "language_loss": 0.81280822,
      "learning_rate": 5.373575298772617e-07,
      "loss": 0.83456242,
      "num_input_tokens_seen": 137473630,
      "step": 6389,
      "time_per_iteration": 2.4467570781707764
    },
    {
      "auxiliary_loss_clip": 0.01059106,
      "auxiliary_loss_mlp": 0.01001136,
      "balance_loss_clip": 1.01214981,
      "balance_loss_mlp": 1.00000322,
      "epoch": 0.7683520711837912,
      "flos": 70072457137920.0,
      "grad_norm": 0.7662733122752389,
      "language_loss": 0.61310136,
      "learning_rate": 5.368263563260689e-07,
      "loss": 0.63370383,
      "num_input_tokens_seen": 137538765,
      "step": 6390,
      "time_per_iteration": 3.9423651695251465
    },
    {
      "auxiliary_loss_clip": 0.01153938,
      "auxiliary_loss_mlp": 0.01022762,
      "balance_loss_clip": 1.04655004,
      "balance_loss_mlp": 1.01564562,
      "epoch": 0.7684723140744304,
      "flos": 18624890332800.0,
      "grad_norm": 1.4744298952076518,
      "language_loss": 0.64075053,
      "learning_rate": 5.362954047371537e-07,
      "loss": 0.66251755,
      "num_input_tokens_seen": 137557875,
      "step": 6391,
      "time_per_iteration": 4.062402009963989
    },
    {
      "auxiliary_loss_clip": 0.01128922,
      "auxiliary_loss_mlp": 0.01027103,
      "balance_loss_clip": 1.05053234,
      "balance_loss_mlp": 1.01964617,
      "epoch": 0.7685925569650695,
      "flos": 27452989532160.0,
      "grad_norm": 2.862261316174651,
      "language_loss": 0.72164237,
      "learning_rate": 5.357646751910627e-07,
      "loss": 0.74320257,
      "num_input_tokens_seen": 137579055,
      "step": 6392,
      "time_per_iteration": 2.5438971519470215
    },
    {
      "auxiliary_loss_clip": 0.0113704,
      "auxiliary_loss_mlp": 0.0103107,
      "balance_loss_clip": 1.04458559,
      "balance_loss_mlp": 1.02365828,
      "epoch": 0.7687127998557085,
      "flos": 24535714642560.0,
      "grad_norm": 2.5933805690484544,
      "language_loss": 0.79862297,
      "learning_rate": 5.352341677683061e-07,
      "loss": 0.8203041,
      "num_input_tokens_seen": 137600355,
      "step": 6393,
      "time_per_iteration": 2.5085439682006836
    },
    {
      "auxiliary_loss_clip": 0.01132052,
      "auxiliary_loss_mlp": 0.010278,
      "balance_loss_clip": 1.04546452,
      "balance_loss_mlp": 1.02042115,
      "epoch": 0.7688330427463477,
      "flos": 25155963717120.0,
      "grad_norm": 1.8067173779569732,
      "language_loss": 0.7894783,
      "learning_rate": 5.347038825493617e-07,
      "loss": 0.81107688,
      "num_input_tokens_seen": 137621885,
      "step": 6394,
      "time_per_iteration": 2.5423941612243652
    },
    {
      "auxiliary_loss_clip": 0.01136182,
      "auxiliary_loss_mlp": 0.01024991,
      "balance_loss_clip": 1.04805613,
      "balance_loss_mlp": 1.01808882,
      "epoch": 0.7689532856369867,
      "flos": 21211284113280.0,
      "grad_norm": 2.348601471180008,
      "language_loss": 0.68609875,
      "learning_rate": 5.341738196146732e-07,
      "loss": 0.7077105,
      "num_input_tokens_seen": 137640230,
      "step": 6395,
      "time_per_iteration": 2.462198257446289
    },
    {
      "auxiliary_loss_clip": 0.01148984,
      "auxiliary_loss_mlp": 0.01022085,
      "balance_loss_clip": 1.04504156,
      "balance_loss_mlp": 1.01493239,
      "epoch": 0.7690735285276258,
      "flos": 25119083427840.0,
      "grad_norm": 2.3378419669585626,
      "language_loss": 0.73600239,
      "learning_rate": 5.336439790446503e-07,
      "loss": 0.75771308,
      "num_input_tokens_seen": 137659330,
      "step": 6396,
      "time_per_iteration": 3.214081048965454
    },
    {
      "auxiliary_loss_clip": 0.01119193,
      "auxiliary_loss_mlp": 0.01026499,
      "balance_loss_clip": 1.04023659,
      "balance_loss_mlp": 1.01849461,
      "epoch": 0.769193771418265,
      "flos": 54744020640000.0,
      "grad_norm": 1.7038441912025504,
      "language_loss": 0.62881178,
      "learning_rate": 5.331143609196711e-07,
      "loss": 0.65026867,
      "num_input_tokens_seen": 137683145,
      "step": 6397,
      "time_per_iteration": 2.8088443279266357
    },
    {
      "auxiliary_loss_clip": 0.01153538,
      "auxiliary_loss_mlp": 0.01024576,
      "balance_loss_clip": 1.04888165,
      "balance_loss_mlp": 1.01738739,
      "epoch": 0.769314014308904,
      "flos": 37341890115840.0,
      "grad_norm": 1.6713104968828993,
      "language_loss": 0.76808703,
      "learning_rate": 5.325849653200758e-07,
      "loss": 0.78986812,
      "num_input_tokens_seen": 137707095,
      "step": 6398,
      "time_per_iteration": 2.5601727962493896
    },
    {
      "auxiliary_loss_clip": 0.0116715,
      "auxiliary_loss_mlp": 0.01023884,
      "balance_loss_clip": 1.0490346,
      "balance_loss_mlp": 1.01664782,
      "epoch": 0.7694342571995431,
      "flos": 20631686256000.0,
      "grad_norm": 1.6415384007702871,
      "language_loss": 0.76343262,
      "learning_rate": 5.32055792326175e-07,
      "loss": 0.78534299,
      "num_input_tokens_seen": 137725520,
      "step": 6399,
      "time_per_iteration": 2.4432456493377686
    },
    {
      "auxiliary_loss_clip": 0.01141757,
      "auxiliary_loss_mlp": 0.01021077,
      "balance_loss_clip": 1.04749393,
      "balance_loss_mlp": 1.01371002,
      "epoch": 0.7695545000901821,
      "flos": 24207706621440.0,
      "grad_norm": 2.3435896626514996,
      "language_loss": 0.73139322,
      "learning_rate": 5.315268420182437e-07,
      "loss": 0.75302148,
      "num_input_tokens_seen": 137744195,
      "step": 6400,
      "time_per_iteration": 2.4782979488372803
    },
    {
      "auxiliary_loss_clip": 0.01129143,
      "auxiliary_loss_mlp": 0.00761597,
      "balance_loss_clip": 1.04450679,
      "balance_loss_mlp": 1.00037694,
      "epoch": 0.7696747429808213,
      "flos": 28001273708160.0,
      "grad_norm": 1.7500450904575275,
      "language_loss": 0.76619947,
      "learning_rate": 5.309981144765221e-07,
      "loss": 0.7851069,
      "num_input_tokens_seen": 137764340,
      "step": 6401,
      "time_per_iteration": 2.6107680797576904
    },
    {
      "auxiliary_loss_clip": 0.01115922,
      "auxiliary_loss_mlp": 0.01020622,
      "balance_loss_clip": 1.04253924,
      "balance_loss_mlp": 1.01420283,
      "epoch": 0.7697949858714603,
      "flos": 11509550323200.0,
      "grad_norm": 3.108236240830104,
      "language_loss": 0.75308383,
      "learning_rate": 5.304696097812196e-07,
      "loss": 0.77444929,
      "num_input_tokens_seen": 137780940,
      "step": 6402,
      "time_per_iteration": 2.5105855464935303
    },
    {
      "auxiliary_loss_clip": 0.01135856,
      "auxiliary_loss_mlp": 0.01029401,
      "balance_loss_clip": 1.04330397,
      "balance_loss_mlp": 1.02156341,
      "epoch": 0.7699152287620994,
      "flos": 26688271956480.0,
      "grad_norm": 3.265144608985128,
      "language_loss": 0.60605782,
      "learning_rate": 5.299413280125078e-07,
      "loss": 0.62771046,
      "num_input_tokens_seen": 137799250,
      "step": 6403,
      "time_per_iteration": 2.5085442066192627
    },
    {
      "auxiliary_loss_clip": 0.01139339,
      "auxiliary_loss_mlp": 0.01029175,
      "balance_loss_clip": 1.04530358,
      "balance_loss_mlp": 1.02209449,
      "epoch": 0.7700354716527386,
      "flos": 16544944362240.0,
      "grad_norm": 2.181327075446862,
      "language_loss": 0.72531736,
      "learning_rate": 5.294132692505284e-07,
      "loss": 0.74700254,
      "num_input_tokens_seen": 137817660,
      "step": 6404,
      "time_per_iteration": 2.466015338897705
    },
    {
      "auxiliary_loss_clip": 0.01101995,
      "auxiliary_loss_mlp": 0.01025704,
      "balance_loss_clip": 1.04092407,
      "balance_loss_mlp": 1.01808906,
      "epoch": 0.7701557145433776,
      "flos": 19242733196160.0,
      "grad_norm": 1.8960630050847007,
      "language_loss": 0.79147637,
      "learning_rate": 5.288854335753861e-07,
      "loss": 0.81275332,
      "num_input_tokens_seen": 137835920,
      "step": 6405,
      "time_per_iteration": 2.5346932411193848
    },
    {
      "auxiliary_loss_clip": 0.01152231,
      "auxiliary_loss_mlp": 0.01020601,
      "balance_loss_clip": 1.044837,
      "balance_loss_mlp": 1.01351082,
      "epoch": 0.7702759574340167,
      "flos": 31685744211840.0,
      "grad_norm": 2.386096919592745,
      "language_loss": 0.75552845,
      "learning_rate": 5.283578210671551e-07,
      "loss": 0.77725673,
      "num_input_tokens_seen": 137858160,
      "step": 6406,
      "time_per_iteration": 2.52358078956604
    },
    {
      "auxiliary_loss_clip": 0.01141899,
      "auxiliary_loss_mlp": 0.01021041,
      "balance_loss_clip": 1.04589963,
      "balance_loss_mlp": 1.01400757,
      "epoch": 0.7703962003246558,
      "flos": 16800089644800.0,
      "grad_norm": 1.9971462237839832,
      "language_loss": 0.76459974,
      "learning_rate": 5.278304318058719e-07,
      "loss": 0.78622913,
      "num_input_tokens_seen": 137876015,
      "step": 6407,
      "time_per_iteration": 2.442455768585205
    },
    {
      "auxiliary_loss_clip": 0.01097723,
      "auxiliary_loss_mlp": 0.01027533,
      "balance_loss_clip": 1.04268956,
      "balance_loss_mlp": 1.01976061,
      "epoch": 0.7705164432152949,
      "flos": 35736072693120.0,
      "grad_norm": 1.9523814734497202,
      "language_loss": 0.79077709,
      "learning_rate": 5.273032658715411e-07,
      "loss": 0.81202972,
      "num_input_tokens_seen": 137898825,
      "step": 6408,
      "time_per_iteration": 2.6898953914642334
    },
    {
      "auxiliary_loss_clip": 0.01107563,
      "auxiliary_loss_mlp": 0.01023865,
      "balance_loss_clip": 1.04177475,
      "balance_loss_mlp": 1.01662326,
      "epoch": 0.7706366861059339,
      "flos": 23365960329600.0,
      "grad_norm": 2.030378492407595,
      "language_loss": 0.76631641,
      "learning_rate": 5.267763233441347e-07,
      "loss": 0.7876308,
      "num_input_tokens_seen": 137919455,
      "step": 6409,
      "time_per_iteration": 2.551865339279175
    },
    {
      "auxiliary_loss_clip": 0.0115581,
      "auxiliary_loss_mlp": 0.01022356,
      "balance_loss_clip": 1.04747939,
      "balance_loss_mlp": 1.01460743,
      "epoch": 0.7707569289965731,
      "flos": 22929897219840.0,
      "grad_norm": 3.1573725336824485,
      "language_loss": 0.6991812,
      "learning_rate": 5.26249604303588e-07,
      "loss": 0.72096288,
      "num_input_tokens_seen": 137937960,
      "step": 6410,
      "time_per_iteration": 2.443208694458008
    },
    {
      "auxiliary_loss_clip": 0.01166478,
      "auxiliary_loss_mlp": 0.01025082,
      "balance_loss_clip": 1.04868722,
      "balance_loss_mlp": 1.01802516,
      "epoch": 0.7708771718872122,
      "flos": 17420661941760.0,
      "grad_norm": 2.1389885977442047,
      "language_loss": 0.78599912,
      "learning_rate": 5.257231088298057e-07,
      "loss": 0.80791473,
      "num_input_tokens_seen": 137956370,
      "step": 6411,
      "time_per_iteration": 2.3761322498321533
    },
    {
      "auxiliary_loss_clip": 0.01032962,
      "auxiliary_loss_mlp": 0.01001088,
      "balance_loss_clip": 1.01138186,
      "balance_loss_mlp": 0.99999762,
      "epoch": 0.7709974147778512,
      "flos": 72241316248320.0,
      "grad_norm": 0.7985975284489502,
      "language_loss": 0.53935266,
      "learning_rate": 5.25196837002655e-07,
      "loss": 0.55969322,
      "num_input_tokens_seen": 138016080,
      "step": 6412,
      "time_per_iteration": 3.1264119148254395
    },
    {
      "auxiliary_loss_clip": 0.01136337,
      "auxiliary_loss_mlp": 0.01033831,
      "balance_loss_clip": 1.04465556,
      "balance_loss_mlp": 1.0263474,
      "epoch": 0.7711176576684904,
      "flos": 39859694876160.0,
      "grad_norm": 2.687812407898376,
      "language_loss": 0.68508303,
      "learning_rate": 5.24670788901971e-07,
      "loss": 0.70678473,
      "num_input_tokens_seen": 138039170,
      "step": 6413,
      "time_per_iteration": 2.6299617290496826
    },
    {
      "auxiliary_loss_clip": 0.01138367,
      "auxiliary_loss_mlp": 0.01026295,
      "balance_loss_clip": 1.04648781,
      "balance_loss_mlp": 1.01796854,
      "epoch": 0.7712379005591294,
      "flos": 36976391274240.0,
      "grad_norm": 2.625574801016011,
      "language_loss": 0.68696755,
      "learning_rate": 5.241449646075557e-07,
      "loss": 0.70861423,
      "num_input_tokens_seen": 138062395,
      "step": 6414,
      "time_per_iteration": 2.5986409187316895
    },
    {
      "auxiliary_loss_clip": 0.01161972,
      "auxiliary_loss_mlp": 0.01026753,
      "balance_loss_clip": 1.0483228,
      "balance_loss_mlp": 1.0195111,
      "epoch": 0.7713581434497685,
      "flos": 22776773541120.0,
      "grad_norm": 2.105796709133199,
      "language_loss": 0.72661221,
      "learning_rate": 5.236193641991762e-07,
      "loss": 0.74849951,
      "num_input_tokens_seen": 138080325,
      "step": 6415,
      "time_per_iteration": 2.4423716068267822
    },
    {
      "auxiliary_loss_clip": 0.01137557,
      "auxiliary_loss_mlp": 0.01024376,
      "balance_loss_clip": 1.04565716,
      "balance_loss_mlp": 1.01741719,
      "epoch": 0.7714783863404077,
      "flos": 24097460803200.0,
      "grad_norm": 2.363846756376933,
      "language_loss": 0.70127535,
      "learning_rate": 5.23093987756565e-07,
      "loss": 0.72289467,
      "num_input_tokens_seen": 138099020,
      "step": 6416,
      "time_per_iteration": 3.344449281692505
    },
    {
      "auxiliary_loss_clip": 0.01128316,
      "auxiliary_loss_mlp": 0.01026605,
      "balance_loss_clip": 1.04139125,
      "balance_loss_mlp": 1.019104,
      "epoch": 0.7715986292310467,
      "flos": 21063655215360.0,
      "grad_norm": 1.8919636641419346,
      "language_loss": 0.75455183,
      "learning_rate": 5.225688353594217e-07,
      "loss": 0.77610105,
      "num_input_tokens_seen": 138118650,
      "step": 6417,
      "time_per_iteration": 3.261704683303833
    },
    {
      "auxiliary_loss_clip": 0.01143051,
      "auxiliary_loss_mlp": 0.0076162,
      "balance_loss_clip": 1.04709172,
      "balance_loss_mlp": 1.0003562,
      "epoch": 0.7717188721216858,
      "flos": 20594877793920.0,
      "grad_norm": 2.5151830443973946,
      "language_loss": 0.77783799,
      "learning_rate": 5.220439070874108e-07,
      "loss": 0.79688466,
      "num_input_tokens_seen": 138137890,
      "step": 6418,
      "time_per_iteration": 3.306267738342285
    },
    {
      "auxiliary_loss_clip": 0.01153546,
      "auxiliary_loss_mlp": 0.01026658,
      "balance_loss_clip": 1.04918706,
      "balance_loss_mlp": 1.01971447,
      "epoch": 0.7718391150123249,
      "flos": 26250951870720.0,
      "grad_norm": 1.8273201541943596,
      "language_loss": 0.71487725,
      "learning_rate": 5.215192030201652e-07,
      "loss": 0.73667926,
      "num_input_tokens_seen": 138158880,
      "step": 6419,
      "time_per_iteration": 2.4934487342834473
    },
    {
      "auxiliary_loss_clip": 0.01110969,
      "auxiliary_loss_mlp": 0.01026385,
      "balance_loss_clip": 1.03986096,
      "balance_loss_mlp": 1.01911354,
      "epoch": 0.771959357902964,
      "flos": 22049762267520.0,
      "grad_norm": 2.124705495135433,
      "language_loss": 0.86177957,
      "learning_rate": 5.209947232372798e-07,
      "loss": 0.88315308,
      "num_input_tokens_seen": 138176370,
      "step": 6420,
      "time_per_iteration": 2.4981281757354736
    },
    {
      "auxiliary_loss_clip": 0.01154832,
      "auxiliary_loss_mlp": 0.00761693,
      "balance_loss_clip": 1.04481959,
      "balance_loss_mlp": 1.00033069,
      "epoch": 0.772079600793603,
      "flos": 30446000248320.0,
      "grad_norm": 1.711911543446483,
      "language_loss": 0.81308931,
      "learning_rate": 5.204704678183196e-07,
      "loss": 0.83225459,
      "num_input_tokens_seen": 138195105,
      "step": 6421,
      "time_per_iteration": 2.511103391647339
    },
    {
      "auxiliary_loss_clip": 0.01167727,
      "auxiliary_loss_mlp": 0.01023517,
      "balance_loss_clip": 1.04906154,
      "balance_loss_mlp": 1.01585793,
      "epoch": 0.7721998436842422,
      "flos": 12969857750400.0,
      "grad_norm": 1.8407963179988214,
      "language_loss": 0.85071194,
      "learning_rate": 5.19946436842813e-07,
      "loss": 0.87262434,
      "num_input_tokens_seen": 138212235,
      "step": 6422,
      "time_per_iteration": 3.145216703414917
    },
    {
      "auxiliary_loss_clip": 0.01129394,
      "auxiliary_loss_mlp": 0.01022093,
      "balance_loss_clip": 1.04907441,
      "balance_loss_mlp": 1.01520836,
      "epoch": 0.7723200865748813,
      "flos": 32635509678720.0,
      "grad_norm": 1.7149964127767598,
      "language_loss": 0.68343997,
      "learning_rate": 5.194226303902546e-07,
      "loss": 0.70495486,
      "num_input_tokens_seen": 138231970,
      "step": 6423,
      "time_per_iteration": 2.592524528503418
    },
    {
      "auxiliary_loss_clip": 0.01136572,
      "auxiliary_loss_mlp": 0.0102629,
      "balance_loss_clip": 1.04452682,
      "balance_loss_mlp": 1.01940012,
      "epoch": 0.7724403294655203,
      "flos": 21105707063040.0,
      "grad_norm": 1.8414754019365667,
      "language_loss": 0.71045929,
      "learning_rate": 5.188990485401072e-07,
      "loss": 0.73208791,
      "num_input_tokens_seen": 138251175,
      "step": 6424,
      "time_per_iteration": 2.4589998722076416
    },
    {
      "auxiliary_loss_clip": 0.01153004,
      "auxiliary_loss_mlp": 0.01020877,
      "balance_loss_clip": 1.04669023,
      "balance_loss_mlp": 1.01413012,
      "epoch": 0.7725605723561595,
      "flos": 22090736707200.0,
      "grad_norm": 1.873710553877904,
      "language_loss": 0.85851347,
      "learning_rate": 5.183756913717954e-07,
      "loss": 0.88025224,
      "num_input_tokens_seen": 138270950,
      "step": 6425,
      "time_per_iteration": 2.457789897918701
    },
    {
      "auxiliary_loss_clip": 0.01134962,
      "auxiliary_loss_mlp": 0.01029954,
      "balance_loss_clip": 1.04533386,
      "balance_loss_mlp": 1.02292693,
      "epoch": 0.7726808152467985,
      "flos": 34495610457600.0,
      "grad_norm": 2.513988629670602,
      "language_loss": 0.73195076,
      "learning_rate": 5.178525589647136e-07,
      "loss": 0.75359988,
      "num_input_tokens_seen": 138292590,
      "step": 6426,
      "time_per_iteration": 2.5548133850097656
    },
    {
      "auxiliary_loss_clip": 0.01142894,
      "auxiliary_loss_mlp": 0.01023533,
      "balance_loss_clip": 1.04459167,
      "balance_loss_mlp": 1.01694965,
      "epoch": 0.7728010581374376,
      "flos": 22306344094080.0,
      "grad_norm": 1.7305743299857412,
      "language_loss": 0.78915191,
      "learning_rate": 5.173296513982197e-07,
      "loss": 0.81081617,
      "num_input_tokens_seen": 138311115,
      "step": 6427,
      "time_per_iteration": 2.4777238368988037
    },
    {
      "auxiliary_loss_clip": 0.01134783,
      "auxiliary_loss_mlp": 0.01027905,
      "balance_loss_clip": 1.04711282,
      "balance_loss_mlp": 1.01950645,
      "epoch": 0.7729213010280768,
      "flos": 27126453968640.0,
      "grad_norm": 1.9938340598284112,
      "language_loss": 0.64725649,
      "learning_rate": 5.168069687516398e-07,
      "loss": 0.66888338,
      "num_input_tokens_seen": 138330885,
      "step": 6428,
      "time_per_iteration": 2.551393985748291
    },
    {
      "auxiliary_loss_clip": 0.01140267,
      "auxiliary_loss_mlp": 0.01021382,
      "balance_loss_clip": 1.04872108,
      "balance_loss_mlp": 1.01407135,
      "epoch": 0.7730415439187158,
      "flos": 18150223080960.0,
      "grad_norm": 1.8176777061761609,
      "language_loss": 0.71884924,
      "learning_rate": 5.16284511104263e-07,
      "loss": 0.7404657,
      "num_input_tokens_seen": 138350020,
      "step": 6429,
      "time_per_iteration": 2.4699227809906006
    },
    {
      "auxiliary_loss_clip": 0.01137933,
      "auxiliary_loss_mlp": 0.01025744,
      "balance_loss_clip": 1.04636443,
      "balance_loss_mlp": 1.01819777,
      "epoch": 0.7731617868093549,
      "flos": 11947480940160.0,
      "grad_norm": 2.564977626660824,
      "language_loss": 0.80906701,
      "learning_rate": 5.157622785353457e-07,
      "loss": 0.83070374,
      "num_input_tokens_seen": 138368135,
      "step": 6430,
      "time_per_iteration": 2.431574821472168
    },
    {
      "auxiliary_loss_clip": 0.01057661,
      "auxiliary_loss_mlp": 0.0100141,
      "balance_loss_clip": 1.01181853,
      "balance_loss_mlp": 1.00029504,
      "epoch": 0.7732820296999939,
      "flos": 64201027069440.0,
      "grad_norm": 0.6419605017811381,
      "language_loss": 0.60350251,
      "learning_rate": 5.152402711241113e-07,
      "loss": 0.62409329,
      "num_input_tokens_seen": 138436040,
      "step": 6431,
      "time_per_iteration": 3.090681791305542
    },
    {
      "auxiliary_loss_clip": 0.01118897,
      "auxiliary_loss_mlp": 0.01021826,
      "balance_loss_clip": 1.03975129,
      "balance_loss_mlp": 1.01476574,
      "epoch": 0.7734022725906331,
      "flos": 25302191984640.0,
      "grad_norm": 1.810689450895245,
      "language_loss": 0.8296504,
      "learning_rate": 5.147184889497465e-07,
      "loss": 0.85105759,
      "num_input_tokens_seen": 138455510,
      "step": 6432,
      "time_per_iteration": 2.5600802898406982
    },
    {
      "auxiliary_loss_clip": 0.01116893,
      "auxiliary_loss_mlp": 0.01024007,
      "balance_loss_clip": 1.04305601,
      "balance_loss_mlp": 1.01667523,
      "epoch": 0.7735225154812722,
      "flos": 17347440067200.0,
      "grad_norm": 2.6104377427408134,
      "language_loss": 0.8023839,
      "learning_rate": 5.141969320914072e-07,
      "loss": 0.82379293,
      "num_input_tokens_seen": 138473015,
      "step": 6433,
      "time_per_iteration": 2.4652419090270996
    },
    {
      "auxiliary_loss_clip": 0.01169906,
      "auxiliary_loss_mlp": 0.0102564,
      "balance_loss_clip": 1.04759467,
      "balance_loss_mlp": 1.01796913,
      "epoch": 0.7736427583719112,
      "flos": 32630086725120.0,
      "grad_norm": 2.1434589849239587,
      "language_loss": 0.62281764,
      "learning_rate": 5.136756006282113e-07,
      "loss": 0.64477313,
      "num_input_tokens_seen": 138491680,
      "step": 6434,
      "time_per_iteration": 2.4845428466796875
    },
    {
      "auxiliary_loss_clip": 0.01168518,
      "auxiliary_loss_mlp": 0.01025235,
      "balance_loss_clip": 1.04943657,
      "balance_loss_mlp": 1.0178616,
      "epoch": 0.7737630012625504,
      "flos": 19860073269120.0,
      "grad_norm": 2.137264941480943,
      "language_loss": 0.85020149,
      "learning_rate": 5.131544946392446e-07,
      "loss": 0.87213904,
      "num_input_tokens_seen": 138506960,
      "step": 6435,
      "time_per_iteration": 2.3737845420837402
    },
    {
      "auxiliary_loss_clip": 0.01143483,
      "auxiliary_loss_mlp": 0.01025696,
      "balance_loss_clip": 1.05161285,
      "balance_loss_mlp": 1.01839757,
      "epoch": 0.7738832441531894,
      "flos": 36022639397760.0,
      "grad_norm": 2.589393978676421,
      "language_loss": 0.6365037,
      "learning_rate": 5.126336142035592e-07,
      "loss": 0.6581955,
      "num_input_tokens_seen": 138526995,
      "step": 6436,
      "time_per_iteration": 2.6106059551239014
    },
    {
      "auxiliary_loss_clip": 0.01138641,
      "auxiliary_loss_mlp": 0.01025322,
      "balance_loss_clip": 1.04440117,
      "balance_loss_mlp": 1.01779437,
      "epoch": 0.7740034870438285,
      "flos": 13405274415360.0,
      "grad_norm": 2.814142197144034,
      "language_loss": 0.72009242,
      "learning_rate": 5.121129594001721e-07,
      "loss": 0.74173212,
      "num_input_tokens_seen": 138541260,
      "step": 6437,
      "time_per_iteration": 2.4573123455047607
    },
    {
      "auxiliary_loss_clip": 0.01153181,
      "auxiliary_loss_mlp": 0.01025154,
      "balance_loss_clip": 1.04819942,
      "balance_loss_mlp": 1.01763141,
      "epoch": 0.7741237299344677,
      "flos": 22086714384000.0,
      "grad_norm": 1.5735089788083296,
      "language_loss": 0.80978227,
      "learning_rate": 5.115925303080661e-07,
      "loss": 0.83156556,
      "num_input_tokens_seen": 138560970,
      "step": 6438,
      "time_per_iteration": 2.476163387298584
    },
    {
      "auxiliary_loss_clip": 0.01141201,
      "auxiliary_loss_mlp": 0.01025642,
      "balance_loss_clip": 1.0459826,
      "balance_loss_mlp": 1.01894593,
      "epoch": 0.7742439728251067,
      "flos": 19864777950720.0,
      "grad_norm": 2.1021872459072335,
      "language_loss": 0.79231328,
      "learning_rate": 5.110723270061899e-07,
      "loss": 0.81398171,
      "num_input_tokens_seen": 138577460,
      "step": 6439,
      "time_per_iteration": 2.441087245941162
    },
    {
      "auxiliary_loss_clip": 0.01163282,
      "auxiliary_loss_mlp": 0.01023818,
      "balance_loss_clip": 1.04651213,
      "balance_loss_mlp": 1.01740479,
      "epoch": 0.7743642157157458,
      "flos": 16690167048960.0,
      "grad_norm": 2.005066207805757,
      "language_loss": 0.79437143,
      "learning_rate": 5.105523495734572e-07,
      "loss": 0.81624246,
      "num_input_tokens_seen": 138594860,
      "step": 6440,
      "time_per_iteration": 2.3974616527557373
    },
    {
      "auxiliary_loss_clip": 0.0116725,
      "auxiliary_loss_mlp": 0.01028219,
      "balance_loss_clip": 1.04725432,
      "balance_loss_mlp": 1.02051222,
      "epoch": 0.7744844586063849,
      "flos": 20304360593280.0,
      "grad_norm": 1.521867159672832,
      "language_loss": 0.7495482,
      "learning_rate": 5.100325980887499e-07,
      "loss": 0.77150285,
      "num_input_tokens_seen": 138614785,
      "step": 6441,
      "time_per_iteration": 2.4130403995513916
    },
    {
      "auxiliary_loss_clip": 0.01148242,
      "auxiliary_loss_mlp": 0.01023467,
      "balance_loss_clip": 1.04713202,
      "balance_loss_mlp": 1.01623058,
      "epoch": 0.774604701497024,
      "flos": 22966705681920.0,
      "grad_norm": 1.7288709964931221,
      "language_loss": 0.83163971,
      "learning_rate": 5.095130726309116e-07,
      "loss": 0.85335678,
      "num_input_tokens_seen": 138634960,
      "step": 6442,
      "time_per_iteration": 2.4730496406555176
    },
    {
      "auxiliary_loss_clip": 0.0106572,
      "auxiliary_loss_mlp": 0.01000898,
      "balance_loss_clip": 1.01094031,
      "balance_loss_mlp": 0.99984854,
      "epoch": 0.774724944387663,
      "flos": 60288523073280.0,
      "grad_norm": 0.8013333376209394,
      "language_loss": 0.59004474,
      "learning_rate": 5.089937732787559e-07,
      "loss": 0.61071086,
      "num_input_tokens_seen": 138699520,
      "step": 6443,
      "time_per_iteration": 3.8609416484832764
    },
    {
      "auxiliary_loss_clip": 0.01125663,
      "auxiliary_loss_mlp": 0.0102824,
      "balance_loss_clip": 1.0437777,
      "balance_loss_mlp": 1.02057743,
      "epoch": 0.7748451872783022,
      "flos": 26761026954240.0,
      "grad_norm": 2.680038287115145,
      "language_loss": 0.66918653,
      "learning_rate": 5.084747001110592e-07,
      "loss": 0.69072556,
      "num_input_tokens_seen": 138719145,
      "step": 6444,
      "time_per_iteration": 3.225616455078125
    },
    {
      "auxiliary_loss_clip": 0.01152046,
      "auxiliary_loss_mlp": 0.00761735,
      "balance_loss_clip": 1.05140138,
      "balance_loss_mlp": 1.00038409,
      "epoch": 0.7749654301689413,
      "flos": 30338627518080.0,
      "grad_norm": 1.7290778018416668,
      "language_loss": 0.70300925,
      "learning_rate": 5.07955853206564e-07,
      "loss": 0.72214717,
      "num_input_tokens_seen": 138743850,
      "step": 6445,
      "time_per_iteration": 3.3695642948150635
    },
    {
      "auxiliary_loss_clip": 0.01156388,
      "auxiliary_loss_mlp": 0.01024178,
      "balance_loss_clip": 1.04747105,
      "balance_loss_mlp": 1.0171237,
      "epoch": 0.7750856730595803,
      "flos": 43179851687040.0,
      "grad_norm": 1.5053421106051914,
      "language_loss": 0.70791072,
      "learning_rate": 5.074372326439807e-07,
      "loss": 0.72971636,
      "num_input_tokens_seen": 138766860,
      "step": 6446,
      "time_per_iteration": 2.6301259994506836
    },
    {
      "auxiliary_loss_clip": 0.01126631,
      "auxiliary_loss_mlp": 0.01024101,
      "balance_loss_clip": 1.04497623,
      "balance_loss_mlp": 1.01697826,
      "epoch": 0.7752059159502195,
      "flos": 17640040256640.0,
      "grad_norm": 2.160646410047585,
      "language_loss": 0.73268616,
      "learning_rate": 5.069188385019814e-07,
      "loss": 0.75419348,
      "num_input_tokens_seen": 138784560,
      "step": 6447,
      "time_per_iteration": 2.468282461166382
    },
    {
      "auxiliary_loss_clip": 0.01116939,
      "auxiliary_loss_mlp": 0.01023589,
      "balance_loss_clip": 1.04172969,
      "balance_loss_mlp": 1.01630569,
      "epoch": 0.7753261588408585,
      "flos": 12677688524160.0,
      "grad_norm": 2.5037220948534036,
      "language_loss": 0.61468381,
      "learning_rate": 5.064006708592077e-07,
      "loss": 0.63608903,
      "num_input_tokens_seen": 138800805,
      "step": 6448,
      "time_per_iteration": 2.5272462368011475
    },
    {
      "auxiliary_loss_clip": 0.01132719,
      "auxiliary_loss_mlp": 0.01021799,
      "balance_loss_clip": 1.04613602,
      "balance_loss_mlp": 1.01481926,
      "epoch": 0.7754464017314976,
      "flos": 16690741666560.0,
      "grad_norm": 3.541752868752098,
      "language_loss": 0.75716442,
      "learning_rate": 5.058827297942641e-07,
      "loss": 0.77870965,
      "num_input_tokens_seen": 138815910,
      "step": 6449,
      "time_per_iteration": 3.141327142715454
    },
    {
      "auxiliary_loss_clip": 0.01145791,
      "auxiliary_loss_mlp": 0.0102569,
      "balance_loss_clip": 1.04628992,
      "balance_loss_mlp": 1.01848936,
      "epoch": 0.7755666446221368,
      "flos": 19718944732800.0,
      "grad_norm": 1.7544060391221827,
      "language_loss": 0.75124407,
      "learning_rate": 5.053650153857237e-07,
      "loss": 0.77295887,
      "num_input_tokens_seen": 138834920,
      "step": 6450,
      "time_per_iteration": 2.501918315887451
    },
    {
      "auxiliary_loss_clip": 0.01152202,
      "auxiliary_loss_mlp": 0.01027634,
      "balance_loss_clip": 1.04784274,
      "balance_loss_mlp": 1.02066088,
      "epoch": 0.7756868875127758,
      "flos": 18693623007360.0,
      "grad_norm": 1.6570058684961895,
      "language_loss": 0.70000458,
      "learning_rate": 5.048475277121214e-07,
      "loss": 0.72180295,
      "num_input_tokens_seen": 138852135,
      "step": 6451,
      "time_per_iteration": 2.40974497795105
    },
    {
      "auxiliary_loss_clip": 0.01151228,
      "auxiliary_loss_mlp": 0.01022735,
      "balance_loss_clip": 1.04468381,
      "balance_loss_mlp": 1.01558852,
      "epoch": 0.7758071304034149,
      "flos": 28404191543040.0,
      "grad_norm": 1.6769634699945188,
      "language_loss": 0.76976824,
      "learning_rate": 5.043302668519598e-07,
      "loss": 0.79150784,
      "num_input_tokens_seen": 138871470,
      "step": 6452,
      "time_per_iteration": 2.4969372749328613
    },
    {
      "auxiliary_loss_clip": 0.01155396,
      "auxiliary_loss_mlp": 0.01020389,
      "balance_loss_clip": 1.04569912,
      "balance_loss_mlp": 1.01350188,
      "epoch": 0.775927373294054,
      "flos": 20595344670720.0,
      "grad_norm": 2.080420852405165,
      "language_loss": 0.71731466,
      "learning_rate": 5.038132328837079e-07,
      "loss": 0.7390725,
      "num_input_tokens_seen": 138889860,
      "step": 6453,
      "time_per_iteration": 2.426710367202759
    },
    {
      "auxiliary_loss_clip": 0.01152916,
      "auxiliary_loss_mlp": 0.01020059,
      "balance_loss_clip": 1.04607224,
      "balance_loss_mlp": 1.01286435,
      "epoch": 0.7760476161846931,
      "flos": 22526368853760.0,
      "grad_norm": 2.033429583875141,
      "language_loss": 0.74034488,
      "learning_rate": 5.032964258857993e-07,
      "loss": 0.76207459,
      "num_input_tokens_seen": 138909955,
      "step": 6454,
      "time_per_iteration": 2.439711809158325
    },
    {
      "auxiliary_loss_clip": 0.01149091,
      "auxiliary_loss_mlp": 0.01024304,
      "balance_loss_clip": 1.04198992,
      "balance_loss_mlp": 1.01694012,
      "epoch": 0.7761678590753321,
      "flos": 48651488403840.0,
      "grad_norm": 1.5741380698309486,
      "language_loss": 0.68368697,
      "learning_rate": 5.027798459366329e-07,
      "loss": 0.70542085,
      "num_input_tokens_seen": 138935320,
      "step": 6455,
      "time_per_iteration": 2.666825294494629
    },
    {
      "auxiliary_loss_clip": 0.01157841,
      "auxiliary_loss_mlp": 0.01026364,
      "balance_loss_clip": 1.04690766,
      "balance_loss_mlp": 1.01894331,
      "epoch": 0.7762881019659713,
      "flos": 26177047637760.0,
      "grad_norm": 1.3340643596533168,
      "language_loss": 0.63644969,
      "learning_rate": 5.02263493114573e-07,
      "loss": 0.6582917,
      "num_input_tokens_seen": 138957115,
      "step": 6456,
      "time_per_iteration": 2.4717013835906982
    },
    {
      "auxiliary_loss_clip": 0.01163778,
      "auxiliary_loss_mlp": 0.01025042,
      "balance_loss_clip": 1.04592049,
      "balance_loss_mlp": 1.0176245,
      "epoch": 0.7764083448566104,
      "flos": 20588341518720.0,
      "grad_norm": 5.4999977880007185,
      "language_loss": 0.7733652,
      "learning_rate": 5.017473674979502e-07,
      "loss": 0.7952534,
      "num_input_tokens_seen": 138973140,
      "step": 6457,
      "time_per_iteration": 2.366192579269409
    },
    {
      "auxiliary_loss_clip": 0.010271,
      "auxiliary_loss_mlp": 0.01002277,
      "balance_loss_clip": 1.0127337,
      "balance_loss_mlp": 1.00142431,
      "epoch": 0.7765285877472494,
      "flos": 67293078560640.0,
      "grad_norm": 0.7388999738542961,
      "language_loss": 0.58362532,
      "learning_rate": 5.01231469165061e-07,
      "loss": 0.60391903,
      "num_input_tokens_seen": 139028965,
      "step": 6458,
      "time_per_iteration": 2.9756686687469482
    },
    {
      "auxiliary_loss_clip": 0.01056278,
      "auxiliary_loss_mlp": 0.01001753,
      "balance_loss_clip": 1.01111376,
      "balance_loss_mlp": 1.00070441,
      "epoch": 0.7766488306378886,
      "flos": 61344476121600.0,
      "grad_norm": 0.8341655305906173,
      "language_loss": 0.56878179,
      "learning_rate": 5.007157981941663e-07,
      "loss": 0.58936214,
      "num_input_tokens_seen": 139094325,
      "step": 6459,
      "time_per_iteration": 3.1317391395568848
    },
    {
      "auxiliary_loss_clip": 0.0104721,
      "auxiliary_loss_mlp": 0.01001525,
      "balance_loss_clip": 1.01143205,
      "balance_loss_mlp": 1.00048208,
      "epoch": 0.7767690735285276,
      "flos": 62946199393920.0,
      "grad_norm": 0.8776113354078642,
      "language_loss": 0.67385995,
      "learning_rate": 5.002003546634928e-07,
      "loss": 0.69434738,
      "num_input_tokens_seen": 139150425,
      "step": 6460,
      "time_per_iteration": 3.0022177696228027
    },
    {
      "auxiliary_loss_clip": 0.01111732,
      "auxiliary_loss_mlp": 0.01024074,
      "balance_loss_clip": 1.04784656,
      "balance_loss_mlp": 1.01729679,
      "epoch": 0.7768893164191667,
      "flos": 20886400575360.0,
      "grad_norm": 1.9647584450303748,
      "language_loss": 0.75964898,
      "learning_rate": 4.996851386512331e-07,
      "loss": 0.78100705,
      "num_input_tokens_seen": 139169130,
      "step": 6461,
      "time_per_iteration": 2.529478073120117
    },
    {
      "auxiliary_loss_clip": 0.01138187,
      "auxiliary_loss_mlp": 0.01026045,
      "balance_loss_clip": 1.04549611,
      "balance_loss_mlp": 1.0183115,
      "epoch": 0.7770095593098058,
      "flos": 20704584908160.0,
      "grad_norm": 1.7860559881382527,
      "language_loss": 0.83080018,
      "learning_rate": 4.991701502355444e-07,
      "loss": 0.85244256,
      "num_input_tokens_seen": 139189595,
      "step": 6462,
      "time_per_iteration": 2.465259552001953
    },
    {
      "auxiliary_loss_clip": 0.01155056,
      "auxiliary_loss_mlp": 0.01023779,
      "balance_loss_clip": 1.04640627,
      "balance_loss_mlp": 1.01743054,
      "epoch": 0.7771298022004449,
      "flos": 24717709877760.0,
      "grad_norm": 1.5090409968500693,
      "language_loss": 0.7594372,
      "learning_rate": 4.986553894945518e-07,
      "loss": 0.78122556,
      "num_input_tokens_seen": 139210805,
      "step": 6463,
      "time_per_iteration": 2.483222723007202
    },
    {
      "auxiliary_loss_clip": 0.01111352,
      "auxiliary_loss_mlp": 0.01023983,
      "balance_loss_clip": 1.0402081,
      "balance_loss_mlp": 1.01786137,
      "epoch": 0.777250045091084,
      "flos": 25009232659200.0,
      "grad_norm": 2.1081996816741966,
      "language_loss": 0.86050177,
      "learning_rate": 4.981408565063416e-07,
      "loss": 0.88185507,
      "num_input_tokens_seen": 139230750,
      "step": 6464,
      "time_per_iteration": 2.5557990074157715
    },
    {
      "auxiliary_loss_clip": 0.01167666,
      "auxiliary_loss_mlp": 0.01022888,
      "balance_loss_clip": 1.04819059,
      "balance_loss_mlp": 1.01584315,
      "epoch": 0.777370287981723,
      "flos": 20119887319680.0,
      "grad_norm": 2.0370118934167842,
      "language_loss": 0.7603848,
      "learning_rate": 4.976265513489701e-07,
      "loss": 0.78229034,
      "num_input_tokens_seen": 139250720,
      "step": 6465,
      "time_per_iteration": 2.4163033962249756
    },
    {
      "auxiliary_loss_clip": 0.01150861,
      "auxiliary_loss_mlp": 0.01024172,
      "balance_loss_clip": 1.04393113,
      "balance_loss_mlp": 1.01696539,
      "epoch": 0.7774905308723622,
      "flos": 21718809331200.0,
      "grad_norm": 1.8639734558211378,
      "language_loss": 0.80701333,
      "learning_rate": 4.971124741004562e-07,
      "loss": 0.82876366,
      "num_input_tokens_seen": 139269720,
      "step": 6466,
      "time_per_iteration": 2.438819169998169
    },
    {
      "auxiliary_loss_clip": 0.01150274,
      "auxiliary_loss_mlp": 0.01021797,
      "balance_loss_clip": 1.04517806,
      "balance_loss_mlp": 1.01493669,
      "epoch": 0.7776107737630013,
      "flos": 16034115093120.0,
      "grad_norm": 2.0543492683344833,
      "language_loss": 0.76488411,
      "learning_rate": 4.965986248387846e-07,
      "loss": 0.78660482,
      "num_input_tokens_seen": 139288035,
      "step": 6467,
      "time_per_iteration": 2.401381015777588
    },
    {
      "auxiliary_loss_clip": 0.01140179,
      "auxiliary_loss_mlp": 0.0102372,
      "balance_loss_clip": 1.04370236,
      "balance_loss_mlp": 1.01692843,
      "epoch": 0.7777310166536403,
      "flos": 24790895838720.0,
      "grad_norm": 1.6802066583015651,
      "language_loss": 0.76982903,
      "learning_rate": 4.960850036419073e-07,
      "loss": 0.79146802,
      "num_input_tokens_seen": 139307135,
      "step": 6468,
      "time_per_iteration": 2.494659423828125
    },
    {
      "auxiliary_loss_clip": 0.01134654,
      "auxiliary_loss_mlp": 0.01021582,
      "balance_loss_clip": 1.0441258,
      "balance_loss_mlp": 1.01435757,
      "epoch": 0.7778512595442795,
      "flos": 17272530253440.0,
      "grad_norm": 2.1862019421500607,
      "language_loss": 0.78757286,
      "learning_rate": 4.955716105877378e-07,
      "loss": 0.80913526,
      "num_input_tokens_seen": 139325905,
      "step": 6469,
      "time_per_iteration": 3.275754928588867
    },
    {
      "auxiliary_loss_clip": 0.01156343,
      "auxiliary_loss_mlp": 0.00761567,
      "balance_loss_clip": 1.04682326,
      "balance_loss_mlp": 1.00036108,
      "epoch": 0.7779715024349185,
      "flos": 17748418567680.0,
      "grad_norm": 1.7528109889418506,
      "language_loss": 0.82955408,
      "learning_rate": 4.950584457541598e-07,
      "loss": 0.84873319,
      "num_input_tokens_seen": 139344370,
      "step": 6470,
      "time_per_iteration": 3.13714861869812
    },
    {
      "auxiliary_loss_clip": 0.01154319,
      "auxiliary_loss_mlp": 0.01025287,
      "balance_loss_clip": 1.04663897,
      "balance_loss_mlp": 1.01850057,
      "epoch": 0.7780917453255576,
      "flos": 24316875031680.0,
      "grad_norm": 1.347970465380064,
      "language_loss": 0.81814688,
      "learning_rate": 4.945455092190183e-07,
      "loss": 0.83994293,
      "num_input_tokens_seen": 139365625,
      "step": 6471,
      "time_per_iteration": 3.3791353702545166
    },
    {
      "auxiliary_loss_clip": 0.01065539,
      "auxiliary_loss_mlp": 0.01000919,
      "balance_loss_clip": 1.01064253,
      "balance_loss_mlp": 0.99985808,
      "epoch": 0.7782119882161967,
      "flos": 56364601530240.0,
      "grad_norm": 0.6771088358195161,
      "language_loss": 0.55961573,
      "learning_rate": 4.940328010601271e-07,
      "loss": 0.5802803,
      "num_input_tokens_seen": 139430540,
      "step": 6472,
      "time_per_iteration": 3.017413377761841
    },
    {
      "auxiliary_loss_clip": 0.01151504,
      "auxiliary_loss_mlp": 0.01028954,
      "balance_loss_clip": 1.05152142,
      "balance_loss_mlp": 1.02137816,
      "epoch": 0.7783322311068358,
      "flos": 46789986994560.0,
      "grad_norm": 1.719036534090131,
      "language_loss": 0.76680022,
      "learning_rate": 4.935203213552621e-07,
      "loss": 0.78860486,
      "num_input_tokens_seen": 139454280,
      "step": 6473,
      "time_per_iteration": 2.691930055618286
    },
    {
      "auxiliary_loss_clip": 0.01141165,
      "auxiliary_loss_mlp": 0.01023115,
      "balance_loss_clip": 1.0461632,
      "balance_loss_mlp": 1.01555395,
      "epoch": 0.7784524739974749,
      "flos": 19057864872960.0,
      "grad_norm": 8.55752309875727,
      "language_loss": 0.66779333,
      "learning_rate": 4.930080701821662e-07,
      "loss": 0.68943614,
      "num_input_tokens_seen": 139471745,
      "step": 6474,
      "time_per_iteration": 2.4473488330841064
    },
    {
      "auxiliary_loss_clip": 0.01139418,
      "auxiliary_loss_mlp": 0.01025608,
      "balance_loss_clip": 1.04459834,
      "balance_loss_mlp": 1.01850629,
      "epoch": 0.778572716888114,
      "flos": 24791111320320.0,
      "grad_norm": 2.3501120765167505,
      "language_loss": 0.77194738,
      "learning_rate": 4.92496047618548e-07,
      "loss": 0.7935977,
      "num_input_tokens_seen": 139491505,
      "step": 6475,
      "time_per_iteration": 3.2657601833343506
    },
    {
      "auxiliary_loss_clip": 0.01157503,
      "auxiliary_loss_mlp": 0.01022966,
      "balance_loss_clip": 1.05059385,
      "balance_loss_mlp": 1.01574206,
      "epoch": 0.7786929597787531,
      "flos": 20078086867200.0,
      "grad_norm": 1.7811780938141082,
      "language_loss": 0.77793294,
      "learning_rate": 4.919842537420811e-07,
      "loss": 0.79973757,
      "num_input_tokens_seen": 139508620,
      "step": 6476,
      "time_per_iteration": 2.42966890335083
    },
    {
      "auxiliary_loss_clip": 0.01140987,
      "auxiliary_loss_mlp": 0.01026943,
      "balance_loss_clip": 1.04864883,
      "balance_loss_mlp": 1.02031803,
      "epoch": 0.7788132026693921,
      "flos": 21872220318720.0,
      "grad_norm": 1.5867789015892955,
      "language_loss": 0.79264104,
      "learning_rate": 4.91472688630404e-07,
      "loss": 0.81432033,
      "num_input_tokens_seen": 139529360,
      "step": 6477,
      "time_per_iteration": 2.474210500717163
    },
    {
      "auxiliary_loss_clip": 0.01163792,
      "auxiliary_loss_mlp": 0.01020733,
      "balance_loss_clip": 1.04691815,
      "balance_loss_mlp": 1.01408958,
      "epoch": 0.7789334455600313,
      "flos": 11181937351680.0,
      "grad_norm": 1.840047594253558,
      "language_loss": 0.74176162,
      "learning_rate": 4.909613523611202e-07,
      "loss": 0.76360685,
      "num_input_tokens_seen": 139546240,
      "step": 6478,
      "time_per_iteration": 2.438084363937378
    },
    {
      "auxiliary_loss_clip": 0.01106869,
      "auxiliary_loss_mlp": 0.00761989,
      "balance_loss_clip": 1.03926635,
      "balance_loss_mlp": 1.00046504,
      "epoch": 0.7790536884506704,
      "flos": 28695427015680.0,
      "grad_norm": 1.7131593213837262,
      "language_loss": 0.74512082,
      "learning_rate": 4.904502450117991e-07,
      "loss": 0.76380944,
      "num_input_tokens_seen": 139567200,
      "step": 6479,
      "time_per_iteration": 2.602010726928711
    },
    {
      "auxiliary_loss_clip": 0.01138148,
      "auxiliary_loss_mlp": 0.0102528,
      "balance_loss_clip": 1.0487994,
      "balance_loss_mlp": 1.01820242,
      "epoch": 0.7791739313413094,
      "flos": 11072302064640.0,
      "grad_norm": 2.2655887327138386,
      "language_loss": 0.7183193,
      "learning_rate": 4.899393666599762e-07,
      "loss": 0.73995364,
      "num_input_tokens_seen": 139583775,
      "step": 6480,
      "time_per_iteration": 2.4409658908843994
    },
    {
      "auxiliary_loss_clip": 0.01164531,
      "auxiliary_loss_mlp": 0.01019843,
      "balance_loss_clip": 1.04551697,
      "balance_loss_mlp": 1.01322973,
      "epoch": 0.7792941742319486,
      "flos": 14679276975360.0,
      "grad_norm": 2.451872446933454,
      "language_loss": 0.72448373,
      "learning_rate": 4.894287173831506e-07,
      "loss": 0.74632752,
      "num_input_tokens_seen": 139599735,
      "step": 6481,
      "time_per_iteration": 2.357515335083008
    },
    {
      "auxiliary_loss_clip": 0.01138921,
      "auxiliary_loss_mlp": 0.01025041,
      "balance_loss_clip": 1.04311824,
      "balance_loss_mlp": 1.01717901,
      "epoch": 0.7794144171225876,
      "flos": 23258874908160.0,
      "grad_norm": 2.7050119523459117,
      "language_loss": 0.84315956,
      "learning_rate": 4.889182972587877e-07,
      "loss": 0.8647992,
      "num_input_tokens_seen": 139619030,
      "step": 6482,
      "time_per_iteration": 2.469722270965576
    },
    {
      "auxiliary_loss_clip": 0.01132318,
      "auxiliary_loss_mlp": 0.01025619,
      "balance_loss_clip": 1.04506886,
      "balance_loss_mlp": 1.01877379,
      "epoch": 0.7795346600132267,
      "flos": 21507080613120.0,
      "grad_norm": 1.7858208545962788,
      "language_loss": 0.66032982,
      "learning_rate": 4.884081063643177e-07,
      "loss": 0.68190914,
      "num_input_tokens_seen": 139637690,
      "step": 6483,
      "time_per_iteration": 2.518369436264038
    },
    {
      "auxiliary_loss_clip": 0.01040475,
      "auxiliary_loss_mlp": 0.0100151,
      "balance_loss_clip": 1.01195192,
      "balance_loss_mlp": 1.00052702,
      "epoch": 0.7796549029038659,
      "flos": 70052273694720.0,
      "grad_norm": 0.8648663077288179,
      "language_loss": 0.52530181,
      "learning_rate": 4.878981447771353e-07,
      "loss": 0.54572165,
      "num_input_tokens_seen": 139692070,
      "step": 6484,
      "time_per_iteration": 3.0795040130615234
    },
    {
      "auxiliary_loss_clip": 0.01118967,
      "auxiliary_loss_mlp": 0.01025686,
      "balance_loss_clip": 1.04298294,
      "balance_loss_mlp": 1.01744246,
      "epoch": 0.7797751457945049,
      "flos": 23989405714560.0,
      "grad_norm": 1.651649588835177,
      "language_loss": 0.73228383,
      "learning_rate": 4.873884125746035e-07,
      "loss": 0.75373042,
      "num_input_tokens_seen": 139713745,
      "step": 6485,
      "time_per_iteration": 2.5281119346618652
    },
    {
      "auxiliary_loss_clip": 0.0113465,
      "auxiliary_loss_mlp": 0.01019535,
      "balance_loss_clip": 1.04448938,
      "balance_loss_mlp": 1.01248991,
      "epoch": 0.779895388685144,
      "flos": 22674751937280.0,
      "grad_norm": 2.448838597906675,
      "language_loss": 0.72434759,
      "learning_rate": 4.868789098340456e-07,
      "loss": 0.74588943,
      "num_input_tokens_seen": 139731650,
      "step": 6486,
      "time_per_iteration": 2.473839521408081
    },
    {
      "auxiliary_loss_clip": 0.01123394,
      "auxiliary_loss_mlp": 0.01023626,
      "balance_loss_clip": 1.04187012,
      "balance_loss_mlp": 1.01695895,
      "epoch": 0.7800156315757831,
      "flos": 23768698596480.0,
      "grad_norm": 4.081162562167619,
      "language_loss": 0.73117936,
      "learning_rate": 4.863696366327543e-07,
      "loss": 0.75264961,
      "num_input_tokens_seen": 139750820,
      "step": 6487,
      "time_per_iteration": 2.516204595565796
    },
    {
      "auxiliary_loss_clip": 0.01153253,
      "auxiliary_loss_mlp": 0.01026933,
      "balance_loss_clip": 1.04431617,
      "balance_loss_mlp": 1.01979876,
      "epoch": 0.7801358744664222,
      "flos": 26429714881920.0,
      "grad_norm": 1.8289264418598066,
      "language_loss": 0.7812897,
      "learning_rate": 4.85860593047986e-07,
      "loss": 0.80309165,
      "num_input_tokens_seen": 139770885,
      "step": 6488,
      "time_per_iteration": 2.492370367050171
    },
    {
      "auxiliary_loss_clip": 0.01115999,
      "auxiliary_loss_mlp": 0.01024151,
      "balance_loss_clip": 1.03857565,
      "balance_loss_mlp": 1.0173409,
      "epoch": 0.7802561173570612,
      "flos": 26322162583680.0,
      "grad_norm": 2.1004130141492454,
      "language_loss": 0.74782205,
      "learning_rate": 4.853517791569613e-07,
      "loss": 0.76922357,
      "num_input_tokens_seen": 139793065,
      "step": 6489,
      "time_per_iteration": 2.557070016860962
    },
    {
      "auxiliary_loss_clip": 0.01144002,
      "auxiliary_loss_mlp": 0.00762092,
      "balance_loss_clip": 1.04451227,
      "balance_loss_mlp": 1.00033188,
      "epoch": 0.7803763602477004,
      "flos": 40333751596800.0,
      "grad_norm": 1.714467869202251,
      "language_loss": 0.66034502,
      "learning_rate": 4.848431950368684e-07,
      "loss": 0.67940599,
      "num_input_tokens_seen": 139815625,
      "step": 6490,
      "time_per_iteration": 2.636080026626587
    },
    {
      "auxiliary_loss_clip": 0.01065659,
      "auxiliary_loss_mlp": 0.00752864,
      "balance_loss_clip": 1.01103115,
      "balance_loss_mlp": 0.99989241,
      "epoch": 0.7804966031383395,
      "flos": 67001448038400.0,
      "grad_norm": 0.712297458260766,
      "language_loss": 0.55715853,
      "learning_rate": 4.843348407648569e-07,
      "loss": 0.57534379,
      "num_input_tokens_seen": 139876905,
      "step": 6491,
      "time_per_iteration": 2.953810453414917
    },
    {
      "auxiliary_loss_clip": 0.01153366,
      "auxiliary_loss_mlp": 0.01024581,
      "balance_loss_clip": 1.04263616,
      "balance_loss_mlp": 1.01674891,
      "epoch": 0.7806168460289785,
      "flos": 17740733057280.0,
      "grad_norm": 2.377935191854429,
      "language_loss": 0.82714969,
      "learning_rate": 4.838267164180457e-07,
      "loss": 0.84892911,
      "num_input_tokens_seen": 139892575,
      "step": 6492,
      "time_per_iteration": 2.399416208267212
    },
    {
      "auxiliary_loss_clip": 0.01168954,
      "auxiliary_loss_mlp": 0.01024881,
      "balance_loss_clip": 1.04785323,
      "balance_loss_mlp": 1.01714444,
      "epoch": 0.7807370889196176,
      "flos": 23946240545280.0,
      "grad_norm": 2.0274687536790608,
      "language_loss": 0.83741719,
      "learning_rate": 4.833188220735156e-07,
      "loss": 0.85935557,
      "num_input_tokens_seen": 139912245,
      "step": 6493,
      "time_per_iteration": 2.419259548187256
    },
    {
      "auxiliary_loss_clip": 0.01151573,
      "auxiliary_loss_mlp": 0.0102186,
      "balance_loss_clip": 1.046157,
      "balance_loss_mlp": 1.01484728,
      "epoch": 0.7808573318102567,
      "flos": 18989024457600.0,
      "grad_norm": 2.023430002560144,
      "language_loss": 0.74718851,
      "learning_rate": 4.828111578083152e-07,
      "loss": 0.76892281,
      "num_input_tokens_seen": 139929150,
      "step": 6494,
      "time_per_iteration": 2.4137871265411377
    },
    {
      "auxiliary_loss_clip": 0.0113444,
      "auxiliary_loss_mlp": 0.01026902,
      "balance_loss_clip": 1.0453341,
      "balance_loss_mlp": 1.01955867,
      "epoch": 0.7809775747008958,
      "flos": 23980750536960.0,
      "grad_norm": 2.6199241854478497,
      "language_loss": 0.8170386,
      "learning_rate": 4.823037236994556e-07,
      "loss": 0.83865201,
      "num_input_tokens_seen": 139947315,
      "step": 6495,
      "time_per_iteration": 2.527867078781128
    },
    {
      "auxiliary_loss_clip": 0.01056259,
      "auxiliary_loss_mlp": 0.01000758,
      "balance_loss_clip": 1.01057625,
      "balance_loss_mlp": 0.99972039,
      "epoch": 0.7810978175915348,
      "flos": 68535875180160.0,
      "grad_norm": 0.7132369122011739,
      "language_loss": 0.56343669,
      "learning_rate": 4.817965198239136e-07,
      "loss": 0.58400691,
      "num_input_tokens_seen": 140013775,
      "step": 6496,
      "time_per_iteration": 3.8623228073120117
    },
    {
      "auxiliary_loss_clip": 0.0112264,
      "auxiliary_loss_mlp": 0.01025794,
      "balance_loss_clip": 1.04158223,
      "balance_loss_mlp": 1.01798308,
      "epoch": 0.781218060482174,
      "flos": 19642131498240.0,
      "grad_norm": 2.1266018211535296,
      "language_loss": 0.74443281,
      "learning_rate": 4.812895462586331e-07,
      "loss": 0.76591718,
      "num_input_tokens_seen": 140031600,
      "step": 6497,
      "time_per_iteration": 3.2123382091522217
    },
    {
      "auxiliary_loss_clip": 0.01127468,
      "auxiliary_loss_mlp": 0.01022041,
      "balance_loss_clip": 1.04589307,
      "balance_loss_mlp": 1.0156728,
      "epoch": 0.7813383033728131,
      "flos": 25627865621760.0,
      "grad_norm": 1.7443390149669613,
      "language_loss": 0.82332885,
      "learning_rate": 4.807828030805207e-07,
      "loss": 0.8448239,
      "num_input_tokens_seen": 140050590,
      "step": 6498,
      "time_per_iteration": 3.3967041969299316
    },
    {
      "auxiliary_loss_clip": 0.01151899,
      "auxiliary_loss_mlp": 0.01033252,
      "balance_loss_clip": 1.04835701,
      "balance_loss_mlp": 1.02563453,
      "epoch": 0.7814585462634521,
      "flos": 20485924865280.0,
      "grad_norm": 1.7733321957795256,
      "language_loss": 0.68363786,
      "learning_rate": 4.802762903664495e-07,
      "loss": 0.70548934,
      "num_input_tokens_seen": 140069770,
      "step": 6499,
      "time_per_iteration": 2.540027618408203
    },
    {
      "auxiliary_loss_clip": 0.01146462,
      "auxiliary_loss_mlp": 0.01028036,
      "balance_loss_clip": 1.04868495,
      "balance_loss_mlp": 1.02036452,
      "epoch": 0.7815787891540913,
      "flos": 22304297018880.0,
      "grad_norm": 2.278488497321996,
      "language_loss": 0.73927927,
      "learning_rate": 4.797700081932565e-07,
      "loss": 0.76102424,
      "num_input_tokens_seen": 140087635,
      "step": 6500,
      "time_per_iteration": 2.468290328979492
    },
    {
      "auxiliary_loss_clip": 0.01089879,
      "auxiliary_loss_mlp": 0.01025554,
      "balance_loss_clip": 1.03757548,
      "balance_loss_mlp": 1.01852059,
      "epoch": 0.7816990320447303,
      "flos": 22600668136320.0,
      "grad_norm": 2.353676989737286,
      "language_loss": 0.82056361,
      "learning_rate": 4.792639566377442e-07,
      "loss": 0.84171796,
      "num_input_tokens_seen": 140105045,
      "step": 6501,
      "time_per_iteration": 2.5547449588775635
    },
    {
      "auxiliary_loss_clip": 0.01146882,
      "auxiliary_loss_mlp": 0.01020525,
      "balance_loss_clip": 1.04369044,
      "balance_loss_mlp": 1.01303852,
      "epoch": 0.7818192749353694,
      "flos": 24935974871040.0,
      "grad_norm": 1.7143268110907315,
      "language_loss": 0.77669728,
      "learning_rate": 4.78758135776681e-07,
      "loss": 0.79837132,
      "num_input_tokens_seen": 140124900,
      "step": 6502,
      "time_per_iteration": 3.249781608581543
    },
    {
      "auxiliary_loss_clip": 0.01139391,
      "auxiliary_loss_mlp": 0.0102537,
      "balance_loss_clip": 1.04566169,
      "balance_loss_mlp": 1.0186348,
      "epoch": 0.7819395178260086,
      "flos": 23733039369600.0,
      "grad_norm": 2.052509956360546,
      "language_loss": 0.78826964,
      "learning_rate": 4.782525456867989e-07,
      "loss": 0.80991721,
      "num_input_tokens_seen": 140143755,
      "step": 6503,
      "time_per_iteration": 2.4841930866241455
    },
    {
      "auxiliary_loss_clip": 0.01125685,
      "auxiliary_loss_mlp": 0.01025145,
      "balance_loss_clip": 1.04330504,
      "balance_loss_mlp": 1.01759315,
      "epoch": 0.7820597607166476,
      "flos": 23221671396480.0,
      "grad_norm": 2.2475392216689696,
      "language_loss": 0.83056802,
      "learning_rate": 4.777471864447959e-07,
      "loss": 0.85207629,
      "num_input_tokens_seen": 140164495,
      "step": 6504,
      "time_per_iteration": 2.5042660236358643
    },
    {
      "auxiliary_loss_clip": 0.01139824,
      "auxiliary_loss_mlp": 0.01029602,
      "balance_loss_clip": 1.04402256,
      "balance_loss_mlp": 1.02259493,
      "epoch": 0.7821800036072867,
      "flos": 22309540404480.0,
      "grad_norm": 2.289048015419642,
      "language_loss": 0.80342984,
      "learning_rate": 4.772420581273344e-07,
      "loss": 0.82512403,
      "num_input_tokens_seen": 140181980,
      "step": 6505,
      "time_per_iteration": 2.4771053791046143
    },
    {
      "auxiliary_loss_clip": 0.01148002,
      "auxiliary_loss_mlp": 0.01022983,
      "balance_loss_clip": 1.0461762,
      "balance_loss_mlp": 1.01564002,
      "epoch": 0.7823002464979258,
      "flos": 21544176384000.0,
      "grad_norm": 1.8111270710972136,
      "language_loss": 0.75861347,
      "learning_rate": 4.7673716081104134e-07,
      "loss": 0.78032333,
      "num_input_tokens_seen": 140202155,
      "step": 6506,
      "time_per_iteration": 2.4428014755249023
    },
    {
      "auxiliary_loss_clip": 0.01153807,
      "auxiliary_loss_mlp": 0.01025059,
      "balance_loss_clip": 1.04919577,
      "balance_loss_mlp": 1.01808512,
      "epoch": 0.7824204893885649,
      "flos": 24535642815360.0,
      "grad_norm": 1.6172844939184783,
      "language_loss": 0.84568745,
      "learning_rate": 4.762324945725109e-07,
      "loss": 0.86747611,
      "num_input_tokens_seen": 140221600,
      "step": 6507,
      "time_per_iteration": 2.466966152191162
    },
    {
      "auxiliary_loss_clip": 0.01134125,
      "auxiliary_loss_mlp": 0.01026296,
      "balance_loss_clip": 1.0478878,
      "balance_loss_mlp": 1.0193938,
      "epoch": 0.782540732279204,
      "flos": 27415211402880.0,
      "grad_norm": 1.6534168662577466,
      "language_loss": 0.75591129,
      "learning_rate": 4.7572805948829844e-07,
      "loss": 0.77751541,
      "num_input_tokens_seen": 140241860,
      "step": 6508,
      "time_per_iteration": 2.501706123352051
    },
    {
      "auxiliary_loss_clip": 0.01115915,
      "auxiliary_loss_mlp": 0.01021402,
      "balance_loss_clip": 1.04416871,
      "balance_loss_mlp": 1.01469302,
      "epoch": 0.7826609751698431,
      "flos": 24353216616960.0,
      "grad_norm": 2.0298483841624435,
      "language_loss": 0.71122849,
      "learning_rate": 4.7522385563492795e-07,
      "loss": 0.7326017,
      "num_input_tokens_seen": 140262160,
      "step": 6509,
      "time_per_iteration": 2.5437140464782715
    },
    {
      "auxiliary_loss_clip": 0.01131003,
      "auxiliary_loss_mlp": 0.01025451,
      "balance_loss_clip": 1.0479269,
      "balance_loss_mlp": 1.01852536,
      "epoch": 0.7827812180604822,
      "flos": 23988543788160.0,
      "grad_norm": 1.8421312027599637,
      "language_loss": 0.70350665,
      "learning_rate": 4.747198830888863e-07,
      "loss": 0.72507119,
      "num_input_tokens_seen": 140282030,
      "step": 6510,
      "time_per_iteration": 2.5157268047332764
    },
    {
      "auxiliary_loss_clip": 0.01133541,
      "auxiliary_loss_mlp": 0.01026722,
      "balance_loss_clip": 1.04472017,
      "balance_loss_mlp": 1.01967096,
      "epoch": 0.7829014609511212,
      "flos": 27454318335360.0,
      "grad_norm": 2.8099577356954786,
      "language_loss": 0.68423414,
      "learning_rate": 4.742161419266251e-07,
      "loss": 0.70583677,
      "num_input_tokens_seen": 140301190,
      "step": 6511,
      "time_per_iteration": 2.5029919147491455
    },
    {
      "auxiliary_loss_clip": 0.0115644,
      "auxiliary_loss_mlp": 0.01027691,
      "balance_loss_clip": 1.04618263,
      "balance_loss_mlp": 1.01999009,
      "epoch": 0.7830217038417604,
      "flos": 29204532432000.0,
      "grad_norm": 3.2335599872123586,
      "language_loss": 0.64929855,
      "learning_rate": 4.7371263222456304e-07,
      "loss": 0.67113984,
      "num_input_tokens_seen": 140318510,
      "step": 6512,
      "time_per_iteration": 2.479785680770874
    },
    {
      "auxiliary_loss_clip": 0.01051348,
      "auxiliary_loss_mlp": 0.01001284,
      "balance_loss_clip": 1.01059556,
      "balance_loss_mlp": 1.00031281,
      "epoch": 0.7831419467323995,
      "flos": 60950895822720.0,
      "grad_norm": 0.8085924692440947,
      "language_loss": 0.61353695,
      "learning_rate": 4.7320935405908004e-07,
      "loss": 0.63406324,
      "num_input_tokens_seen": 140379380,
      "step": 6513,
      "time_per_iteration": 2.9893112182617188
    },
    {
      "auxiliary_loss_clip": 0.01168898,
      "auxiliary_loss_mlp": 0.01025646,
      "balance_loss_clip": 1.04777169,
      "balance_loss_mlp": 1.01790285,
      "epoch": 0.7832621896230385,
      "flos": 19682531320320.0,
      "grad_norm": 2.0746706653203,
      "language_loss": 0.84062672,
      "learning_rate": 4.7270630750652475e-07,
      "loss": 0.86257219,
      "num_input_tokens_seen": 140395335,
      "step": 6514,
      "time_per_iteration": 2.3778579235076904
    },
    {
      "auxiliary_loss_clip": 0.01149626,
      "auxiliary_loss_mlp": 0.01020532,
      "balance_loss_clip": 1.04434681,
      "balance_loss_mlp": 1.01382351,
      "epoch": 0.7833824325136777,
      "flos": 25009232659200.0,
      "grad_norm": 1.6879473377035084,
      "language_loss": 0.80741167,
      "learning_rate": 4.7220349264320746e-07,
      "loss": 0.82911325,
      "num_input_tokens_seen": 140414420,
      "step": 6515,
      "time_per_iteration": 2.4611709117889404
    },
    {
      "auxiliary_loss_clip": 0.01054559,
      "auxiliary_loss_mlp": 0.01000669,
      "balance_loss_clip": 1.01051331,
      "balance_loss_mlp": 0.99963778,
      "epoch": 0.7835026754043167,
      "flos": 68800142517120.0,
      "grad_norm": 0.7329849561181503,
      "language_loss": 0.54916102,
      "learning_rate": 4.71700909545407e-07,
      "loss": 0.56971329,
      "num_input_tokens_seen": 140477365,
      "step": 6516,
      "time_per_iteration": 3.0265395641326904
    },
    {
      "auxiliary_loss_clip": 0.01154139,
      "auxiliary_loss_mlp": 0.01021434,
      "balance_loss_clip": 1.04637933,
      "balance_loss_mlp": 1.01434982,
      "epoch": 0.7836229182949558,
      "flos": 19864598382720.0,
      "grad_norm": 1.8738135362949782,
      "language_loss": 0.77109027,
      "learning_rate": 4.711985582893627e-07,
      "loss": 0.79284596,
      "num_input_tokens_seen": 140495885,
      "step": 6517,
      "time_per_iteration": 2.413577079772949
    },
    {
      "auxiliary_loss_clip": 0.01112179,
      "auxiliary_loss_mlp": 0.01021841,
      "balance_loss_clip": 1.04055035,
      "balance_loss_mlp": 1.0145036,
      "epoch": 0.783743161185595,
      "flos": 22965843755520.0,
      "grad_norm": 1.6402273495031856,
      "language_loss": 0.71610463,
      "learning_rate": 4.706964389512811e-07,
      "loss": 0.73744488,
      "num_input_tokens_seen": 140515920,
      "step": 6518,
      "time_per_iteration": 2.5515036582946777
    },
    {
      "auxiliary_loss_clip": 0.01166309,
      "auxiliary_loss_mlp": 0.01020135,
      "balance_loss_clip": 1.04967618,
      "balance_loss_mlp": 1.01331663,
      "epoch": 0.783863404076234,
      "flos": 12458489777280.0,
      "grad_norm": 1.9073058921881056,
      "language_loss": 0.87306106,
      "learning_rate": 4.701945516073345e-07,
      "loss": 0.89492559,
      "num_input_tokens_seen": 140533395,
      "step": 6519,
      "time_per_iteration": 2.369558572769165
    },
    {
      "auxiliary_loss_clip": 0.01122572,
      "auxiliary_loss_mlp": 0.01020777,
      "balance_loss_clip": 1.04431868,
      "balance_loss_mlp": 1.01414037,
      "epoch": 0.7839836469668731,
      "flos": 24243940465920.0,
      "grad_norm": 1.7950173901195265,
      "language_loss": 0.75393915,
      "learning_rate": 4.696928963336577e-07,
      "loss": 0.77537262,
      "num_input_tokens_seen": 140552825,
      "step": 6520,
      "time_per_iteration": 2.5475728511810303
    },
    {
      "auxiliary_loss_clip": 0.01051419,
      "auxiliary_loss_mlp": 0.01001331,
      "balance_loss_clip": 1.01031482,
      "balance_loss_mlp": 1.00031209,
      "epoch": 0.7841038898575122,
      "flos": 62121978938880.0,
      "grad_norm": 0.8534967495046033,
      "language_loss": 0.61017454,
      "learning_rate": 4.6919147320635224e-07,
      "loss": 0.63070214,
      "num_input_tokens_seen": 140615535,
      "step": 6521,
      "time_per_iteration": 2.9999747276306152
    },
    {
      "auxiliary_loss_clip": 0.01154218,
      "auxiliary_loss_mlp": 0.01024606,
      "balance_loss_clip": 1.04555976,
      "balance_loss_mlp": 1.01783764,
      "epoch": 0.7842241327481513,
      "flos": 20193899293440.0,
      "grad_norm": 2.342963454017732,
      "language_loss": 0.72994047,
      "learning_rate": 4.6869028230148286e-07,
      "loss": 0.75172877,
      "num_input_tokens_seen": 140633330,
      "step": 6522,
      "time_per_iteration": 3.2662525177001953
    },
    {
      "auxiliary_loss_clip": 0.01118504,
      "auxiliary_loss_mlp": 0.01024257,
      "balance_loss_clip": 1.04025996,
      "balance_loss_mlp": 1.01652002,
      "epoch": 0.7843443756387903,
      "flos": 28074531496320.0,
      "grad_norm": 2.34381135662973,
      "language_loss": 0.59847081,
      "learning_rate": 4.6818932369507957e-07,
      "loss": 0.61989844,
      "num_input_tokens_seen": 140652830,
      "step": 6523,
      "time_per_iteration": 3.2541370391845703
    },
    {
      "auxiliary_loss_clip": 0.01153993,
      "auxiliary_loss_mlp": 0.01026472,
      "balance_loss_clip": 1.04981923,
      "balance_loss_mlp": 1.01931953,
      "epoch": 0.7844646185294295,
      "flos": 21323397438720.0,
      "grad_norm": 2.20927760909072,
      "language_loss": 0.88968349,
      "learning_rate": 4.676885974631386e-07,
      "loss": 0.91148812,
      "num_input_tokens_seen": 140671190,
      "step": 6524,
      "time_per_iteration": 2.43854022026062
    },
    {
      "auxiliary_loss_clip": 0.01153788,
      "auxiliary_loss_mlp": 0.01022298,
      "balance_loss_clip": 1.04735041,
      "balance_loss_mlp": 1.01559854,
      "epoch": 0.7845848614200686,
      "flos": 23656585271040.0,
      "grad_norm": 1.9212038267190354,
      "language_loss": 0.81113577,
      "learning_rate": 4.67188103681619e-07,
      "loss": 0.83289659,
      "num_input_tokens_seen": 140690975,
      "step": 6525,
      "time_per_iteration": 3.256356716156006
    },
    {
      "auxiliary_loss_clip": 0.01152803,
      "auxiliary_loss_mlp": 0.00761716,
      "balance_loss_clip": 1.05069482,
      "balance_loss_mlp": 1.00032806,
      "epoch": 0.7847051043107076,
      "flos": 23402194174080.0,
      "grad_norm": 4.33744629859235,
      "language_loss": 0.6875186,
      "learning_rate": 4.666878424264453e-07,
      "loss": 0.70666373,
      "num_input_tokens_seen": 140710930,
      "step": 6526,
      "time_per_iteration": 2.444742441177368
    },
    {
      "auxiliary_loss_clip": 0.01130504,
      "auxiliary_loss_mlp": 0.01019782,
      "balance_loss_clip": 1.0446167,
      "balance_loss_mlp": 1.01362491,
      "epoch": 0.7848253472013467,
      "flos": 19022277473280.0,
      "grad_norm": 1.8657727553705894,
      "language_loss": 0.73486632,
      "learning_rate": 4.661878137735069e-07,
      "loss": 0.75636923,
      "num_input_tokens_seen": 140729120,
      "step": 6527,
      "time_per_iteration": 2.444528102874756
    },
    {
      "auxiliary_loss_clip": 0.01137561,
      "auxiliary_loss_mlp": 0.01021878,
      "balance_loss_clip": 1.04501319,
      "balance_loss_mlp": 1.01512766,
      "epoch": 0.7849455900919858,
      "flos": 21179180332800.0,
      "grad_norm": 2.172175631590525,
      "language_loss": 0.74662203,
      "learning_rate": 4.656880177986571e-07,
      "loss": 0.76821643,
      "num_input_tokens_seen": 140747665,
      "step": 6528,
      "time_per_iteration": 2.4619085788726807
    },
    {
      "auxiliary_loss_clip": 0.01140805,
      "auxiliary_loss_mlp": 0.01024067,
      "balance_loss_clip": 1.04357326,
      "balance_loss_mlp": 1.0165273,
      "epoch": 0.7850658329826249,
      "flos": 19536482620800.0,
      "grad_norm": 1.8828479089549728,
      "language_loss": 0.81276512,
      "learning_rate": 4.6518845457771607e-07,
      "loss": 0.83441389,
      "num_input_tokens_seen": 140766525,
      "step": 6529,
      "time_per_iteration": 3.2345666885375977
    },
    {
      "auxiliary_loss_clip": 0.01144647,
      "auxiliary_loss_mlp": 0.00761501,
      "balance_loss_clip": 1.04440784,
      "balance_loss_mlp": 1.00037599,
      "epoch": 0.7851860758732639,
      "flos": 12495334152960.0,
      "grad_norm": 1.7715898218398218,
      "language_loss": 0.78886473,
      "learning_rate": 4.646891241864652e-07,
      "loss": 0.8079263,
      "num_input_tokens_seen": 140785090,
      "step": 6530,
      "time_per_iteration": 2.420994520187378
    },
    {
      "auxiliary_loss_clip": 0.01151563,
      "auxiliary_loss_mlp": 0.01027857,
      "balance_loss_clip": 1.04540133,
      "balance_loss_mlp": 1.02001297,
      "epoch": 0.7853063187639031,
      "flos": 22960959505920.0,
      "grad_norm": 2.0034811796725855,
      "language_loss": 0.73514444,
      "learning_rate": 4.6419002670065397e-07,
      "loss": 0.75693864,
      "num_input_tokens_seen": 140804670,
      "step": 6531,
      "time_per_iteration": 2.4273664951324463
    },
    {
      "auxiliary_loss_clip": 0.01129057,
      "auxiliary_loss_mlp": 0.01028411,
      "balance_loss_clip": 1.04596663,
      "balance_loss_mlp": 1.02072835,
      "epoch": 0.7854265616545422,
      "flos": 17347260499200.0,
      "grad_norm": 3.4046561018796853,
      "language_loss": 0.86753833,
      "learning_rate": 4.6369116219599445e-07,
      "loss": 0.88911295,
      "num_input_tokens_seen": 140820655,
      "step": 6532,
      "time_per_iteration": 2.481265068054199
    },
    {
      "auxiliary_loss_clip": 0.01123649,
      "auxiliary_loss_mlp": 0.0102251,
      "balance_loss_clip": 1.04358661,
      "balance_loss_mlp": 1.01591802,
      "epoch": 0.7855468045451812,
      "flos": 23838293197440.0,
      "grad_norm": 1.657003108576288,
      "language_loss": 0.79092884,
      "learning_rate": 4.631925307481637e-07,
      "loss": 0.81239045,
      "num_input_tokens_seen": 140840470,
      "step": 6533,
      "time_per_iteration": 2.51924467086792
    },
    {
      "auxiliary_loss_clip": 0.01137392,
      "auxiliary_loss_mlp": 0.01022508,
      "balance_loss_clip": 1.04688919,
      "balance_loss_mlp": 1.01583219,
      "epoch": 0.7856670474358204,
      "flos": 25666792986240.0,
      "grad_norm": 1.94763198897157,
      "language_loss": 0.75912499,
      "learning_rate": 4.6269413243280533e-07,
      "loss": 0.78072405,
      "num_input_tokens_seen": 140859890,
      "step": 6534,
      "time_per_iteration": 2.507854461669922
    },
    {
      "auxiliary_loss_clip": 0.01146527,
      "auxiliary_loss_mlp": 0.0102237,
      "balance_loss_clip": 1.05017853,
      "balance_loss_mlp": 1.01415014,
      "epoch": 0.7857872903264594,
      "flos": 18144656472960.0,
      "grad_norm": 2.7598073972263037,
      "language_loss": 0.74365258,
      "learning_rate": 4.621959673255236e-07,
      "loss": 0.76534152,
      "num_input_tokens_seen": 140876190,
      "step": 6535,
      "time_per_iteration": 2.435556411743164
    },
    {
      "auxiliary_loss_clip": 0.01107208,
      "auxiliary_loss_mlp": 0.01026101,
      "balance_loss_clip": 1.04113913,
      "balance_loss_mlp": 1.01904368,
      "epoch": 0.7859075332170985,
      "flos": 14386138081920.0,
      "grad_norm": 2.0730766876488294,
      "language_loss": 0.90412748,
      "learning_rate": 4.6169803550189135e-07,
      "loss": 0.92546058,
      "num_input_tokens_seen": 140891885,
      "step": 6536,
      "time_per_iteration": 2.4852495193481445
    },
    {
      "auxiliary_loss_clip": 0.01104958,
      "auxiliary_loss_mlp": 0.01028422,
      "balance_loss_clip": 1.04470956,
      "balance_loss_mlp": 1.02053666,
      "epoch": 0.7860277761077377,
      "flos": 19864059678720.0,
      "grad_norm": 2.048396562476462,
      "language_loss": 0.77154273,
      "learning_rate": 4.6120033703744355e-07,
      "loss": 0.7928766,
      "num_input_tokens_seen": 140910780,
      "step": 6537,
      "time_per_iteration": 2.513505697250366
    },
    {
      "auxiliary_loss_clip": 0.01126414,
      "auxiliary_loss_mlp": 0.0102377,
      "balance_loss_clip": 1.04220009,
      "balance_loss_mlp": 1.01680577,
      "epoch": 0.7861480189983767,
      "flos": 26396174557440.0,
      "grad_norm": 1.7797252122112697,
      "language_loss": 0.78226602,
      "learning_rate": 4.607028720076822e-07,
      "loss": 0.80376792,
      "num_input_tokens_seen": 140927460,
      "step": 6538,
      "time_per_iteration": 2.5055649280548096
    },
    {
      "auxiliary_loss_clip": 0.01152233,
      "auxiliary_loss_mlp": 0.01024347,
      "balance_loss_clip": 1.04643404,
      "balance_loss_mlp": 1.0172925,
      "epoch": 0.7862682618890158,
      "flos": 24236578177920.0,
      "grad_norm": 1.882501517667879,
      "language_loss": 0.73534894,
      "learning_rate": 4.6020564048807074e-07,
      "loss": 0.75711471,
      "num_input_tokens_seen": 140945135,
      "step": 6539,
      "time_per_iteration": 2.4453020095825195
    },
    {
      "auxiliary_loss_clip": 0.01156743,
      "auxiliary_loss_mlp": 0.01024812,
      "balance_loss_clip": 1.04839015,
      "balance_loss_mlp": 1.01742077,
      "epoch": 0.7863885047796549,
      "flos": 47551508259840.0,
      "grad_norm": 2.340499658063534,
      "language_loss": 0.71783292,
      "learning_rate": 4.5970864255403883e-07,
      "loss": 0.73964846,
      "num_input_tokens_seen": 140966660,
      "step": 6540,
      "time_per_iteration": 2.655489444732666
    },
    {
      "auxiliary_loss_clip": 0.01140207,
      "auxiliary_loss_mlp": 0.01022249,
      "balance_loss_clip": 1.04365444,
      "balance_loss_mlp": 1.01574016,
      "epoch": 0.786508747670294,
      "flos": 24389234979840.0,
      "grad_norm": 1.8667333097532215,
      "language_loss": 0.81985426,
      "learning_rate": 4.59211878280982e-07,
      "loss": 0.84147882,
      "num_input_tokens_seen": 140986175,
      "step": 6541,
      "time_per_iteration": 2.4602620601654053
    },
    {
      "auxiliary_loss_clip": 0.01138426,
      "auxiliary_loss_mlp": 0.01021688,
      "balance_loss_clip": 1.04481506,
      "balance_loss_mlp": 1.0146513,
      "epoch": 0.786628990560933,
      "flos": 18041234238720.0,
      "grad_norm": 2.638287957537487,
      "language_loss": 0.70001912,
      "learning_rate": 4.587153477442578e-07,
      "loss": 0.72162032,
      "num_input_tokens_seen": 141002490,
      "step": 6542,
      "time_per_iteration": 2.4550085067749023
    },
    {
      "auxiliary_loss_clip": 0.01169613,
      "auxiliary_loss_mlp": 0.01025984,
      "balance_loss_clip": 1.04906929,
      "balance_loss_mlp": 1.01834321,
      "epoch": 0.7867492334515722,
      "flos": 25848860048640.0,
      "grad_norm": 3.045472342088166,
      "language_loss": 0.81581938,
      "learning_rate": 4.582190510191899e-07,
      "loss": 0.83777535,
      "num_input_tokens_seen": 141021150,
      "step": 6543,
      "time_per_iteration": 2.4252097606658936
    },
    {
      "auxiliary_loss_clip": 0.01121494,
      "auxiliary_loss_mlp": 0.01024218,
      "balance_loss_clip": 1.04519296,
      "balance_loss_mlp": 1.01729512,
      "epoch": 0.7868694763422113,
      "flos": 16580819070720.0,
      "grad_norm": 2.0068830520879373,
      "language_loss": 0.87285471,
      "learning_rate": 4.5772298818106625e-07,
      "loss": 0.89431179,
      "num_input_tokens_seen": 141036940,
      "step": 6544,
      "time_per_iteration": 2.4546384811401367
    },
    {
      "auxiliary_loss_clip": 0.01133911,
      "auxiliary_loss_mlp": 0.01025222,
      "balance_loss_clip": 1.04806626,
      "balance_loss_mlp": 1.01776505,
      "epoch": 0.7869897192328503,
      "flos": 29386276272000.0,
      "grad_norm": 2.2657619764442587,
      "language_loss": 0.71594775,
      "learning_rate": 4.572271593051384e-07,
      "loss": 0.73753911,
      "num_input_tokens_seen": 141054295,
      "step": 6545,
      "time_per_iteration": 2.540421962738037
    },
    {
      "auxiliary_loss_clip": 0.01105133,
      "auxiliary_loss_mlp": 0.0102248,
      "balance_loss_clip": 1.04405975,
      "balance_loss_mlp": 1.01569057,
      "epoch": 0.7871099621234895,
      "flos": 17128923678720.0,
      "grad_norm": 1.7652684842074522,
      "language_loss": 0.78288555,
      "learning_rate": 4.567315644666245e-07,
      "loss": 0.80416167,
      "num_input_tokens_seen": 141073090,
      "step": 6546,
      "time_per_iteration": 2.508835554122925
    },
    {
      "auxiliary_loss_clip": 0.01120778,
      "auxiliary_loss_mlp": 0.01020691,
      "balance_loss_clip": 1.04563427,
      "balance_loss_mlp": 1.01372063,
      "epoch": 0.7872302050141285,
      "flos": 23440187784960.0,
      "grad_norm": 1.9833915426105033,
      "language_loss": 0.84761536,
      "learning_rate": 4.5623620374070507e-07,
      "loss": 0.86903006,
      "num_input_tokens_seen": 141092405,
      "step": 6547,
      "time_per_iteration": 2.4985992908477783
    },
    {
      "auxiliary_loss_clip": 0.01032081,
      "auxiliary_loss_mlp": 0.01001732,
      "balance_loss_clip": 1.00973082,
      "balance_loss_mlp": 1.00065935,
      "epoch": 0.7873504479047676,
      "flos": 65959752689280.0,
      "grad_norm": 0.7646426487743536,
      "language_loss": 0.58440018,
      "learning_rate": 4.557410772025263e-07,
      "loss": 0.60473835,
      "num_input_tokens_seen": 141154355,
      "step": 6548,
      "time_per_iteration": 3.1714861392974854
    },
    {
      "auxiliary_loss_clip": 0.0113385,
      "auxiliary_loss_mlp": 0.01026978,
      "balance_loss_clip": 1.04336679,
      "balance_loss_mlp": 1.01981091,
      "epoch": 0.7874706907954068,
      "flos": 23258336204160.0,
      "grad_norm": 2.0820165254770107,
      "language_loss": 0.66226792,
      "learning_rate": 4.5524618492719803e-07,
      "loss": 0.68387616,
      "num_input_tokens_seen": 141173575,
      "step": 6549,
      "time_per_iteration": 3.360734462738037
    },
    {
      "auxiliary_loss_clip": 0.01151419,
      "auxiliary_loss_mlp": 0.01022071,
      "balance_loss_clip": 1.04533255,
      "balance_loss_mlp": 1.01524878,
      "epoch": 0.7875909336860458,
      "flos": 28767786963840.0,
      "grad_norm": 1.6711485690248198,
      "language_loss": 0.79356635,
      "learning_rate": 4.54751526989795e-07,
      "loss": 0.8153013,
      "num_input_tokens_seen": 141195415,
      "step": 6550,
      "time_per_iteration": 3.2458090782165527
    },
    {
      "auxiliary_loss_clip": 0.01154321,
      "auxiliary_loss_mlp": 0.01024548,
      "balance_loss_clip": 1.04614449,
      "balance_loss_mlp": 1.01723158,
      "epoch": 0.7877111765766849,
      "flos": 18697286194560.0,
      "grad_norm": 2.0306043773456373,
      "language_loss": 0.79386109,
      "learning_rate": 4.5425710346535775e-07,
      "loss": 0.81564975,
      "num_input_tokens_seen": 141213360,
      "step": 6551,
      "time_per_iteration": 2.397141456604004
    },
    {
      "auxiliary_loss_clip": 0.01153103,
      "auxiliary_loss_mlp": 0.01024801,
      "balance_loss_clip": 1.04562199,
      "balance_loss_mlp": 1.01686788,
      "epoch": 0.787831419467324,
      "flos": 27592968833280.0,
      "grad_norm": 1.9140507309289247,
      "language_loss": 0.81705105,
      "learning_rate": 4.537629144288877e-07,
      "loss": 0.83883017,
      "num_input_tokens_seen": 141230815,
      "step": 6552,
      "time_per_iteration": 3.270242929458618
    },
    {
      "auxiliary_loss_clip": 0.01115908,
      "auxiliary_loss_mlp": 0.01025015,
      "balance_loss_clip": 1.0404191,
      "balance_loss_mlp": 1.01798737,
      "epoch": 0.7879516623579631,
      "flos": 18150187167360.0,
      "grad_norm": 2.170461918890001,
      "language_loss": 0.75071132,
      "learning_rate": 4.5326895995535477e-07,
      "loss": 0.77212059,
      "num_input_tokens_seen": 141249715,
      "step": 6553,
      "time_per_iteration": 2.5500857830047607
    },
    {
      "auxiliary_loss_clip": 0.01147228,
      "auxiliary_loss_mlp": 0.01023523,
      "balance_loss_clip": 1.0442307,
      "balance_loss_mlp": 1.01643324,
      "epoch": 0.7880719052486022,
      "flos": 20339193807360.0,
      "grad_norm": 2.3007570624229223,
      "language_loss": 0.84597278,
      "learning_rate": 4.527752401196907e-07,
      "loss": 0.86768031,
      "num_input_tokens_seen": 141267730,
      "step": 6554,
      "time_per_iteration": 2.426267385482788
    },
    {
      "auxiliary_loss_clip": 0.01133772,
      "auxiliary_loss_mlp": 0.01026927,
      "balance_loss_clip": 1.04474401,
      "balance_loss_mlp": 1.01949465,
      "epoch": 0.7881921481392413,
      "flos": 21653237053440.0,
      "grad_norm": 1.7210816373088333,
      "language_loss": 0.66440421,
      "learning_rate": 4.5228175499679254e-07,
      "loss": 0.6860112,
      "num_input_tokens_seen": 141287315,
      "step": 6555,
      "time_per_iteration": 3.2407355308532715
    },
    {
      "auxiliary_loss_clip": 0.01055194,
      "auxiliary_loss_mlp": 0.01001301,
      "balance_loss_clip": 1.01077652,
      "balance_loss_mlp": 1.00024033,
      "epoch": 0.7883123910298804,
      "flos": 68565860058240.0,
      "grad_norm": 0.8151351645725482,
      "language_loss": 0.54515839,
      "learning_rate": 4.5178850466152174e-07,
      "loss": 0.5657233,
      "num_input_tokens_seen": 141346145,
      "step": 6556,
      "time_per_iteration": 3.0649099349975586
    },
    {
      "auxiliary_loss_clip": 0.0113277,
      "auxiliary_loss_mlp": 0.01023415,
      "balance_loss_clip": 1.04187489,
      "balance_loss_mlp": 1.01644158,
      "epoch": 0.7884326339205194,
      "flos": 19318217627520.0,
      "grad_norm": 1.7963096579371622,
      "language_loss": 0.81914318,
      "learning_rate": 4.512954891887031e-07,
      "loss": 0.84070504,
      "num_input_tokens_seen": 141364445,
      "step": 6557,
      "time_per_iteration": 2.4567461013793945
    },
    {
      "auxiliary_loss_clip": 0.01136199,
      "auxiliary_loss_mlp": 0.01029725,
      "balance_loss_clip": 1.04742813,
      "balance_loss_mlp": 1.02181304,
      "epoch": 0.7885528768111585,
      "flos": 17784903807360.0,
      "grad_norm": 2.2733152396122227,
      "language_loss": 0.83459806,
      "learning_rate": 4.5080270865312806e-07,
      "loss": 0.85625732,
      "num_input_tokens_seen": 141381640,
      "step": 6558,
      "time_per_iteration": 2.43115496635437
    },
    {
      "auxiliary_loss_clip": 0.01150972,
      "auxiliary_loss_mlp": 0.01022017,
      "balance_loss_clip": 1.04605997,
      "balance_loss_mlp": 1.01529968,
      "epoch": 0.7886731197017977,
      "flos": 18807639753600.0,
      "grad_norm": 2.0857913693067376,
      "language_loss": 0.71319795,
      "learning_rate": 4.5031016312954985e-07,
      "loss": 0.73492777,
      "num_input_tokens_seen": 141399955,
      "step": 6559,
      "time_per_iteration": 2.4247934818267822
    },
    {
      "auxiliary_loss_clip": 0.01161901,
      "auxiliary_loss_mlp": 0.01023811,
      "balance_loss_clip": 1.04975438,
      "balance_loss_mlp": 1.0168258,
      "epoch": 0.7887933625924367,
      "flos": 33365358126720.0,
      "grad_norm": 4.1630067051630775,
      "language_loss": 0.74909467,
      "learning_rate": 4.498178526926886e-07,
      "loss": 0.77095181,
      "num_input_tokens_seen": 141420820,
      "step": 6560,
      "time_per_iteration": 2.553723096847534
    },
    {
      "auxiliary_loss_clip": 0.0116526,
      "auxiliary_loss_mlp": 0.01027075,
      "balance_loss_clip": 1.04810011,
      "balance_loss_mlp": 1.02041388,
      "epoch": 0.7889136054830758,
      "flos": 17019360218880.0,
      "grad_norm": 2.758014043534276,
      "language_loss": 0.72061914,
      "learning_rate": 4.4932577741722635e-07,
      "loss": 0.74254251,
      "num_input_tokens_seen": 141439350,
      "step": 6561,
      "time_per_iteration": 2.380889654159546
    },
    {
      "auxiliary_loss_clip": 0.01134743,
      "auxiliary_loss_mlp": 0.01028906,
      "balance_loss_clip": 1.04419732,
      "balance_loss_mlp": 1.02115107,
      "epoch": 0.7890338483737149,
      "flos": 29424629018880.0,
      "grad_norm": 1.9559362193656162,
      "language_loss": 0.74352396,
      "learning_rate": 4.4883393737780985e-07,
      "loss": 0.76516044,
      "num_input_tokens_seen": 141460300,
      "step": 6562,
      "time_per_iteration": 2.524661064147949
    },
    {
      "auxiliary_loss_clip": 0.01146446,
      "auxiliary_loss_mlp": 0.01024985,
      "balance_loss_clip": 1.04456294,
      "balance_loss_mlp": 1.01792789,
      "epoch": 0.789154091264354,
      "flos": 19971576063360.0,
      "grad_norm": 1.922886281588387,
      "language_loss": 0.7861793,
      "learning_rate": 4.4834233264905254e-07,
      "loss": 0.80789363,
      "num_input_tokens_seen": 141477315,
      "step": 6563,
      "time_per_iteration": 2.416424512863159
    },
    {
      "auxiliary_loss_clip": 0.01116282,
      "auxiliary_loss_mlp": 0.0102847,
      "balance_loss_clip": 1.04081166,
      "balance_loss_mlp": 1.0210191,
      "epoch": 0.789274334154993,
      "flos": 14537825216640.0,
      "grad_norm": 2.6226564838199424,
      "language_loss": 0.7143079,
      "learning_rate": 4.478509633055294e-07,
      "loss": 0.73575538,
      "num_input_tokens_seen": 141495025,
      "step": 6564,
      "time_per_iteration": 2.471700668334961
    },
    {
      "auxiliary_loss_clip": 0.01140841,
      "auxiliary_loss_mlp": 0.01028025,
      "balance_loss_clip": 1.04515147,
      "balance_loss_mlp": 1.0203836,
      "epoch": 0.7893945770456322,
      "flos": 21827403123840.0,
      "grad_norm": 3.3680339608858048,
      "language_loss": 0.79994029,
      "learning_rate": 4.473598294217813e-07,
      "loss": 0.82162899,
      "num_input_tokens_seen": 141510450,
      "step": 6565,
      "time_per_iteration": 2.4578559398651123
    },
    {
      "auxiliary_loss_clip": 0.0115129,
      "auxiliary_loss_mlp": 0.01022398,
      "balance_loss_clip": 1.04835749,
      "balance_loss_mlp": 1.01551068,
      "epoch": 0.7895148199362713,
      "flos": 20740639184640.0,
      "grad_norm": 2.0752877753709598,
      "language_loss": 0.71948761,
      "learning_rate": 4.468689310723124e-07,
      "loss": 0.74122441,
      "num_input_tokens_seen": 141528265,
      "step": 6566,
      "time_per_iteration": 2.4193062782287598
    },
    {
      "auxiliary_loss_clip": 0.01127121,
      "auxiliary_loss_mlp": 0.01024212,
      "balance_loss_clip": 1.04337358,
      "balance_loss_mlp": 1.01701736,
      "epoch": 0.7896350628269103,
      "flos": 16690669839360.0,
      "grad_norm": 1.68254047823207,
      "language_loss": 0.78595316,
      "learning_rate": 4.463782683315913e-07,
      "loss": 0.80746651,
      "num_input_tokens_seen": 141547270,
      "step": 6567,
      "time_per_iteration": 2.4683709144592285
    },
    {
      "auxiliary_loss_clip": 0.0116324,
      "auxiliary_loss_mlp": 0.0102462,
      "balance_loss_clip": 1.04682946,
      "balance_loss_mlp": 1.01782513,
      "epoch": 0.7897553057175495,
      "flos": 22638374438400.0,
      "grad_norm": 1.7791749198296285,
      "language_loss": 0.73169696,
      "learning_rate": 4.458878412740523e-07,
      "loss": 0.75357556,
      "num_input_tokens_seen": 141566050,
      "step": 6568,
      "time_per_iteration": 2.400585651397705
    },
    {
      "auxiliary_loss_clip": 0.01147715,
      "auxiliary_loss_mlp": 0.01023001,
      "balance_loss_clip": 1.04556549,
      "balance_loss_mlp": 1.0158428,
      "epoch": 0.7898755486081885,
      "flos": 14537573821440.0,
      "grad_norm": 2.809623088500836,
      "language_loss": 0.77538687,
      "learning_rate": 4.453976499740919e-07,
      "loss": 0.79709399,
      "num_input_tokens_seen": 141583695,
      "step": 6569,
      "time_per_iteration": 2.3946003913879395
    },
    {
      "auxiliary_loss_clip": 0.01148466,
      "auxiliary_loss_mlp": 0.01021776,
      "balance_loss_clip": 1.04703081,
      "balance_loss_mlp": 1.01494503,
      "epoch": 0.7899957914988276,
      "flos": 17238487138560.0,
      "grad_norm": 1.6439343101453636,
      "language_loss": 0.77846432,
      "learning_rate": 4.4490769450607215e-07,
      "loss": 0.80016673,
      "num_input_tokens_seen": 141601320,
      "step": 6570,
      "time_per_iteration": 2.4071273803710938
    },
    {
      "auxiliary_loss_clip": 0.01120218,
      "auxiliary_loss_mlp": 0.0102223,
      "balance_loss_clip": 1.03963912,
      "balance_loss_mlp": 1.01531291,
      "epoch": 0.7901160343894668,
      "flos": 41279351086080.0,
      "grad_norm": 1.8300896281009391,
      "language_loss": 0.72620595,
      "learning_rate": 4.4441797494431845e-07,
      "loss": 0.74763042,
      "num_input_tokens_seen": 141623125,
      "step": 6571,
      "time_per_iteration": 2.662055253982544
    },
    {
      "auxiliary_loss_clip": 0.01149012,
      "auxiliary_loss_mlp": 0.01024046,
      "balance_loss_clip": 1.04602754,
      "balance_loss_mlp": 1.01655662,
      "epoch": 0.7902362772801058,
      "flos": 16837005847680.0,
      "grad_norm": 1.9374060795423802,
      "language_loss": 0.7782433,
      "learning_rate": 4.439284913631207e-07,
      "loss": 0.79997391,
      "num_input_tokens_seen": 141640335,
      "step": 6572,
      "time_per_iteration": 2.397940158843994
    },
    {
      "auxiliary_loss_clip": 0.01127686,
      "auxiliary_loss_mlp": 0.01029301,
      "balance_loss_clip": 1.04582214,
      "balance_loss_mlp": 1.02207732,
      "epoch": 0.7903565201707449,
      "flos": 27125987091840.0,
      "grad_norm": 2.702596699673583,
      "language_loss": 0.83565795,
      "learning_rate": 4.434392438367347e-07,
      "loss": 0.8572278,
      "num_input_tokens_seen": 141659760,
      "step": 6573,
      "time_per_iteration": 2.531533718109131
    },
    {
      "auxiliary_loss_clip": 0.0115622,
      "auxiliary_loss_mlp": 0.0102143,
      "balance_loss_clip": 1.04528213,
      "balance_loss_mlp": 1.01421762,
      "epoch": 0.790476763061384,
      "flos": 31025167142400.0,
      "grad_norm": 2.966522421635397,
      "language_loss": 0.74172604,
      "learning_rate": 4.4295023243937677e-07,
      "loss": 0.76350248,
      "num_input_tokens_seen": 141679965,
      "step": 6574,
      "time_per_iteration": 2.5097243785858154
    },
    {
      "auxiliary_loss_clip": 0.01157439,
      "auxiliary_loss_mlp": 0.01027073,
      "balance_loss_clip": 1.05093598,
      "balance_loss_mlp": 1.01953292,
      "epoch": 0.7905970059520231,
      "flos": 22089084681600.0,
      "grad_norm": 1.7204156251130005,
      "language_loss": 0.80530035,
      "learning_rate": 4.4246145724523123e-07,
      "loss": 0.82714558,
      "num_input_tokens_seen": 141697710,
      "step": 6575,
      "time_per_iteration": 2.4417030811309814
    },
    {
      "auxiliary_loss_clip": 0.01122534,
      "auxiliary_loss_mlp": 0.01023488,
      "balance_loss_clip": 1.04502881,
      "balance_loss_mlp": 1.01640725,
      "epoch": 0.7907172488426621,
      "flos": 20558141159040.0,
      "grad_norm": 2.096361457670146,
      "language_loss": 0.77651459,
      "learning_rate": 4.41972918328444e-07,
      "loss": 0.79797482,
      "num_input_tokens_seen": 141715145,
      "step": 6576,
      "time_per_iteration": 3.3165934085845947
    },
    {
      "auxiliary_loss_clip": 0.01149275,
      "auxiliary_loss_mlp": 0.01029651,
      "balance_loss_clip": 1.0459249,
      "balance_loss_mlp": 1.02227449,
      "epoch": 0.7908374917333013,
      "flos": 30081542901120.0,
      "grad_norm": 3.214441896016704,
      "language_loss": 0.7731806,
      "learning_rate": 4.4148461576312646e-07,
      "loss": 0.79496992,
      "num_input_tokens_seen": 141734810,
      "step": 6577,
      "time_per_iteration": 3.2404873371124268
    },
    {
      "auxiliary_loss_clip": 0.01153834,
      "auxiliary_loss_mlp": 0.01022512,
      "balance_loss_clip": 1.0492245,
      "balance_loss_mlp": 1.01559448,
      "epoch": 0.7909577346239404,
      "flos": 20996359084800.0,
      "grad_norm": 1.3785762635557686,
      "language_loss": 0.74699754,
      "learning_rate": 4.4099654962335343e-07,
      "loss": 0.76876098,
      "num_input_tokens_seen": 141755260,
      "step": 6578,
      "time_per_iteration": 3.3011975288391113
    },
    {
      "auxiliary_loss_clip": 0.01145444,
      "auxiliary_loss_mlp": 0.0102379,
      "balance_loss_clip": 1.04795718,
      "balance_loss_mlp": 1.01679218,
      "epoch": 0.7910779775145794,
      "flos": 26247935128320.0,
      "grad_norm": 1.8128132770653769,
      "language_loss": 0.75138372,
      "learning_rate": 4.405087199831636e-07,
      "loss": 0.77307606,
      "num_input_tokens_seen": 141775500,
      "step": 6579,
      "time_per_iteration": 2.5055079460144043
    },
    {
      "auxiliary_loss_clip": 0.01140659,
      "auxiliary_loss_mlp": 0.00761408,
      "balance_loss_clip": 1.04465985,
      "balance_loss_mlp": 1.00035143,
      "epoch": 0.7911982204052186,
      "flos": 22564434291840.0,
      "grad_norm": 2.102353587808442,
      "language_loss": 0.67287648,
      "learning_rate": 4.400211269165619e-07,
      "loss": 0.69189715,
      "num_input_tokens_seen": 141791955,
      "step": 6580,
      "time_per_iteration": 2.4734010696411133
    },
    {
      "auxiliary_loss_clip": 0.01171218,
      "auxiliary_loss_mlp": 0.01023561,
      "balance_loss_clip": 1.05281544,
      "balance_loss_mlp": 1.01724887,
      "epoch": 0.7913184632958576,
      "flos": 23112538899840.0,
      "grad_norm": 1.4725045682308275,
      "language_loss": 0.7670927,
      "learning_rate": 4.3953377049751416e-07,
      "loss": 0.78904045,
      "num_input_tokens_seen": 141812380,
      "step": 6581,
      "time_per_iteration": 3.1848576068878174
    },
    {
      "auxiliary_loss_clip": 0.01143829,
      "auxiliary_loss_mlp": 0.01025948,
      "balance_loss_clip": 1.04724455,
      "balance_loss_mlp": 1.01895678,
      "epoch": 0.7914387061864967,
      "flos": 12311758719360.0,
      "grad_norm": 2.777868849081666,
      "language_loss": 0.77703995,
      "learning_rate": 4.390466507999537e-07,
      "loss": 0.79873765,
      "num_input_tokens_seen": 141828130,
      "step": 6582,
      "time_per_iteration": 2.41615629196167
    },
    {
      "auxiliary_loss_clip": 0.01120402,
      "auxiliary_loss_mlp": 0.01028229,
      "balance_loss_clip": 1.04160357,
      "balance_loss_mlp": 1.0213058,
      "epoch": 0.7915589490771359,
      "flos": 17603267708160.0,
      "grad_norm": 2.2040057104361357,
      "language_loss": 0.75313091,
      "learning_rate": 4.385597678977748e-07,
      "loss": 0.7746172,
      "num_input_tokens_seen": 141846965,
      "step": 6583,
      "time_per_iteration": 2.477984666824341
    },
    {
      "auxiliary_loss_clip": 0.01138378,
      "auxiliary_loss_mlp": 0.01021882,
      "balance_loss_clip": 1.04486108,
      "balance_loss_mlp": 1.01464927,
      "epoch": 0.7916791919677749,
      "flos": 25591272641280.0,
      "grad_norm": 1.8946348671589661,
      "language_loss": 0.75569862,
      "learning_rate": 4.3807312186483726e-07,
      "loss": 0.77730131,
      "num_input_tokens_seen": 141867685,
      "step": 6584,
      "time_per_iteration": 2.496858596801758
    },
    {
      "auxiliary_loss_clip": 0.0114917,
      "auxiliary_loss_mlp": 0.01023636,
      "balance_loss_clip": 1.04819489,
      "balance_loss_mlp": 1.01640296,
      "epoch": 0.791799434858414,
      "flos": 18844340474880.0,
      "grad_norm": 2.648147981521643,
      "language_loss": 0.78432012,
      "learning_rate": 4.375867127749655e-07,
      "loss": 0.80604827,
      "num_input_tokens_seen": 141885960,
      "step": 6585,
      "time_per_iteration": 2.4113566875457764
    },
    {
      "auxiliary_loss_clip": 0.01128374,
      "auxiliary_loss_mlp": 0.01025166,
      "balance_loss_clip": 1.04844689,
      "balance_loss_mlp": 1.01781094,
      "epoch": 0.7919196777490531,
      "flos": 25812015672960.0,
      "grad_norm": 2.411423776021973,
      "language_loss": 0.6719209,
      "learning_rate": 4.3710054070194744e-07,
      "loss": 0.69345629,
      "num_input_tokens_seen": 141905655,
      "step": 6586,
      "time_per_iteration": 2.531641721725464
    },
    {
      "auxiliary_loss_clip": 0.01168053,
      "auxiliary_loss_mlp": 0.00761989,
      "balance_loss_clip": 1.04852152,
      "balance_loss_mlp": 1.0004313,
      "epoch": 0.7920399206396922,
      "flos": 11947624594560.0,
      "grad_norm": 2.8417938874174347,
      "language_loss": 0.66800284,
      "learning_rate": 4.3661460571953455e-07,
      "loss": 0.68730319,
      "num_input_tokens_seen": 141922390,
      "step": 6587,
      "time_per_iteration": 2.3614377975463867
    },
    {
      "auxiliary_loss_clip": 0.011509,
      "auxiliary_loss_mlp": 0.01020981,
      "balance_loss_clip": 1.04291511,
      "balance_loss_mlp": 1.01420724,
      "epoch": 0.7921601635303313,
      "flos": 21579907438080.0,
      "grad_norm": 1.5125766647095769,
      "language_loss": 0.68470734,
      "learning_rate": 4.36128907901443e-07,
      "loss": 0.70642608,
      "num_input_tokens_seen": 141941985,
      "step": 6588,
      "time_per_iteration": 2.4317331314086914
    },
    {
      "auxiliary_loss_clip": 0.0112554,
      "auxiliary_loss_mlp": 0.0102264,
      "balance_loss_clip": 1.04207182,
      "balance_loss_mlp": 1.01519573,
      "epoch": 0.7922804064209703,
      "flos": 18113989236480.0,
      "grad_norm": 3.0670378583880904,
      "language_loss": 0.72801197,
      "learning_rate": 4.356434473213519e-07,
      "loss": 0.74949372,
      "num_input_tokens_seen": 141959435,
      "step": 6589,
      "time_per_iteration": 2.463629961013794
    },
    {
      "auxiliary_loss_clip": 0.01135478,
      "auxiliary_loss_mlp": 0.01023051,
      "balance_loss_clip": 1.04511428,
      "balance_loss_mlp": 1.01650667,
      "epoch": 0.7924006493116095,
      "flos": 21652806090240.0,
      "grad_norm": 1.6623448842462079,
      "language_loss": 0.79864168,
      "learning_rate": 4.351582240529068e-07,
      "loss": 0.82022691,
      "num_input_tokens_seen": 141980265,
      "step": 6590,
      "time_per_iteration": 2.4719135761260986
    },
    {
      "auxiliary_loss_clip": 0.01048226,
      "auxiliary_loss_mlp": 0.0100223,
      "balance_loss_clip": 1.01133204,
      "balance_loss_mlp": 1.00105023,
      "epoch": 0.7925208922022485,
      "flos": 64242755694720.0,
      "grad_norm": 0.8190078717215722,
      "language_loss": 0.58210039,
      "learning_rate": 4.346732381697149e-07,
      "loss": 0.60260493,
      "num_input_tokens_seen": 142044395,
      "step": 6591,
      "time_per_iteration": 3.0929298400878906
    },
    {
      "auxiliary_loss_clip": 0.01131311,
      "auxiliary_loss_mlp": 0.01025845,
      "balance_loss_clip": 1.0438385,
      "balance_loss_mlp": 1.01848435,
      "epoch": 0.7926411350928876,
      "flos": 16941541403520.0,
      "grad_norm": 2.0617446319843036,
      "language_loss": 0.81260121,
      "learning_rate": 4.3418848974534825e-07,
      "loss": 0.83417273,
      "num_input_tokens_seen": 142061335,
      "step": 6592,
      "time_per_iteration": 2.4242377281188965
    },
    {
      "auxiliary_loss_clip": 0.01129228,
      "auxiliary_loss_mlp": 0.01023377,
      "balance_loss_clip": 1.04533362,
      "balance_loss_mlp": 1.01690149,
      "epoch": 0.7927613779835267,
      "flos": 34459987144320.0,
      "grad_norm": 1.7948372383114815,
      "language_loss": 0.68892765,
      "learning_rate": 4.3370397885334276e-07,
      "loss": 0.71045375,
      "num_input_tokens_seen": 142081965,
      "step": 6593,
      "time_per_iteration": 2.5983049869537354
    },
    {
      "auxiliary_loss_clip": 0.01147252,
      "auxiliary_loss_mlp": 0.01028533,
      "balance_loss_clip": 1.04742765,
      "balance_loss_mlp": 1.02138102,
      "epoch": 0.7928816208741658,
      "flos": 18951174501120.0,
      "grad_norm": 1.8713264344573266,
      "language_loss": 0.75638378,
      "learning_rate": 4.3321970556719777e-07,
      "loss": 0.77814162,
      "num_input_tokens_seen": 142100260,
      "step": 6594,
      "time_per_iteration": 2.4202566146850586
    },
    {
      "auxiliary_loss_clip": 0.01166175,
      "auxiliary_loss_mlp": 0.01027012,
      "balance_loss_clip": 1.04855394,
      "balance_loss_mlp": 1.01957309,
      "epoch": 0.7930018637648049,
      "flos": 18623022825600.0,
      "grad_norm": 2.6766589335515127,
      "language_loss": 0.718279,
      "learning_rate": 4.3273566996037856e-07,
      "loss": 0.74021089,
      "num_input_tokens_seen": 142116955,
      "step": 6595,
      "time_per_iteration": 2.3635802268981934
    },
    {
      "auxiliary_loss_clip": 0.01137154,
      "auxiliary_loss_mlp": 0.01025349,
      "balance_loss_clip": 1.04545259,
      "balance_loss_mlp": 1.01849699,
      "epoch": 0.793122106655444,
      "flos": 24530650824960.0,
      "grad_norm": 1.9060626493223316,
      "language_loss": 0.80084467,
      "learning_rate": 4.322518721063113e-07,
      "loss": 0.82246965,
      "num_input_tokens_seen": 142135505,
      "step": 6596,
      "time_per_iteration": 2.4905996322631836
    },
    {
      "auxiliary_loss_clip": 0.01152518,
      "auxiliary_loss_mlp": 0.01026542,
      "balance_loss_clip": 1.04818416,
      "balance_loss_mlp": 1.01936543,
      "epoch": 0.7932423495460831,
      "flos": 34421203434240.0,
      "grad_norm": 2.032777346926402,
      "language_loss": 0.70500243,
      "learning_rate": 4.3176831207838906e-07,
      "loss": 0.72679299,
      "num_input_tokens_seen": 142158915,
      "step": 6597,
      "time_per_iteration": 2.5546207427978516
    },
    {
      "auxiliary_loss_clip": 0.01151181,
      "auxiliary_loss_mlp": 0.01022238,
      "balance_loss_clip": 1.04924536,
      "balance_loss_mlp": 1.01523197,
      "epoch": 0.7933625924367221,
      "flos": 26980333441920.0,
      "grad_norm": 2.064778182947546,
      "language_loss": 0.74736714,
      "learning_rate": 4.3128498994996685e-07,
      "loss": 0.76910138,
      "num_input_tokens_seen": 142178390,
      "step": 6598,
      "time_per_iteration": 2.48626971244812
    },
    {
      "auxiliary_loss_clip": 0.0115717,
      "auxiliary_loss_mlp": 0.01026676,
      "balance_loss_clip": 1.04825103,
      "balance_loss_mlp": 1.01898718,
      "epoch": 0.7934828353273613,
      "flos": 29568630643200.0,
      "grad_norm": 2.188865164993516,
      "language_loss": 0.71513712,
      "learning_rate": 4.308019057943646e-07,
      "loss": 0.73697561,
      "num_input_tokens_seen": 142200115,
      "step": 6599,
      "time_per_iteration": 2.5224437713623047
    },
    {
      "auxiliary_loss_clip": 0.0111875,
      "auxiliary_loss_mlp": 0.01027022,
      "balance_loss_clip": 1.0437125,
      "balance_loss_mlp": 1.02027774,
      "epoch": 0.7936030782180004,
      "flos": 28615381557120.0,
      "grad_norm": 1.566567151418977,
      "language_loss": 0.74606913,
      "learning_rate": 4.3031905968486535e-07,
      "loss": 0.76752687,
      "num_input_tokens_seen": 142220945,
      "step": 6600,
      "time_per_iteration": 2.617225408554077
    },
    {
      "auxiliary_loss_clip": 0.01106504,
      "auxiliary_loss_mlp": 0.01022823,
      "balance_loss_clip": 1.04422569,
      "balance_loss_mlp": 1.01592636,
      "epoch": 0.7937233211086394,
      "flos": 16392574869120.0,
      "grad_norm": 2.4639726675608062,
      "language_loss": 0.68404102,
      "learning_rate": 4.298364516947162e-07,
      "loss": 0.70533431,
      "num_input_tokens_seen": 142238175,
      "step": 6601,
      "time_per_iteration": 2.509373903274536
    },
    {
      "auxiliary_loss_clip": 0.01105193,
      "auxiliary_loss_mlp": 0.01021577,
      "balance_loss_clip": 1.0403955,
      "balance_loss_mlp": 1.01476693,
      "epoch": 0.7938435639992786,
      "flos": 22013420682240.0,
      "grad_norm": 1.8733447522045494,
      "language_loss": 0.65741765,
      "learning_rate": 4.293540818971295e-07,
      "loss": 0.67868537,
      "num_input_tokens_seen": 142255980,
      "step": 6602,
      "time_per_iteration": 3.3912975788116455
    },
    {
      "auxiliary_loss_clip": 0.01156994,
      "auxiliary_loss_mlp": 0.01018569,
      "balance_loss_clip": 1.04711223,
      "balance_loss_mlp": 1.01138937,
      "epoch": 0.7939638068899176,
      "flos": 22197032029440.0,
      "grad_norm": 2.044910526542651,
      "language_loss": 0.76782644,
      "learning_rate": 4.2887195036527934e-07,
      "loss": 0.78958201,
      "num_input_tokens_seen": 142274785,
      "step": 6603,
      "time_per_iteration": 3.1848556995391846
    },
    {
      "auxiliary_loss_clip": 0.01141112,
      "auxiliary_loss_mlp": 0.01022265,
      "balance_loss_clip": 1.04158592,
      "balance_loss_mlp": 1.01513028,
      "epoch": 0.7940840497805567,
      "flos": 17745186343680.0,
      "grad_norm": 3.092064818015703,
      "language_loss": 0.73251331,
      "learning_rate": 4.28390057172306e-07,
      "loss": 0.75414705,
      "num_input_tokens_seen": 142291290,
      "step": 6604,
      "time_per_iteration": 2.3862674236297607
    },
    {
      "auxiliary_loss_clip": 0.01118427,
      "auxiliary_loss_mlp": 0.01026807,
      "balance_loss_clip": 1.04036212,
      "balance_loss_mlp": 1.01908207,
      "epoch": 0.7942042926711959,
      "flos": 23805435231360.0,
      "grad_norm": 2.348865616828976,
      "language_loss": 0.72159493,
      "learning_rate": 4.279084023913111e-07,
      "loss": 0.74304724,
      "num_input_tokens_seen": 142309165,
      "step": 6605,
      "time_per_iteration": 3.351757049560547
    },
    {
      "auxiliary_loss_clip": 0.01151842,
      "auxiliary_loss_mlp": 0.01023228,
      "balance_loss_clip": 1.04752648,
      "balance_loss_mlp": 1.01621902,
      "epoch": 0.7943245355618349,
      "flos": 19244959839360.0,
      "grad_norm": 1.8509945064574276,
      "language_loss": 0.69608033,
      "learning_rate": 4.2742698609536096e-07,
      "loss": 0.71783102,
      "num_input_tokens_seen": 142327475,
      "step": 6606,
      "time_per_iteration": 2.411729574203491
    },
    {
      "auxiliary_loss_clip": 0.01141759,
      "auxiliary_loss_mlp": 0.01026158,
      "balance_loss_clip": 1.04604197,
      "balance_loss_mlp": 1.01916051,
      "epoch": 0.794444778452474,
      "flos": 25007616547200.0,
      "grad_norm": 1.8525007965634,
      "language_loss": 0.78784835,
      "learning_rate": 4.2694580835748706e-07,
      "loss": 0.80952752,
      "num_input_tokens_seen": 142347335,
      "step": 6607,
      "time_per_iteration": 2.507347583770752
    },
    {
      "auxiliary_loss_clip": 0.01137214,
      "auxiliary_loss_mlp": 0.01024264,
      "balance_loss_clip": 1.04394722,
      "balance_loss_mlp": 1.01677167,
      "epoch": 0.7945650213431131,
      "flos": 23221491828480.0,
      "grad_norm": 2.2480442197891364,
      "language_loss": 0.74014509,
      "learning_rate": 4.264648692506836e-07,
      "loss": 0.76175988,
      "num_input_tokens_seen": 142366125,
      "step": 6608,
      "time_per_iteration": 3.192314624786377
    },
    {
      "auxiliary_loss_clip": 0.01130992,
      "auxiliary_loss_mlp": 0.01025439,
      "balance_loss_clip": 1.04236984,
      "balance_loss_mlp": 1.01782703,
      "epoch": 0.7946852642337522,
      "flos": 26062887237120.0,
      "grad_norm": 1.8070844801424428,
      "language_loss": 0.72065431,
      "learning_rate": 4.2598416884790824e-07,
      "loss": 0.74221861,
      "num_input_tokens_seen": 142385175,
      "step": 6609,
      "time_per_iteration": 2.48479962348938
    },
    {
      "auxiliary_loss_clip": 0.0114658,
      "auxiliary_loss_mlp": 0.01022373,
      "balance_loss_clip": 1.04435027,
      "balance_loss_mlp": 1.0144453,
      "epoch": 0.7948055071243912,
      "flos": 23769704177280.0,
      "grad_norm": 1.8588221875126303,
      "language_loss": 0.81020772,
      "learning_rate": 4.255037072220828e-07,
      "loss": 0.83189726,
      "num_input_tokens_seen": 142406545,
      "step": 6610,
      "time_per_iteration": 2.4791934490203857
    },
    {
      "auxiliary_loss_clip": 0.01163863,
      "auxiliary_loss_mlp": 0.01021872,
      "balance_loss_clip": 1.047575,
      "balance_loss_mlp": 1.01540196,
      "epoch": 0.7949257500150304,
      "flos": 21980814111360.0,
      "grad_norm": 1.6322089106261355,
      "language_loss": 0.72071111,
      "learning_rate": 4.2502348444609293e-07,
      "loss": 0.74256849,
      "num_input_tokens_seen": 142426165,
      "step": 6611,
      "time_per_iteration": 2.393824338912964
    },
    {
      "auxiliary_loss_clip": 0.01107092,
      "auxiliary_loss_mlp": 0.01024829,
      "balance_loss_clip": 1.03946376,
      "balance_loss_mlp": 1.01811194,
      "epoch": 0.7950459929056695,
      "flos": 25774129802880.0,
      "grad_norm": 2.0735636942778717,
      "language_loss": 0.69263506,
      "learning_rate": 4.2454350059278844e-07,
      "loss": 0.71395433,
      "num_input_tokens_seen": 142447225,
      "step": 6612,
      "time_per_iteration": 2.576458692550659
    },
    {
      "auxiliary_loss_clip": 0.0113145,
      "auxiliary_loss_mlp": 0.01022896,
      "balance_loss_clip": 1.04071593,
      "balance_loss_mlp": 1.01608574,
      "epoch": 0.7951662357963085,
      "flos": 22158068751360.0,
      "grad_norm": 1.8668637044677727,
      "language_loss": 0.84584939,
      "learning_rate": 4.240637557349824e-07,
      "loss": 0.8673929,
      "num_input_tokens_seen": 142464440,
      "step": 6613,
      "time_per_iteration": 2.454233169555664
    },
    {
      "auxiliary_loss_clip": 0.01125005,
      "auxiliary_loss_mlp": 0.01024335,
      "balance_loss_clip": 1.04366422,
      "balance_loss_mlp": 1.01713729,
      "epoch": 0.7952864786869477,
      "flos": 24641938137600.0,
      "grad_norm": 1.9281627515450201,
      "language_loss": 0.66895097,
      "learning_rate": 4.235842499454516e-07,
      "loss": 0.69044435,
      "num_input_tokens_seen": 142484355,
      "step": 6614,
      "time_per_iteration": 2.4768826961517334
    },
    {
      "auxiliary_loss_clip": 0.01139714,
      "auxiliary_loss_mlp": 0.01027459,
      "balance_loss_clip": 1.04618716,
      "balance_loss_mlp": 1.02038097,
      "epoch": 0.7954067215775867,
      "flos": 21830922656640.0,
      "grad_norm": 2.052894538503174,
      "language_loss": 0.83102804,
      "learning_rate": 4.2310498329693687e-07,
      "loss": 0.85269976,
      "num_input_tokens_seen": 142505255,
      "step": 6615,
      "time_per_iteration": 2.5034849643707275
    },
    {
      "auxiliary_loss_clip": 0.01155481,
      "auxiliary_loss_mlp": 0.01024506,
      "balance_loss_clip": 1.04739833,
      "balance_loss_mlp": 1.01675487,
      "epoch": 0.7955269644682258,
      "flos": 24060652341120.0,
      "grad_norm": 1.8714757143023533,
      "language_loss": 0.80673945,
      "learning_rate": 4.2262595586214164e-07,
      "loss": 0.82853931,
      "num_input_tokens_seen": 142526350,
      "step": 6616,
      "time_per_iteration": 2.4590518474578857
    },
    {
      "auxiliary_loss_clip": 0.0115553,
      "auxiliary_loss_mlp": 0.01026049,
      "balance_loss_clip": 1.04738343,
      "balance_loss_mlp": 1.01852143,
      "epoch": 0.795647207358865,
      "flos": 25010741030400.0,
      "grad_norm": 1.8369499267244507,
      "language_loss": 0.76683843,
      "learning_rate": 4.221471677137358e-07,
      "loss": 0.78865427,
      "num_input_tokens_seen": 142547165,
      "step": 6617,
      "time_per_iteration": 2.4743330478668213
    },
    {
      "auxiliary_loss_clip": 0.01126113,
      "auxiliary_loss_mlp": 0.01022762,
      "balance_loss_clip": 1.04303741,
      "balance_loss_mlp": 1.01626253,
      "epoch": 0.795767450249504,
      "flos": 14648358343680.0,
      "grad_norm": 1.8995220252151008,
      "language_loss": 0.70113754,
      "learning_rate": 4.216686189243492e-07,
      "loss": 0.72262627,
      "num_input_tokens_seen": 142565955,
      "step": 6618,
      "time_per_iteration": 2.4523582458496094
    },
    {
      "auxiliary_loss_clip": 0.01122303,
      "auxiliary_loss_mlp": 0.01021656,
      "balance_loss_clip": 1.04486299,
      "balance_loss_mlp": 1.01412189,
      "epoch": 0.7958876931401431,
      "flos": 18547897530240.0,
      "grad_norm": 1.8052228019774563,
      "language_loss": 0.72656584,
      "learning_rate": 4.211903095665785e-07,
      "loss": 0.74800545,
      "num_input_tokens_seen": 142585340,
      "step": 6619,
      "time_per_iteration": 2.480369806289673
    },
    {
      "auxiliary_loss_clip": 0.01145166,
      "auxiliary_loss_mlp": 0.01028525,
      "balance_loss_clip": 1.04394758,
      "balance_loss_mlp": 1.02140558,
      "epoch": 0.7960079360307821,
      "flos": 21543960902400.0,
      "grad_norm": 1.917538964622712,
      "language_loss": 0.75246668,
      "learning_rate": 4.2071223971298277e-07,
      "loss": 0.77420354,
      "num_input_tokens_seen": 142602525,
      "step": 6620,
      "time_per_iteration": 2.462979555130005
    },
    {
      "auxiliary_loss_clip": 0.01152625,
      "auxiliary_loss_mlp": 0.01023628,
      "balance_loss_clip": 1.04575241,
      "balance_loss_mlp": 1.01584041,
      "epoch": 0.7961281789214213,
      "flos": 25481745095040.0,
      "grad_norm": 2.0093183304856908,
      "language_loss": 0.61201036,
      "learning_rate": 4.2023440943608433e-07,
      "loss": 0.63377297,
      "num_input_tokens_seen": 142622490,
      "step": 6621,
      "time_per_iteration": 2.4907608032226562
    },
    {
      "auxiliary_loss_clip": 0.011482,
      "auxiliary_loss_mlp": 0.0102103,
      "balance_loss_clip": 1.04299212,
      "balance_loss_mlp": 1.0142591,
      "epoch": 0.7962484218120603,
      "flos": 21944436612480.0,
      "grad_norm": 1.471355302036172,
      "language_loss": 0.77995121,
      "learning_rate": 4.1975681880837023e-07,
      "loss": 0.80164349,
      "num_input_tokens_seen": 142642495,
      "step": 6622,
      "time_per_iteration": 2.4658966064453125
    },
    {
      "auxiliary_loss_clip": 0.0111791,
      "auxiliary_loss_mlp": 0.01024361,
      "balance_loss_clip": 1.03971243,
      "balance_loss_mlp": 1.01717281,
      "epoch": 0.7963686647026994,
      "flos": 18876264687360.0,
      "grad_norm": 1.7483037580313359,
      "language_loss": 0.82218623,
      "learning_rate": 4.192794679022895e-07,
      "loss": 0.84360898,
      "num_input_tokens_seen": 142660820,
      "step": 6623,
      "time_per_iteration": 2.4928228855133057
    },
    {
      "auxiliary_loss_clip": 0.01152496,
      "auxiliary_loss_mlp": 0.01026655,
      "balance_loss_clip": 1.04489589,
      "balance_loss_mlp": 1.02009594,
      "epoch": 0.7964889075933386,
      "flos": 29716582763520.0,
      "grad_norm": 1.7625971478868678,
      "language_loss": 0.7220723,
      "learning_rate": 4.1880235679025743e-07,
      "loss": 0.74386382,
      "num_input_tokens_seen": 142680915,
      "step": 6624,
      "time_per_iteration": 2.5325610637664795
    },
    {
      "auxiliary_loss_clip": 0.01095573,
      "auxiliary_loss_mlp": 0.01031653,
      "balance_loss_clip": 1.03922272,
      "balance_loss_mlp": 1.02478313,
      "epoch": 0.7966091504839776,
      "flos": 29491458272640.0,
      "grad_norm": 1.9780163457741171,
      "language_loss": 0.63591146,
      "learning_rate": 4.1832548554464986e-07,
      "loss": 0.65718377,
      "num_input_tokens_seen": 142699210,
      "step": 6625,
      "time_per_iteration": 2.616722583770752
    },
    {
      "auxiliary_loss_clip": 0.01048986,
      "auxiliary_loss_mlp": 0.01002833,
      "balance_loss_clip": 1.0090121,
      "balance_loss_mlp": 1.00172412,
      "epoch": 0.7967293933746167,
      "flos": 67288697101440.0,
      "grad_norm": 0.7447212932907569,
      "language_loss": 0.58800209,
      "learning_rate": 4.178488542378098e-07,
      "loss": 0.60852027,
      "num_input_tokens_seen": 142756790,
      "step": 6626,
      "time_per_iteration": 2.9727652072906494
    },
    {
      "auxiliary_loss_clip": 0.01169683,
      "auxiliary_loss_mlp": 0.01025511,
      "balance_loss_clip": 1.04920316,
      "balance_loss_mlp": 1.01832235,
      "epoch": 0.7968496362652558,
      "flos": 25554679660800.0,
      "grad_norm": 1.6894785489034347,
      "language_loss": 0.89079356,
      "learning_rate": 4.173724629420401e-07,
      "loss": 0.91274548,
      "num_input_tokens_seen": 142778150,
      "step": 6627,
      "time_per_iteration": 2.4321272373199463
    },
    {
      "auxiliary_loss_clip": 0.01145601,
      "auxiliary_loss_mlp": 0.01026488,
      "balance_loss_clip": 1.04684985,
      "balance_loss_mlp": 1.01886463,
      "epoch": 0.7969698791558949,
      "flos": 14501088581760.0,
      "grad_norm": 2.6422234536385565,
      "language_loss": 0.68216497,
      "learning_rate": 4.168963117296087e-07,
      "loss": 0.70388585,
      "num_input_tokens_seen": 142795485,
      "step": 6628,
      "time_per_iteration": 2.4320638179779053
    },
    {
      "auxiliary_loss_clip": 0.01166011,
      "auxiliary_loss_mlp": 0.01021662,
      "balance_loss_clip": 1.04813099,
      "balance_loss_mlp": 1.01493895,
      "epoch": 0.797090122046534,
      "flos": 22127545169280.0,
      "grad_norm": 2.09721354716106,
      "language_loss": 0.76210493,
      "learning_rate": 4.1642040067274876e-07,
      "loss": 0.78398168,
      "num_input_tokens_seen": 142815155,
      "step": 6629,
      "time_per_iteration": 3.2515742778778076
    },
    {
      "auxiliary_loss_clip": 0.0114023,
      "auxiliary_loss_mlp": 0.01019083,
      "balance_loss_clip": 1.04439282,
      "balance_loss_mlp": 1.01215971,
      "epoch": 0.7972103649371731,
      "flos": 19897671830400.0,
      "grad_norm": 1.584353341051865,
      "language_loss": 0.72362602,
      "learning_rate": 4.1594472984365493e-07,
      "loss": 0.74521911,
      "num_input_tokens_seen": 142833840,
      "step": 6630,
      "time_per_iteration": 3.233954906463623
    },
    {
      "auxiliary_loss_clip": 0.01148549,
      "auxiliary_loss_mlp": 0.01027875,
      "balance_loss_clip": 1.04648972,
      "balance_loss_mlp": 1.02128029,
      "epoch": 0.7973306078278122,
      "flos": 36058621847040.0,
      "grad_norm": 1.9469795785085844,
      "language_loss": 0.77993846,
      "learning_rate": 4.154692993144862e-07,
      "loss": 0.80170268,
      "num_input_tokens_seen": 142853610,
      "step": 6631,
      "time_per_iteration": 3.4039735794067383
    },
    {
      "auxiliary_loss_clip": 0.01164139,
      "auxiliary_loss_mlp": 0.00761675,
      "balance_loss_clip": 1.04683375,
      "balance_loss_mlp": 1.00043726,
      "epoch": 0.7974508507184512,
      "flos": 21360600950400.0,
      "grad_norm": 2.6279990431078666,
      "language_loss": 0.71698678,
      "learning_rate": 4.1499410915736476e-07,
      "loss": 0.73624492,
      "num_input_tokens_seen": 142872540,
      "step": 6632,
      "time_per_iteration": 2.421588897705078
    },
    {
      "auxiliary_loss_clip": 0.01058281,
      "auxiliary_loss_mlp": 0.01000898,
      "balance_loss_clip": 1.01310456,
      "balance_loss_mlp": 0.99992037,
      "epoch": 0.7975710936090904,
      "flos": 68253115317120.0,
      "grad_norm": 0.7752791990943658,
      "language_loss": 0.6431545,
      "learning_rate": 4.145191594443762e-07,
      "loss": 0.6637463,
      "num_input_tokens_seen": 142936895,
      "step": 6633,
      "time_per_iteration": 3.1609485149383545
    },
    {
      "auxiliary_loss_clip": 0.01117424,
      "auxiliary_loss_mlp": 0.01026095,
      "balance_loss_clip": 1.04175329,
      "balance_loss_mlp": 1.01909709,
      "epoch": 0.7976913364997295,
      "flos": 22492433479680.0,
      "grad_norm": 1.6771579006444906,
      "language_loss": 0.70526063,
      "learning_rate": 4.140444502475713e-07,
      "loss": 0.72669578,
      "num_input_tokens_seen": 142956445,
      "step": 6634,
      "time_per_iteration": 2.549887180328369
    },
    {
      "auxiliary_loss_clip": 0.01146464,
      "auxiliary_loss_mlp": 0.01028909,
      "balance_loss_clip": 1.04296589,
      "balance_loss_mlp": 1.02203321,
      "epoch": 0.7978115793903685,
      "flos": 15263220378240.0,
      "grad_norm": 1.7871226904047188,
      "language_loss": 0.70029974,
      "learning_rate": 4.1356998163896216e-07,
      "loss": 0.72205341,
      "num_input_tokens_seen": 142973495,
      "step": 6635,
      "time_per_iteration": 3.19233775138855
    },
    {
      "auxiliary_loss_clip": 0.01127764,
      "auxiliary_loss_mlp": 0.01023754,
      "balance_loss_clip": 1.04546535,
      "balance_loss_mlp": 1.01692009,
      "epoch": 0.7979318222810077,
      "flos": 19719232041600.0,
      "grad_norm": 1.865828625478986,
      "language_loss": 0.74809849,
      "learning_rate": 4.130957536905255e-07,
      "loss": 0.76961368,
      "num_input_tokens_seen": 142991510,
      "step": 6636,
      "time_per_iteration": 2.5263454914093018
    },
    {
      "auxiliary_loss_clip": 0.01146819,
      "auxiliary_loss_mlp": 0.01028785,
      "balance_loss_clip": 1.04677951,
      "balance_loss_mlp": 1.0209589,
      "epoch": 0.7980520651716467,
      "flos": 15560273854080.0,
      "grad_norm": 2.3695581695747987,
      "language_loss": 0.71548969,
      "learning_rate": 4.1262176647420134e-07,
      "loss": 0.73724574,
      "num_input_tokens_seen": 143009675,
      "step": 6637,
      "time_per_iteration": 2.4899373054504395
    },
    {
      "auxiliary_loss_clip": 0.01145188,
      "auxiliary_loss_mlp": 0.01025695,
      "balance_loss_clip": 1.04762423,
      "balance_loss_mlp": 1.0186677,
      "epoch": 0.7981723080622858,
      "flos": 22309432663680.0,
      "grad_norm": 1.747200761284665,
      "language_loss": 0.79858661,
      "learning_rate": 4.121480200618923e-07,
      "loss": 0.82029545,
      "num_input_tokens_seen": 143029330,
      "step": 6638,
      "time_per_iteration": 2.630326986312866
    },
    {
      "auxiliary_loss_clip": 0.01133709,
      "auxiliary_loss_mlp": 0.01022972,
      "balance_loss_clip": 1.044855,
      "balance_loss_mlp": 1.01623678,
      "epoch": 0.798292550952925,
      "flos": 22929573997440.0,
      "grad_norm": 1.673504942036836,
      "language_loss": 0.80067152,
      "learning_rate": 4.116745145254674e-07,
      "loss": 0.82223833,
      "num_input_tokens_seen": 143048865,
      "step": 6639,
      "time_per_iteration": 2.527531623840332
    },
    {
      "auxiliary_loss_clip": 0.01039239,
      "auxiliary_loss_mlp": 0.0100207,
      "balance_loss_clip": 1.00960457,
      "balance_loss_mlp": 1.00106227,
      "epoch": 0.798412793843564,
      "flos": 64497936890880.0,
      "grad_norm": 0.7656802468105876,
      "language_loss": 0.58011341,
      "learning_rate": 4.1120124993675476e-07,
      "loss": 0.60052645,
      "num_input_tokens_seen": 143113295,
      "step": 6640,
      "time_per_iteration": 3.1508359909057617
    },
    {
      "auxiliary_loss_clip": 0.01145018,
      "auxiliary_loss_mlp": 0.01022817,
      "balance_loss_clip": 1.04478443,
      "balance_loss_mlp": 1.0151875,
      "epoch": 0.7985330367342031,
      "flos": 13586910514560.0,
      "grad_norm": 2.1090898927963853,
      "language_loss": 0.61771172,
      "learning_rate": 4.107282263675498e-07,
      "loss": 0.63939011,
      "num_input_tokens_seen": 143130965,
      "step": 6641,
      "time_per_iteration": 2.501239538192749
    },
    {
      "auxiliary_loss_clip": 0.01039931,
      "auxiliary_loss_mlp": 0.00752985,
      "balance_loss_clip": 1.01194334,
      "balance_loss_mlp": 0.99999315,
      "epoch": 0.7986532796248422,
      "flos": 67698797656320.0,
      "grad_norm": 0.7910024269912419,
      "language_loss": 0.52526909,
      "learning_rate": 4.1025544388960907e-07,
      "loss": 0.54319823,
      "num_input_tokens_seen": 143192005,
      "step": 6642,
      "time_per_iteration": 3.082073450088501
    },
    {
      "auxiliary_loss_clip": 0.01151596,
      "auxiliary_loss_mlp": 0.01026218,
      "balance_loss_clip": 1.04854918,
      "balance_loss_mlp": 1.01911342,
      "epoch": 0.7987735225154813,
      "flos": 22455373622400.0,
      "grad_norm": 1.7941724437790638,
      "language_loss": 0.7171185,
      "learning_rate": 4.097829025746538e-07,
      "loss": 0.73889667,
      "num_input_tokens_seen": 143213550,
      "step": 6643,
      "time_per_iteration": 2.5058670043945312
    },
    {
      "auxiliary_loss_clip": 0.01054208,
      "auxiliary_loss_mlp": 0.01003208,
      "balance_loss_clip": 1.01105118,
      "balance_loss_mlp": 1.00221813,
      "epoch": 0.7988937654061203,
      "flos": 68864098682880.0,
      "grad_norm": 0.657115815200203,
      "language_loss": 0.60978001,
      "learning_rate": 4.0931060249436757e-07,
      "loss": 0.63035417,
      "num_input_tokens_seen": 143277390,
      "step": 6644,
      "time_per_iteration": 3.07392954826355
    },
    {
      "auxiliary_loss_clip": 0.0114965,
      "auxiliary_loss_mlp": 0.01030269,
      "balance_loss_clip": 1.04651618,
      "balance_loss_mlp": 1.0229193,
      "epoch": 0.7990140082967595,
      "flos": 20806893820800.0,
      "grad_norm": 1.9034701140918886,
      "language_loss": 0.69732344,
      "learning_rate": 4.088385437203978e-07,
      "loss": 0.71912265,
      "num_input_tokens_seen": 143294400,
      "step": 6645,
      "time_per_iteration": 2.4290459156036377
    },
    {
      "auxiliary_loss_clip": 0.01165878,
      "auxiliary_loss_mlp": 0.01023936,
      "balance_loss_clip": 1.04676008,
      "balance_loss_mlp": 1.01672649,
      "epoch": 0.7991342511873986,
      "flos": 18985289443200.0,
      "grad_norm": 2.137935974741238,
      "language_loss": 0.77576292,
      "learning_rate": 4.083667263243564e-07,
      "loss": 0.79766107,
      "num_input_tokens_seen": 143312745,
      "step": 6646,
      "time_per_iteration": 2.4386520385742188
    },
    {
      "auxiliary_loss_clip": 0.01150429,
      "auxiliary_loss_mlp": 0.01025181,
      "balance_loss_clip": 1.04839861,
      "balance_loss_mlp": 1.018502,
      "epoch": 0.7992544940780376,
      "flos": 20816805974400.0,
      "grad_norm": 1.9925427171261334,
      "language_loss": 0.71968889,
      "learning_rate": 4.0789515037781653e-07,
      "loss": 0.741445,
      "num_input_tokens_seen": 143333470,
      "step": 6647,
      "time_per_iteration": 2.4792494773864746
    },
    {
      "auxiliary_loss_clip": 0.01155546,
      "auxiliary_loss_mlp": 0.01022919,
      "balance_loss_clip": 1.04755127,
      "balance_loss_mlp": 1.01606393,
      "epoch": 0.7993747369686768,
      "flos": 12640772321280.0,
      "grad_norm": 1.7584999545295306,
      "language_loss": 0.82385117,
      "learning_rate": 4.0742381595231755e-07,
      "loss": 0.84563583,
      "num_input_tokens_seen": 143350195,
      "step": 6648,
      "time_per_iteration": 2.477421522140503
    },
    {
      "auxiliary_loss_clip": 0.01127987,
      "auxiliary_loss_mlp": 0.01025056,
      "balance_loss_clip": 1.04492748,
      "balance_loss_mlp": 1.01849341,
      "epoch": 0.7994949798593158,
      "flos": 20078769225600.0,
      "grad_norm": 2.2203364577199,
      "language_loss": 0.78458756,
      "learning_rate": 4.06952723119359e-07,
      "loss": 0.80611795,
      "num_input_tokens_seen": 143370070,
      "step": 6649,
      "time_per_iteration": 2.5374505519866943
    },
    {
      "auxiliary_loss_clip": 0.01127601,
      "auxiliary_loss_mlp": 0.01026033,
      "balance_loss_clip": 1.04227853,
      "balance_loss_mlp": 1.0186007,
      "epoch": 0.7996152227499549,
      "flos": 38654209509120.0,
      "grad_norm": 1.8569947275147876,
      "language_loss": 0.67374372,
      "learning_rate": 4.0648187195040504e-07,
      "loss": 0.69528008,
      "num_input_tokens_seen": 143392275,
      "step": 6650,
      "time_per_iteration": 2.6422083377838135
    },
    {
      "auxiliary_loss_clip": 0.01049539,
      "auxiliary_loss_mlp": 0.01002018,
      "balance_loss_clip": 1.00941777,
      "balance_loss_mlp": 1.00083816,
      "epoch": 0.799735465640594,
      "flos": 70243821947520.0,
      "grad_norm": 0.8104505219086647,
      "language_loss": 0.6760093,
      "learning_rate": 4.060112625168848e-07,
      "loss": 0.69652486,
      "num_input_tokens_seen": 143457385,
      "step": 6651,
      "time_per_iteration": 3.1255669593811035
    },
    {
      "auxiliary_loss_clip": 0.01166879,
      "auxiliary_loss_mlp": 0.01023976,
      "balance_loss_clip": 1.04918313,
      "balance_loss_mlp": 1.01689172,
      "epoch": 0.7998557085312331,
      "flos": 24240995550720.0,
      "grad_norm": 1.8984052468269064,
      "language_loss": 0.73776877,
      "learning_rate": 4.055408948901886e-07,
      "loss": 0.75967729,
      "num_input_tokens_seen": 143478785,
      "step": 6652,
      "time_per_iteration": 2.4691245555877686
    },
    {
      "auxiliary_loss_clip": 0.01155409,
      "auxiliary_loss_mlp": 0.01023496,
      "balance_loss_clip": 1.04716694,
      "balance_loss_mlp": 1.01551473,
      "epoch": 0.7999759514218722,
      "flos": 27564025449600.0,
      "grad_norm": 2.258278549137495,
      "language_loss": 0.71474397,
      "learning_rate": 4.050707691416708e-07,
      "loss": 0.73653305,
      "num_input_tokens_seen": 143500095,
      "step": 6653,
      "time_per_iteration": 2.503244161605835
    },
    {
      "auxiliary_loss_clip": 0.01049821,
      "auxiliary_loss_mlp": 0.01001798,
      "balance_loss_clip": 1.0097363,
      "balance_loss_mlp": 1.0006417,
      "epoch": 0.8000961943125112,
      "flos": 67337428878720.0,
      "grad_norm": 0.6762803425296061,
      "language_loss": 0.59772789,
      "learning_rate": 4.046008853426495e-07,
      "loss": 0.61824417,
      "num_input_tokens_seen": 143563410,
      "step": 6654,
      "time_per_iteration": 3.1881251335144043
    },
    {
      "auxiliary_loss_clip": 0.01118773,
      "auxiliary_loss_mlp": 0.01025273,
      "balance_loss_clip": 1.04262495,
      "balance_loss_mlp": 1.017977,
      "epoch": 0.8002164372031504,
      "flos": 28733815676160.0,
      "grad_norm": 4.545113583885504,
      "language_loss": 0.62991476,
      "learning_rate": 4.0413124356440464e-07,
      "loss": 0.65135515,
      "num_input_tokens_seen": 143587455,
      "step": 6655,
      "time_per_iteration": 3.439544200897217
    },
    {
      "auxiliary_loss_clip": 0.01113118,
      "auxiliary_loss_mlp": 0.01025344,
      "balance_loss_clip": 1.04144979,
      "balance_loss_mlp": 1.01833725,
      "epoch": 0.8003366800937894,
      "flos": 17639429725440.0,
      "grad_norm": 1.8082924126450304,
      "language_loss": 0.82063144,
      "learning_rate": 4.0366184387818223e-07,
      "loss": 0.8420161,
      "num_input_tokens_seen": 143605915,
      "step": 6656,
      "time_per_iteration": 2.5539333820343018
    },
    {
      "auxiliary_loss_clip": 0.01171186,
      "auxiliary_loss_mlp": 0.01022304,
      "balance_loss_clip": 1.04884708,
      "balance_loss_mlp": 1.01445675,
      "epoch": 0.8004569229844285,
      "flos": 25995303797760.0,
      "grad_norm": 1.7958345363539914,
      "language_loss": 0.85095513,
      "learning_rate": 4.0319268635518797e-07,
      "loss": 0.87289011,
      "num_input_tokens_seen": 143626490,
      "step": 6657,
      "time_per_iteration": 2.4990313053131104
    },
    {
      "auxiliary_loss_clip": 0.01151284,
      "auxiliary_loss_mlp": 0.01018873,
      "balance_loss_clip": 1.0458653,
      "balance_loss_mlp": 1.01235521,
      "epoch": 0.8005771658750677,
      "flos": 20812352688000.0,
      "grad_norm": 1.628504546613032,
      "language_loss": 0.75346142,
      "learning_rate": 4.027237710665943e-07,
      "loss": 0.77516305,
      "num_input_tokens_seen": 143644955,
      "step": 6658,
      "time_per_iteration": 3.2879130840301514
    },
    {
      "auxiliary_loss_clip": 0.01125352,
      "auxiliary_loss_mlp": 0.01025662,
      "balance_loss_clip": 1.04112196,
      "balance_loss_mlp": 1.01829529,
      "epoch": 0.8006974087657067,
      "flos": 25812626204160.0,
      "grad_norm": 2.697019971360361,
      "language_loss": 0.69361913,
      "learning_rate": 4.022550980835344e-07,
      "loss": 0.71512932,
      "num_input_tokens_seen": 143667200,
      "step": 6659,
      "time_per_iteration": 2.60933256149292
    },
    {
      "auxiliary_loss_clip": 0.01121099,
      "auxiliary_loss_mlp": 0.01022905,
      "balance_loss_clip": 1.04010904,
      "balance_loss_mlp": 1.0158838,
      "epoch": 0.8008176516563458,
      "flos": 17164690646400.0,
      "grad_norm": 2.2842929689764855,
      "language_loss": 0.79404354,
      "learning_rate": 4.017866674771051e-07,
      "loss": 0.81548357,
      "num_input_tokens_seen": 143684685,
      "step": 6660,
      "time_per_iteration": 2.5000317096710205
    },
    {
      "auxiliary_loss_clip": 0.01101012,
      "auxiliary_loss_mlp": 0.01027672,
      "balance_loss_clip": 1.03868544,
      "balance_loss_mlp": 1.02040291,
      "epoch": 0.8009378945469849,
      "flos": 24207311571840.0,
      "grad_norm": 1.8961428877580324,
      "language_loss": 0.74707818,
      "learning_rate": 4.013184793183688e-07,
      "loss": 0.76836503,
      "num_input_tokens_seen": 143706780,
      "step": 6661,
      "time_per_iteration": 3.3352625370025635
    },
    {
      "auxiliary_loss_clip": 0.01149144,
      "auxiliary_loss_mlp": 0.01025466,
      "balance_loss_clip": 1.04364705,
      "balance_loss_mlp": 1.01876688,
      "epoch": 0.801058137437624,
      "flos": 19787318271360.0,
      "grad_norm": 1.8881118416660736,
      "language_loss": 0.72744876,
      "learning_rate": 4.008505336783472e-07,
      "loss": 0.74919486,
      "num_input_tokens_seen": 143724505,
      "step": 6662,
      "time_per_iteration": 2.44775128364563
    },
    {
      "auxiliary_loss_clip": 0.01141981,
      "auxiliary_loss_mlp": 0.01024167,
      "balance_loss_clip": 1.04392195,
      "balance_loss_mlp": 1.01752436,
      "epoch": 0.801178380328263,
      "flos": 18659400324480.0,
      "grad_norm": 1.9128885411117909,
      "language_loss": 0.80724239,
      "learning_rate": 4.003828306280284e-07,
      "loss": 0.82890391,
      "num_input_tokens_seen": 143742180,
      "step": 6663,
      "time_per_iteration": 2.4275035858154297
    },
    {
      "auxiliary_loss_clip": 0.01152745,
      "auxiliary_loss_mlp": 0.01022696,
      "balance_loss_clip": 1.0462929,
      "balance_loss_mlp": 1.01620793,
      "epoch": 0.8012986232189022,
      "flos": 15706573948800.0,
      "grad_norm": 1.8353789082887482,
      "language_loss": 0.7816295,
      "learning_rate": 3.999153702383626e-07,
      "loss": 0.80338395,
      "num_input_tokens_seen": 143760070,
      "step": 6664,
      "time_per_iteration": 2.4279041290283203
    },
    {
      "auxiliary_loss_clip": 0.01154723,
      "auxiliary_loss_mlp": 0.01023236,
      "balance_loss_clip": 1.04591537,
      "balance_loss_mlp": 1.0157851,
      "epoch": 0.8014188661095413,
      "flos": 28584139703040.0,
      "grad_norm": 1.786727106013486,
      "language_loss": 0.73600835,
      "learning_rate": 3.9944815258026263e-07,
      "loss": 0.757788,
      "num_input_tokens_seen": 143781890,
      "step": 6665,
      "time_per_iteration": 2.5283732414245605
    },
    {
      "auxiliary_loss_clip": 0.01155398,
      "auxiliary_loss_mlp": 0.01025454,
      "balance_loss_clip": 1.04737949,
      "balance_loss_mlp": 1.01794958,
      "epoch": 0.8015391090001803,
      "flos": 29310360877440.0,
      "grad_norm": 1.7974828799829023,
      "language_loss": 0.83458436,
      "learning_rate": 3.989811777246057e-07,
      "loss": 0.85639292,
      "num_input_tokens_seen": 143802060,
      "step": 6666,
      "time_per_iteration": 2.5059494972229004
    },
    {
      "auxiliary_loss_clip": 0.01064879,
      "auxiliary_loss_mlp": 0.01000984,
      "balance_loss_clip": 1.0104183,
      "balance_loss_mlp": 0.99997026,
      "epoch": 0.8016593518908195,
      "flos": 70397340675840.0,
      "grad_norm": 0.8520184815143531,
      "language_loss": 0.66284394,
      "learning_rate": 3.985144457422305e-07,
      "loss": 0.68350255,
      "num_input_tokens_seen": 143856345,
      "step": 6667,
      "time_per_iteration": 2.964301586151123
    },
    {
      "auxiliary_loss_clip": 0.01166185,
      "auxiliary_loss_mlp": 0.01022195,
      "balance_loss_clip": 1.04805779,
      "balance_loss_mlp": 1.01520085,
      "epoch": 0.8017795947814585,
      "flos": 26026114688640.0,
      "grad_norm": 2.0642245157263837,
      "language_loss": 0.76509959,
      "learning_rate": 3.9804795670394096e-07,
      "loss": 0.78698337,
      "num_input_tokens_seen": 143876470,
      "step": 6668,
      "time_per_iteration": 2.4591193199157715
    },
    {
      "auxiliary_loss_clip": 0.0112999,
      "auxiliary_loss_mlp": 0.01023939,
      "balance_loss_clip": 1.04341531,
      "balance_loss_mlp": 1.01714134,
      "epoch": 0.8018998376720976,
      "flos": 22087181260800.0,
      "grad_norm": 1.5969724554204254,
      "language_loss": 0.70731962,
      "learning_rate": 3.975817106805022e-07,
      "loss": 0.72885883,
      "num_input_tokens_seen": 143895170,
      "step": 6669,
      "time_per_iteration": 2.4780917167663574
    },
    {
      "auxiliary_loss_clip": 0.01124782,
      "auxiliary_loss_mlp": 0.01027451,
      "balance_loss_clip": 1.04396296,
      "balance_loss_mlp": 1.0199343,
      "epoch": 0.8020200805627368,
      "flos": 34568545023360.0,
      "grad_norm": 2.1789523549842444,
      "language_loss": 0.64707994,
      "learning_rate": 3.97115707742645e-07,
      "loss": 0.66860229,
      "num_input_tokens_seen": 143915845,
      "step": 6670,
      "time_per_iteration": 2.634737491607666
    },
    {
      "auxiliary_loss_clip": 0.01140208,
      "auxiliary_loss_mlp": 0.01021172,
      "balance_loss_clip": 1.04683733,
      "balance_loss_mlp": 1.01445496,
      "epoch": 0.8021403234533758,
      "flos": 20120354196480.0,
      "grad_norm": 2.0240657156320894,
      "language_loss": 0.65097427,
      "learning_rate": 3.966499479610599e-07,
      "loss": 0.67258811,
      "num_input_tokens_seen": 143933940,
      "step": 6671,
      "time_per_iteration": 2.4889557361602783
    },
    {
      "auxiliary_loss_clip": 0.01120821,
      "auxiliary_loss_mlp": 0.01022906,
      "balance_loss_clip": 1.04516745,
      "balance_loss_mlp": 1.01626611,
      "epoch": 0.8022605663440149,
      "flos": 27746200252800.0,
      "grad_norm": 1.772648516596436,
      "language_loss": 0.65133452,
      "learning_rate": 3.9618443140640225e-07,
      "loss": 0.67277181,
      "num_input_tokens_seen": 143952850,
      "step": 6672,
      "time_per_iteration": 2.549924850463867
    },
    {
      "auxiliary_loss_clip": 0.0102025,
      "auxiliary_loss_mlp": 0.0100202,
      "balance_loss_clip": 1.00861204,
      "balance_loss_mlp": 1.0009768,
      "epoch": 0.802380809234654,
      "flos": 60244998768000.0,
      "grad_norm": 0.6882311653034767,
      "language_loss": 0.51398444,
      "learning_rate": 3.957191581492918e-07,
      "loss": 0.53420717,
      "num_input_tokens_seen": 144013610,
      "step": 6673,
      "time_per_iteration": 3.1349406242370605
    },
    {
      "auxiliary_loss_clip": 0.0113414,
      "auxiliary_loss_mlp": 0.01022172,
      "balance_loss_clip": 1.04385209,
      "balance_loss_mlp": 1.01503181,
      "epoch": 0.8025010521252931,
      "flos": 15080722352640.0,
      "grad_norm": 2.6127279831027876,
      "language_loss": 0.70470041,
      "learning_rate": 3.952541282603097e-07,
      "loss": 0.72626352,
      "num_input_tokens_seen": 144028715,
      "step": 6674,
      "time_per_iteration": 2.43558669090271
    },
    {
      "auxiliary_loss_clip": 0.01150407,
      "auxiliary_loss_mlp": 0.01021929,
      "balance_loss_clip": 1.04614091,
      "balance_loss_mlp": 1.01501822,
      "epoch": 0.8026212950159322,
      "flos": 22163527618560.0,
      "grad_norm": 2.0698465936192596,
      "language_loss": 0.83650875,
      "learning_rate": 3.9478934181000013e-07,
      "loss": 0.85823214,
      "num_input_tokens_seen": 144048740,
      "step": 6675,
      "time_per_iteration": 2.449049234390259
    },
    {
      "auxiliary_loss_clip": 0.01169052,
      "auxiliary_loss_mlp": 0.01022378,
      "balance_loss_clip": 1.0476408,
      "balance_loss_mlp": 1.01522565,
      "epoch": 0.8027415379065713,
      "flos": 17675986792320.0,
      "grad_norm": 2.487618029224469,
      "language_loss": 0.84529996,
      "learning_rate": 3.943247988688714e-07,
      "loss": 0.8672142,
      "num_input_tokens_seen": 144067435,
      "step": 6676,
      "time_per_iteration": 2.3897159099578857
    },
    {
      "auxiliary_loss_clip": 0.01152194,
      "auxiliary_loss_mlp": 0.01022045,
      "balance_loss_clip": 1.0463289,
      "balance_loss_mlp": 1.01578331,
      "epoch": 0.8028617807972104,
      "flos": 21979593048960.0,
      "grad_norm": 1.6999376240464537,
      "language_loss": 0.72038507,
      "learning_rate": 3.938604995073933e-07,
      "loss": 0.74212742,
      "num_input_tokens_seen": 144085905,
      "step": 6677,
      "time_per_iteration": 2.531378984451294
    },
    {
      "auxiliary_loss_clip": 0.01140265,
      "auxiliary_loss_mlp": 0.01024485,
      "balance_loss_clip": 1.04425871,
      "balance_loss_mlp": 1.0176903,
      "epoch": 0.8029820236878494,
      "flos": 26428457905920.0,
      "grad_norm": 1.8971618848262934,
      "language_loss": 0.65273595,
      "learning_rate": 3.9339644379600157e-07,
      "loss": 0.67438352,
      "num_input_tokens_seen": 144105735,
      "step": 6678,
      "time_per_iteration": 2.527109384536743
    },
    {
      "auxiliary_loss_clip": 0.01156656,
      "auxiliary_loss_mlp": 0.01025693,
      "balance_loss_clip": 1.04819679,
      "balance_loss_mlp": 1.01890421,
      "epoch": 0.8031022665784886,
      "flos": 17676489582720.0,
      "grad_norm": 1.7219975960449956,
      "language_loss": 0.71130437,
      "learning_rate": 3.929326318050907e-07,
      "loss": 0.73312783,
      "num_input_tokens_seen": 144123405,
      "step": 6679,
      "time_per_iteration": 2.4120123386383057
    },
    {
      "auxiliary_loss_clip": 0.0116192,
      "auxiliary_loss_mlp": 0.01024445,
      "balance_loss_clip": 1.04530418,
      "balance_loss_mlp": 1.01745081,
      "epoch": 0.8032225094691277,
      "flos": 15450279431040.0,
      "grad_norm": 2.2371876749609902,
      "language_loss": 0.78963286,
      "learning_rate": 3.924690636050225e-07,
      "loss": 0.8114965,
      "num_input_tokens_seen": 144140815,
      "step": 6680,
      "time_per_iteration": 2.3645241260528564
    },
    {
      "auxiliary_loss_clip": 0.01154145,
      "auxiliary_loss_mlp": 0.01026626,
      "balance_loss_clip": 1.04784143,
      "balance_loss_mlp": 1.01846623,
      "epoch": 0.8033427523597667,
      "flos": 26179202453760.0,
      "grad_norm": 2.23077868817721,
      "language_loss": 0.73061395,
      "learning_rate": 3.9200573926611915e-07,
      "loss": 0.75242168,
      "num_input_tokens_seen": 144162230,
      "step": 6681,
      "time_per_iteration": 2.476757526397705
    },
    {
      "auxiliary_loss_clip": 0.01152857,
      "auxiliary_loss_mlp": 0.01025499,
      "balance_loss_clip": 1.04971278,
      "balance_loss_mlp": 1.01810765,
      "epoch": 0.8034629952504058,
      "flos": 21324905809920.0,
      "grad_norm": 1.9271129731830863,
      "language_loss": 0.72951698,
      "learning_rate": 3.9154265885866613e-07,
      "loss": 0.75130057,
      "num_input_tokens_seen": 144181540,
      "step": 6682,
      "time_per_iteration": 3.3160345554351807
    },
    {
      "auxiliary_loss_clip": 0.01149921,
      "auxiliary_loss_mlp": 0.01025662,
      "balance_loss_clip": 1.04675198,
      "balance_loss_mlp": 1.0178771,
      "epoch": 0.8035832381410449,
      "flos": 21651585027840.0,
      "grad_norm": 2.7474751679527434,
      "language_loss": 0.75135398,
      "learning_rate": 3.9107982245291394e-07,
      "loss": 0.77310973,
      "num_input_tokens_seen": 144199665,
      "step": 6683,
      "time_per_iteration": 3.273348331451416
    },
    {
      "auxiliary_loss_clip": 0.01124769,
      "auxiliary_loss_mlp": 0.01025883,
      "balance_loss_clip": 1.04568338,
      "balance_loss_mlp": 1.01844466,
      "epoch": 0.803703481031684,
      "flos": 20518818744960.0,
      "grad_norm": 2.376624634240618,
      "language_loss": 0.77103812,
      "learning_rate": 3.9061723011907245e-07,
      "loss": 0.79254466,
      "num_input_tokens_seen": 144219020,
      "step": 6684,
      "time_per_iteration": 2.5193569660186768
    },
    {
      "auxiliary_loss_clip": 0.011371,
      "auxiliary_loss_mlp": 0.01023423,
      "balance_loss_clip": 1.04436541,
      "balance_loss_mlp": 1.01645815,
      "epoch": 0.803823723922323,
      "flos": 22854807838080.0,
      "grad_norm": 1.78926650257332,
      "language_loss": 0.79217517,
      "learning_rate": 3.901548819273179e-07,
      "loss": 0.81378043,
      "num_input_tokens_seen": 144239035,
      "step": 6685,
      "time_per_iteration": 3.3258824348449707
    },
    {
      "auxiliary_loss_clip": 0.01153621,
      "auxiliary_loss_mlp": 0.01025708,
      "balance_loss_clip": 1.04795897,
      "balance_loss_mlp": 1.01842475,
      "epoch": 0.8039439668129622,
      "flos": 21362145235200.0,
      "grad_norm": 2.420503086853807,
      "language_loss": 0.69310957,
      "learning_rate": 3.896927779477881e-07,
      "loss": 0.71490288,
      "num_input_tokens_seen": 144258295,
      "step": 6686,
      "time_per_iteration": 2.447969436645508
    },
    {
      "auxiliary_loss_clip": 0.01123779,
      "auxiliary_loss_mlp": 0.01022723,
      "balance_loss_clip": 1.04276121,
      "balance_loss_mlp": 1.01528442,
      "epoch": 0.8040642097036013,
      "flos": 23802382575360.0,
      "grad_norm": 2.0869171929170998,
      "language_loss": 0.66904062,
      "learning_rate": 3.892309182505833e-07,
      "loss": 0.69050562,
      "num_input_tokens_seen": 144276110,
      "step": 6687,
      "time_per_iteration": 2.5395350456237793
    },
    {
      "auxiliary_loss_clip": 0.01163261,
      "auxiliary_loss_mlp": 0.01024587,
      "balance_loss_clip": 1.04517627,
      "balance_loss_mlp": 1.01760769,
      "epoch": 0.8041844525942403,
      "flos": 25922046009600.0,
      "grad_norm": 2.2044442396972013,
      "language_loss": 0.86270356,
      "learning_rate": 3.887693029057675e-07,
      "loss": 0.8845821,
      "num_input_tokens_seen": 144295620,
      "step": 6688,
      "time_per_iteration": 3.204310178756714
    },
    {
      "auxiliary_loss_clip": 0.01136067,
      "auxiliary_loss_mlp": 0.01023722,
      "balance_loss_clip": 1.04409742,
      "balance_loss_mlp": 1.01681995,
      "epoch": 0.8043046954848795,
      "flos": 25191120153600.0,
      "grad_norm": 1.621480719087694,
      "language_loss": 0.81293631,
      "learning_rate": 3.8830793198336684e-07,
      "loss": 0.83453417,
      "num_input_tokens_seen": 144315210,
      "step": 6689,
      "time_per_iteration": 2.5239224433898926
    },
    {
      "auxiliary_loss_clip": 0.0115652,
      "auxiliary_loss_mlp": 0.0102652,
      "balance_loss_clip": 1.04580259,
      "balance_loss_mlp": 1.01953149,
      "epoch": 0.8044249383755185,
      "flos": 41719185123840.0,
      "grad_norm": 1.6535277724128856,
      "language_loss": 0.70343477,
      "learning_rate": 3.878468055533721e-07,
      "loss": 0.7252652,
      "num_input_tokens_seen": 144337750,
      "step": 6690,
      "time_per_iteration": 2.612668514251709
    },
    {
      "auxiliary_loss_clip": 0.01129922,
      "auxiliary_loss_mlp": 0.01026217,
      "balance_loss_clip": 1.04634726,
      "balance_loss_mlp": 1.01890326,
      "epoch": 0.8045451812661576,
      "flos": 20631434860800.0,
      "grad_norm": 2.4727561753325453,
      "language_loss": 0.84178805,
      "learning_rate": 3.8738592368573464e-07,
      "loss": 0.86334938,
      "num_input_tokens_seen": 144355305,
      "step": 6691,
      "time_per_iteration": 2.504849672317505
    },
    {
      "auxiliary_loss_clip": 0.01114367,
      "auxiliary_loss_mlp": 0.01023368,
      "balance_loss_clip": 1.04309714,
      "balance_loss_mlp": 1.01620388,
      "epoch": 0.8046654241567968,
      "flos": 29711806254720.0,
      "grad_norm": 2.2698168009613346,
      "language_loss": 0.87997055,
      "learning_rate": 3.8692528645037137e-07,
      "loss": 0.90134788,
      "num_input_tokens_seen": 144374485,
      "step": 6692,
      "time_per_iteration": 2.547819137573242
    },
    {
      "auxiliary_loss_clip": 0.01165869,
      "auxiliary_loss_mlp": 0.01024266,
      "balance_loss_clip": 1.04839516,
      "balance_loss_mlp": 1.01750088,
      "epoch": 0.8047856670474358,
      "flos": 17671389851520.0,
      "grad_norm": 2.1837270677836833,
      "language_loss": 0.77791119,
      "learning_rate": 3.8646489391715907e-07,
      "loss": 0.79981256,
      "num_input_tokens_seen": 144388780,
      "step": 6693,
      "time_per_iteration": 2.38718581199646
    },
    {
      "auxiliary_loss_clip": 0.01136008,
      "auxiliary_loss_mlp": 0.0102805,
      "balance_loss_clip": 1.04445839,
      "balance_loss_mlp": 1.02090669,
      "epoch": 0.8049059099380749,
      "flos": 17120699464320.0,
      "grad_norm": 2.3759723085219804,
      "language_loss": 0.8833952,
      "learning_rate": 3.8600474615593903e-07,
      "loss": 0.90503579,
      "num_input_tokens_seen": 144403395,
      "step": 6694,
      "time_per_iteration": 2.420403242111206
    },
    {
      "auxiliary_loss_clip": 0.01034693,
      "auxiliary_loss_mlp": 0.01001139,
      "balance_loss_clip": 1.01007438,
      "balance_loss_mlp": 1.00019681,
      "epoch": 0.805026152828714,
      "flos": 62212903240320.0,
      "grad_norm": 0.7862004234973567,
      "language_loss": 0.59673691,
      "learning_rate": 3.8554484323651605e-07,
      "loss": 0.61709523,
      "num_input_tokens_seen": 144465265,
      "step": 6695,
      "time_per_iteration": 3.1427159309387207
    },
    {
      "auxiliary_loss_clip": 0.01151022,
      "auxiliary_loss_mlp": 0.00761606,
      "balance_loss_clip": 1.04756498,
      "balance_loss_mlp": 1.00050426,
      "epoch": 0.8051463957193531,
      "flos": 21688608971520.0,
      "grad_norm": 1.6207591652070437,
      "language_loss": 0.79389042,
      "learning_rate": 3.85085185228657e-07,
      "loss": 0.81301665,
      "num_input_tokens_seen": 144484235,
      "step": 6696,
      "time_per_iteration": 2.4662370681762695
    },
    {
      "auxiliary_loss_clip": 0.01133228,
      "auxiliary_loss_mlp": 0.01026799,
      "balance_loss_clip": 1.04390085,
      "balance_loss_mlp": 1.0196048,
      "epoch": 0.8052666386099921,
      "flos": 32051458535040.0,
      "grad_norm": 2.027045296516793,
      "language_loss": 0.73432875,
      "learning_rate": 3.8462577220209114e-07,
      "loss": 0.75592899,
      "num_input_tokens_seen": 144504610,
      "step": 6697,
      "time_per_iteration": 2.560473918914795
    },
    {
      "auxiliary_loss_clip": 0.01064645,
      "auxiliary_loss_mlp": 0.01001232,
      "balance_loss_clip": 1.01025391,
      "balance_loss_mlp": 1.00022495,
      "epoch": 0.8053868815006313,
      "flos": 67157875768320.0,
      "grad_norm": 0.7226530099967193,
      "language_loss": 0.59050536,
      "learning_rate": 3.8416660422651127e-07,
      "loss": 0.61116409,
      "num_input_tokens_seen": 144574260,
      "step": 6698,
      "time_per_iteration": 3.09350848197937
    },
    {
      "auxiliary_loss_clip": 0.01126881,
      "auxiliary_loss_mlp": 0.01027739,
      "balance_loss_clip": 1.04254925,
      "balance_loss_mlp": 1.02010632,
      "epoch": 0.8055071243912704,
      "flos": 23837000307840.0,
      "grad_norm": 2.1423675499563575,
      "language_loss": 0.67922997,
      "learning_rate": 3.837076813715723e-07,
      "loss": 0.70077622,
      "num_input_tokens_seen": 144594145,
      "step": 6699,
      "time_per_iteration": 2.529806613922119
    },
    {
      "auxiliary_loss_clip": 0.01121408,
      "auxiliary_loss_mlp": 0.01022613,
      "balance_loss_clip": 1.04094076,
      "balance_loss_mlp": 1.01503682,
      "epoch": 0.8056273672819094,
      "flos": 21324510760320.0,
      "grad_norm": 1.6593974194596193,
      "language_loss": 0.75310874,
      "learning_rate": 3.832490037068941e-07,
      "loss": 0.77454895,
      "num_input_tokens_seen": 144612935,
      "step": 6700,
      "time_per_iteration": 2.5115110874176025
    },
    {
      "auxiliary_loss_clip": 0.01094882,
      "auxiliary_loss_mlp": 0.01020787,
      "balance_loss_clip": 1.0410533,
      "balance_loss_mlp": 1.0135628,
      "epoch": 0.8057476101725486,
      "flos": 25768383626880.0,
      "grad_norm": 1.8502713842368181,
      "language_loss": 0.75766021,
      "learning_rate": 3.827905713020554e-07,
      "loss": 0.77881694,
      "num_input_tokens_seen": 144630580,
      "step": 6701,
      "time_per_iteration": 2.5936341285705566
    },
    {
      "auxiliary_loss_clip": 0.01126223,
      "auxiliary_loss_mlp": 0.01027587,
      "balance_loss_clip": 1.04121804,
      "balance_loss_mlp": 1.01959968,
      "epoch": 0.8058678530631876,
      "flos": 24535283679360.0,
      "grad_norm": 2.9949510074060215,
      "language_loss": 0.68960822,
      "learning_rate": 3.823323842266017e-07,
      "loss": 0.71114624,
      "num_input_tokens_seen": 144649975,
      "step": 6702,
      "time_per_iteration": 2.5584065914154053
    },
    {
      "auxiliary_loss_clip": 0.01151998,
      "auxiliary_loss_mlp": 0.01026027,
      "balance_loss_clip": 1.04330182,
      "balance_loss_mlp": 1.01885414,
      "epoch": 0.8059880959538267,
      "flos": 24753728240640.0,
      "grad_norm": 2.5664519967402004,
      "language_loss": 0.7320441,
      "learning_rate": 3.818744425500393e-07,
      "loss": 0.75382435,
      "num_input_tokens_seen": 144667990,
      "step": 6703,
      "time_per_iteration": 2.468947172164917
    },
    {
      "auxiliary_loss_clip": 0.01117729,
      "auxiliary_loss_mlp": 0.0102817,
      "balance_loss_clip": 1.04089403,
      "balance_loss_mlp": 1.02005482,
      "epoch": 0.8061083388444659,
      "flos": 22196349671040.0,
      "grad_norm": 1.8267384998821306,
      "language_loss": 0.80989528,
      "learning_rate": 3.8141674634183675e-07,
      "loss": 0.83135426,
      "num_input_tokens_seen": 144687020,
      "step": 6704,
      "time_per_iteration": 2.511730670928955
    },
    {
      "auxiliary_loss_clip": 0.01107883,
      "auxiliary_loss_mlp": 0.01024559,
      "balance_loss_clip": 1.04325318,
      "balance_loss_mlp": 1.01813102,
      "epoch": 0.8062285817351049,
      "flos": 30044195735040.0,
      "grad_norm": 1.7698414397158995,
      "language_loss": 0.66323364,
      "learning_rate": 3.809592956714278e-07,
      "loss": 0.68455803,
      "num_input_tokens_seen": 144710255,
      "step": 6705,
      "time_per_iteration": 2.619598627090454
    },
    {
      "auxiliary_loss_clip": 0.01156786,
      "auxiliary_loss_mlp": 0.01027432,
      "balance_loss_clip": 1.04809058,
      "balance_loss_mlp": 1.02072668,
      "epoch": 0.806348824625744,
      "flos": 22782591544320.0,
      "grad_norm": 1.907890647993773,
      "language_loss": 0.74567449,
      "learning_rate": 3.805020906082057e-07,
      "loss": 0.76751667,
      "num_input_tokens_seen": 144728830,
      "step": 6706,
      "time_per_iteration": 2.48353910446167
    },
    {
      "auxiliary_loss_clip": 0.01141351,
      "auxiliary_loss_mlp": 0.01023647,
      "balance_loss_clip": 1.04517627,
      "balance_loss_mlp": 1.01607728,
      "epoch": 0.8064690675163831,
      "flos": 23404600385280.0,
      "grad_norm": 2.3937022476838288,
      "language_loss": 0.80873859,
      "learning_rate": 3.8004513122152917e-07,
      "loss": 0.83038855,
      "num_input_tokens_seen": 144747140,
      "step": 6707,
      "time_per_iteration": 2.474058151245117
    },
    {
      "auxiliary_loss_clip": 0.0113162,
      "auxiliary_loss_mlp": 0.01032052,
      "balance_loss_clip": 1.04780769,
      "balance_loss_mlp": 1.02523577,
      "epoch": 0.8065893104070222,
      "flos": 24060903736320.0,
      "grad_norm": 1.685016397270131,
      "language_loss": 0.66997236,
      "learning_rate": 3.79588417580718e-07,
      "loss": 0.69160903,
      "num_input_tokens_seen": 144765250,
      "step": 6708,
      "time_per_iteration": 3.3262388706207275
    },
    {
      "auxiliary_loss_clip": 0.01153963,
      "auxiliary_loss_mlp": 0.01023628,
      "balance_loss_clip": 1.04774809,
      "balance_loss_mlp": 1.01699686,
      "epoch": 0.8067095532976613,
      "flos": 22305410340480.0,
      "grad_norm": 2.1075360374718954,
      "language_loss": 0.76660264,
      "learning_rate": 3.791319497550558e-07,
      "loss": 0.78837848,
      "num_input_tokens_seen": 144783080,
      "step": 6709,
      "time_per_iteration": 2.4473893642425537
    },
    {
      "auxiliary_loss_clip": 0.01130252,
      "auxiliary_loss_mlp": 0.00761406,
      "balance_loss_clip": 1.04489422,
      "balance_loss_mlp": 1.00036776,
      "epoch": 0.8068297961883004,
      "flos": 17129498296320.0,
      "grad_norm": 1.9592145636576255,
      "language_loss": 0.70970368,
      "learning_rate": 3.78675727813788e-07,
      "loss": 0.72862029,
      "num_input_tokens_seen": 144800645,
      "step": 6710,
      "time_per_iteration": 3.3032350540161133
    },
    {
      "auxiliary_loss_clip": 0.0113773,
      "auxiliary_loss_mlp": 0.01020867,
      "balance_loss_clip": 1.04538417,
      "balance_loss_mlp": 1.01394057,
      "epoch": 0.8069500390789395,
      "flos": 22018843635840.0,
      "grad_norm": 1.8471727289918745,
      "language_loss": 0.73801577,
      "learning_rate": 3.782197518261225e-07,
      "loss": 0.75960177,
      "num_input_tokens_seen": 144820085,
      "step": 6711,
      "time_per_iteration": 3.321425199508667
    },
    {
      "auxiliary_loss_clip": 0.01144001,
      "auxiliary_loss_mlp": 0.01030216,
      "balance_loss_clip": 1.04665875,
      "balance_loss_mlp": 1.0232482,
      "epoch": 0.8070702819695785,
      "flos": 19244241567360.0,
      "grad_norm": 1.9601434379668574,
      "language_loss": 0.95249069,
      "learning_rate": 3.777640218612319e-07,
      "loss": 0.97423285,
      "num_input_tokens_seen": 144838070,
      "step": 6712,
      "time_per_iteration": 2.44708251953125
    },
    {
      "auxiliary_loss_clip": 0.01144317,
      "auxiliary_loss_mlp": 0.01025124,
      "balance_loss_clip": 1.0443778,
      "balance_loss_mlp": 1.01818633,
      "epoch": 0.8071905248602176,
      "flos": 21544320038400.0,
      "grad_norm": 2.1807629174518945,
      "language_loss": 0.71770018,
      "learning_rate": 3.773085379882488e-07,
      "loss": 0.73939461,
      "num_input_tokens_seen": 144857125,
      "step": 6713,
      "time_per_iteration": 2.441237211227417
    },
    {
      "auxiliary_loss_clip": 0.01150555,
      "auxiliary_loss_mlp": 0.00761898,
      "balance_loss_clip": 1.04338217,
      "balance_loss_mlp": 1.00043595,
      "epoch": 0.8073107677508568,
      "flos": 37268309105280.0,
      "grad_norm": 5.107299336549649,
      "language_loss": 0.76135564,
      "learning_rate": 3.768533002762715e-07,
      "loss": 0.78048015,
      "num_input_tokens_seen": 144880660,
      "step": 6714,
      "time_per_iteration": 2.5883443355560303
    },
    {
      "auxiliary_loss_clip": 0.01136512,
      "auxiliary_loss_mlp": 0.01020915,
      "balance_loss_clip": 1.04164505,
      "balance_loss_mlp": 1.01426852,
      "epoch": 0.8074310106414958,
      "flos": 28366269759360.0,
      "grad_norm": 1.8073421214175733,
      "language_loss": 0.76833713,
      "learning_rate": 3.763983087943572e-07,
      "loss": 0.78991139,
      "num_input_tokens_seen": 144900050,
      "step": 6715,
      "time_per_iteration": 3.2253429889678955
    },
    {
      "auxiliary_loss_clip": 0.01142234,
      "auxiliary_loss_mlp": 0.00761875,
      "balance_loss_clip": 1.04312479,
      "balance_loss_mlp": 1.00046432,
      "epoch": 0.8075512535321349,
      "flos": 24281646768000.0,
      "grad_norm": 5.668014207877506,
      "language_loss": 0.80547422,
      "learning_rate": 3.759435636115282e-07,
      "loss": 0.82451534,
      "num_input_tokens_seen": 144920835,
      "step": 6716,
      "time_per_iteration": 2.5130603313446045
    },
    {
      "auxiliary_loss_clip": 0.01095494,
      "auxiliary_loss_mlp": 0.00761484,
      "balance_loss_clip": 1.04471576,
      "balance_loss_mlp": 1.0004586,
      "epoch": 0.807671496422774,
      "flos": 26030855283840.0,
      "grad_norm": 1.690034293635486,
      "language_loss": 0.72882104,
      "learning_rate": 3.7548906479676967e-07,
      "loss": 0.74739081,
      "num_input_tokens_seen": 144940430,
      "step": 6717,
      "time_per_iteration": 2.62199068069458
    },
    {
      "auxiliary_loss_clip": 0.01155652,
      "auxiliary_loss_mlp": 0.01022447,
      "balance_loss_clip": 1.04553843,
      "balance_loss_mlp": 1.01581025,
      "epoch": 0.8077917393134131,
      "flos": 23730740899200.0,
      "grad_norm": 1.7878890026842318,
      "language_loss": 0.71957392,
      "learning_rate": 3.7503481241902855e-07,
      "loss": 0.74135494,
      "num_input_tokens_seen": 144960405,
      "step": 6718,
      "time_per_iteration": 2.470048189163208
    },
    {
      "auxiliary_loss_clip": 0.01139283,
      "auxiliary_loss_mlp": 0.00761381,
      "balance_loss_clip": 1.04501224,
      "balance_loss_mlp": 1.0004518,
      "epoch": 0.8079119822040521,
      "flos": 18402028398720.0,
      "grad_norm": 1.7789515554819748,
      "language_loss": 0.80462229,
      "learning_rate": 3.745808065472145e-07,
      "loss": 0.82362896,
      "num_input_tokens_seen": 144977700,
      "step": 6719,
      "time_per_iteration": 2.4562885761260986
    },
    {
      "auxiliary_loss_clip": 0.01151322,
      "auxiliary_loss_mlp": 0.01028168,
      "balance_loss_clip": 1.05156875,
      "balance_loss_mlp": 1.02140331,
      "epoch": 0.8080322250946913,
      "flos": 23621787970560.0,
      "grad_norm": 1.5855762524968087,
      "language_loss": 0.76182711,
      "learning_rate": 3.741270472501994e-07,
      "loss": 0.78362209,
      "num_input_tokens_seen": 144998340,
      "step": 6720,
      "time_per_iteration": 2.4687557220458984
    },
    {
      "auxiliary_loss_clip": 0.01135863,
      "auxiliary_loss_mlp": 0.01021836,
      "balance_loss_clip": 1.04576635,
      "balance_loss_mlp": 1.01557755,
      "epoch": 0.8081524679853304,
      "flos": 22820692896000.0,
      "grad_norm": 1.6951240409153494,
      "language_loss": 0.72786272,
      "learning_rate": 3.736735345968183e-07,
      "loss": 0.74943966,
      "num_input_tokens_seen": 145017950,
      "step": 6721,
      "time_per_iteration": 2.491544723510742
    },
    {
      "auxiliary_loss_clip": 0.0115311,
      "auxiliary_loss_mlp": 0.0102307,
      "balance_loss_clip": 1.04740298,
      "balance_loss_mlp": 1.01637042,
      "epoch": 0.8082727108759694,
      "flos": 17640004343040.0,
      "grad_norm": 2.2561155494325416,
      "language_loss": 0.78834271,
      "learning_rate": 3.7322026865586986e-07,
      "loss": 0.81010449,
      "num_input_tokens_seen": 145036985,
      "step": 6722,
      "time_per_iteration": 2.431783437728882
    },
    {
      "auxiliary_loss_clip": 0.01158348,
      "auxiliary_loss_mlp": 0.01023599,
      "balance_loss_clip": 1.0487026,
      "balance_loss_mlp": 1.01633954,
      "epoch": 0.8083929537666086,
      "flos": 25958172113280.0,
      "grad_norm": 2.373226396603151,
      "language_loss": 0.73515564,
      "learning_rate": 3.7276724949611206e-07,
      "loss": 0.75697517,
      "num_input_tokens_seen": 145057095,
      "step": 6723,
      "time_per_iteration": 2.480518102645874
    },
    {
      "auxiliary_loss_clip": 0.01141912,
      "auxiliary_loss_mlp": 0.0102297,
      "balance_loss_clip": 1.04560447,
      "balance_loss_mlp": 1.0155189,
      "epoch": 0.8085131966572476,
      "flos": 27089178629760.0,
      "grad_norm": 1.8693078251011823,
      "language_loss": 0.75131035,
      "learning_rate": 3.723144771862694e-07,
      "loss": 0.77295911,
      "num_input_tokens_seen": 145077735,
      "step": 6724,
      "time_per_iteration": 2.5143120288848877
    },
    {
      "auxiliary_loss_clip": 0.01126972,
      "auxiliary_loss_mlp": 0.0102167,
      "balance_loss_clip": 1.04233742,
      "balance_loss_mlp": 1.01441061,
      "epoch": 0.8086334395478867,
      "flos": 23988543788160.0,
      "grad_norm": 1.5281396431524967,
      "language_loss": 0.76768053,
      "learning_rate": 3.718619517950263e-07,
      "loss": 0.78916693,
      "num_input_tokens_seen": 145098330,
      "step": 6725,
      "time_per_iteration": 2.533939838409424
    },
    {
      "auxiliary_loss_clip": 0.01166834,
      "auxiliary_loss_mlp": 0.0102911,
      "balance_loss_clip": 1.04898512,
      "balance_loss_mlp": 1.02228868,
      "epoch": 0.8087536824385259,
      "flos": 20405879406720.0,
      "grad_norm": 1.9280259871147754,
      "language_loss": 0.76751697,
      "learning_rate": 3.714096733910301e-07,
      "loss": 0.78947645,
      "num_input_tokens_seen": 145115855,
      "step": 6726,
      "time_per_iteration": 2.4907848834991455
    },
    {
      "auxiliary_loss_clip": 0.01160154,
      "auxiliary_loss_mlp": 0.01024471,
      "balance_loss_clip": 1.047575,
      "balance_loss_mlp": 1.01678181,
      "epoch": 0.8088739253291649,
      "flos": 25919639798400.0,
      "grad_norm": 1.8955932804268432,
      "language_loss": 0.705755,
      "learning_rate": 3.709576420428926e-07,
      "loss": 0.72760129,
      "num_input_tokens_seen": 145136655,
      "step": 6727,
      "time_per_iteration": 2.5362765789031982
    },
    {
      "auxiliary_loss_clip": 0.01137981,
      "auxiliary_loss_mlp": 0.01022948,
      "balance_loss_clip": 1.04221606,
      "balance_loss_mlp": 1.01615,
      "epoch": 0.808994168219804,
      "flos": 28402072640640.0,
      "grad_norm": 2.3979954908928223,
      "language_loss": 0.73727179,
      "learning_rate": 3.7050585781918463e-07,
      "loss": 0.75888109,
      "num_input_tokens_seen": 145156955,
      "step": 6728,
      "time_per_iteration": 2.522301435470581
    },
    {
      "auxiliary_loss_clip": 0.01155645,
      "auxiliary_loss_mlp": 0.01025312,
      "balance_loss_clip": 1.04588485,
      "balance_loss_mlp": 1.01780176,
      "epoch": 0.8091144111104431,
      "flos": 17421056991360.0,
      "grad_norm": 1.959254908405037,
      "language_loss": 0.68922156,
      "learning_rate": 3.700543207884428e-07,
      "loss": 0.71103108,
      "num_input_tokens_seen": 145173865,
      "step": 6729,
      "time_per_iteration": 2.4134814739227295
    },
    {
      "auxiliary_loss_clip": 0.0115155,
      "auxiliary_loss_mlp": 0.01023111,
      "balance_loss_clip": 1.04756951,
      "balance_loss_mlp": 1.01617646,
      "epoch": 0.8092346540010822,
      "flos": 32153803361280.0,
      "grad_norm": 2.802261506359597,
      "language_loss": 0.71408498,
      "learning_rate": 3.6960303101916466e-07,
      "loss": 0.73583168,
      "num_input_tokens_seen": 145193780,
      "step": 6730,
      "time_per_iteration": 2.5015552043914795
    },
    {
      "auxiliary_loss_clip": 0.01064216,
      "auxiliary_loss_mlp": 0.00753147,
      "balance_loss_clip": 1.00998366,
      "balance_loss_mlp": 1.00010478,
      "epoch": 0.8093548968917212,
      "flos": 58035093390720.0,
      "grad_norm": 0.7580382329229355,
      "language_loss": 0.5553233,
      "learning_rate": 3.6915198857981047e-07,
      "loss": 0.57349694,
      "num_input_tokens_seen": 145258980,
      "step": 6731,
      "time_per_iteration": 3.040640115737915
    },
    {
      "auxiliary_loss_clip": 0.01123919,
      "auxiliary_loss_mlp": 0.01024858,
      "balance_loss_clip": 1.04523206,
      "balance_loss_mlp": 1.0169307,
      "epoch": 0.8094751397823604,
      "flos": 27381599251200.0,
      "grad_norm": 1.7873658474159342,
      "language_loss": 0.68177879,
      "learning_rate": 3.687011935388027e-07,
      "loss": 0.70326662,
      "num_input_tokens_seen": 145281875,
      "step": 6732,
      "time_per_iteration": 2.5652506351470947
    },
    {
      "auxiliary_loss_clip": 0.01151315,
      "auxiliary_loss_mlp": 0.01019669,
      "balance_loss_clip": 1.0462985,
      "balance_loss_mlp": 1.01307106,
      "epoch": 0.8095953826729995,
      "flos": 24061083304320.0,
      "grad_norm": 1.9111601097421276,
      "language_loss": 0.72748965,
      "learning_rate": 3.6825064596452646e-07,
      "loss": 0.74919951,
      "num_input_tokens_seen": 145302220,
      "step": 6733,
      "time_per_iteration": 2.4660708904266357
    },
    {
      "auxiliary_loss_clip": 0.01151579,
      "auxiliary_loss_mlp": 0.01022666,
      "balance_loss_clip": 1.04503703,
      "balance_loss_mlp": 1.0158236,
      "epoch": 0.8097156255636385,
      "flos": 23951412103680.0,
      "grad_norm": 1.6899498514628872,
      "language_loss": 0.7094565,
      "learning_rate": 3.678003459253305e-07,
      "loss": 0.73119903,
      "num_input_tokens_seen": 145323070,
      "step": 6734,
      "time_per_iteration": 3.3349273204803467
    },
    {
      "auxiliary_loss_clip": 0.01124053,
      "auxiliary_loss_mlp": 0.01021403,
      "balance_loss_clip": 1.04286206,
      "balance_loss_mlp": 1.01383352,
      "epoch": 0.8098358684542777,
      "flos": 21799142098560.0,
      "grad_norm": 2.3888308610252293,
      "language_loss": 0.7422775,
      "learning_rate": 3.673502934895236e-07,
      "loss": 0.76373208,
      "num_input_tokens_seen": 145342575,
      "step": 6735,
      "time_per_iteration": 2.5073490142822266
    },
    {
      "auxiliary_loss_clip": 0.01063695,
      "auxiliary_loss_mlp": 0.0100167,
      "balance_loss_clip": 1.00955558,
      "balance_loss_mlp": 1.00069845,
      "epoch": 0.8099561113449167,
      "flos": 68809515966720.0,
      "grad_norm": 0.6919123196324478,
      "language_loss": 0.57949078,
      "learning_rate": 3.669004887253802e-07,
      "loss": 0.60014439,
      "num_input_tokens_seen": 145408865,
      "step": 6736,
      "time_per_iteration": 3.1324520111083984
    },
    {
      "auxiliary_loss_clip": 0.01140924,
      "auxiliary_loss_mlp": 0.0102314,
      "balance_loss_clip": 1.04639959,
      "balance_loss_mlp": 1.01644981,
      "epoch": 0.8100763542355558,
      "flos": 23586056916480.0,
      "grad_norm": 1.5942867345826883,
      "language_loss": 0.78958714,
      "learning_rate": 3.664509317011335e-07,
      "loss": 0.8112278,
      "num_input_tokens_seen": 145429200,
      "step": 6737,
      "time_per_iteration": 3.324171781539917
    },
    {
      "auxiliary_loss_clip": 0.01153624,
      "auxiliary_loss_mlp": 0.01026404,
      "balance_loss_clip": 1.04933083,
      "balance_loss_mlp": 1.01877201,
      "epoch": 0.810196597126195,
      "flos": 31650408207360.0,
      "grad_norm": 2.221953221788739,
      "language_loss": 0.73493624,
      "learning_rate": 3.6600162248498134e-07,
      "loss": 0.75673652,
      "num_input_tokens_seen": 145452830,
      "step": 6738,
      "time_per_iteration": 3.3799679279327393
    },
    {
      "auxiliary_loss_clip": 0.01081264,
      "auxiliary_loss_mlp": 0.01022846,
      "balance_loss_clip": 1.03793836,
      "balance_loss_mlp": 1.01630449,
      "epoch": 0.810316840016834,
      "flos": 24900459298560.0,
      "grad_norm": 1.7505623178314402,
      "language_loss": 0.75748158,
      "learning_rate": 3.6555256114508426e-07,
      "loss": 0.77852267,
      "num_input_tokens_seen": 145472625,
      "step": 6739,
      "time_per_iteration": 2.589928388595581
    },
    {
      "auxiliary_loss_clip": 0.01136107,
      "auxiliary_loss_mlp": 0.01024665,
      "balance_loss_clip": 1.04095316,
      "balance_loss_mlp": 1.0174644,
      "epoch": 0.8104370829074731,
      "flos": 27965003950080.0,
      "grad_norm": 2.152393013104217,
      "language_loss": 0.73196304,
      "learning_rate": 3.651037477495642e-07,
      "loss": 0.75357068,
      "num_input_tokens_seen": 145494075,
      "step": 6740,
      "time_per_iteration": 2.5469439029693604
    },
    {
      "auxiliary_loss_clip": 0.01165486,
      "auxiliary_loss_mlp": 0.01023075,
      "balance_loss_clip": 1.04593062,
      "balance_loss_mlp": 1.01602674,
      "epoch": 0.8105573257981122,
      "flos": 24640752988800.0,
      "grad_norm": 2.214866630627599,
      "language_loss": 0.68145937,
      "learning_rate": 3.6465518236650584e-07,
      "loss": 0.703345,
      "num_input_tokens_seen": 145514220,
      "step": 6741,
      "time_per_iteration": 3.183682441711426
    },
    {
      "auxiliary_loss_clip": 0.01123966,
      "auxiliary_loss_mlp": 0.01027906,
      "balance_loss_clip": 1.04270828,
      "balance_loss_mlp": 1.0214386,
      "epoch": 0.8106775686887513,
      "flos": 26358935132160.0,
      "grad_norm": 1.6831597036051975,
      "language_loss": 0.78348041,
      "learning_rate": 3.642068650639558e-07,
      "loss": 0.80499911,
      "num_input_tokens_seen": 145533965,
      "step": 6742,
      "time_per_iteration": 2.5536162853240967
    },
    {
      "auxiliary_loss_clip": 0.01129906,
      "auxiliary_loss_mlp": 0.01027939,
      "balance_loss_clip": 1.03913927,
      "balance_loss_mlp": 1.02102757,
      "epoch": 0.8107978115793903,
      "flos": 27271892136960.0,
      "grad_norm": 5.969952903569133,
      "language_loss": 0.64504516,
      "learning_rate": 3.6375879590992334e-07,
      "loss": 0.66662353,
      "num_input_tokens_seen": 145554310,
      "step": 6743,
      "time_per_iteration": 2.519423246383667
    },
    {
      "auxiliary_loss_clip": 0.01133314,
      "auxiliary_loss_mlp": 0.0102539,
      "balance_loss_clip": 1.04300809,
      "balance_loss_mlp": 1.01820803,
      "epoch": 0.8109180544700295,
      "flos": 24934322845440.0,
      "grad_norm": 1.76173379754774,
      "language_loss": 0.81191272,
      "learning_rate": 3.6331097497238173e-07,
      "loss": 0.83349979,
      "num_input_tokens_seen": 145573755,
      "step": 6744,
      "time_per_iteration": 2.5046942234039307
    },
    {
      "auxiliary_loss_clip": 0.0112124,
      "auxiliary_loss_mlp": 0.01019148,
      "balance_loss_clip": 1.04242122,
      "balance_loss_mlp": 1.0123229,
      "epoch": 0.8110382973606686,
      "flos": 21105383840640.0,
      "grad_norm": 2.112121223684056,
      "language_loss": 0.8019529,
      "learning_rate": 3.628634023192627e-07,
      "loss": 0.82335675,
      "num_input_tokens_seen": 145594000,
      "step": 6745,
      "time_per_iteration": 2.5282111167907715
    },
    {
      "auxiliary_loss_clip": 0.011539,
      "auxiliary_loss_mlp": 0.01026568,
      "balance_loss_clip": 1.04620337,
      "balance_loss_mlp": 1.01866996,
      "epoch": 0.8111585402513076,
      "flos": 15414081500160.0,
      "grad_norm": 2.0551271466630947,
      "language_loss": 0.75035942,
      "learning_rate": 3.624160780184644e-07,
      "loss": 0.77216411,
      "num_input_tokens_seen": 145611215,
      "step": 6746,
      "time_per_iteration": 2.4097530841827393
    },
    {
      "auxiliary_loss_clip": 0.0113142,
      "auxiliary_loss_mlp": 0.01025055,
      "balance_loss_clip": 1.04324627,
      "balance_loss_mlp": 1.01795936,
      "epoch": 0.8112787831419467,
      "flos": 24095736950400.0,
      "grad_norm": 2.179920322205396,
      "language_loss": 0.74357545,
      "learning_rate": 3.6196900213784496e-07,
      "loss": 0.76514018,
      "num_input_tokens_seen": 145630530,
      "step": 6747,
      "time_per_iteration": 2.516167640686035
    },
    {
      "auxiliary_loss_clip": 0.01150903,
      "auxiliary_loss_mlp": 0.01025265,
      "balance_loss_clip": 1.04547477,
      "balance_loss_mlp": 1.01844358,
      "epoch": 0.8113990260325858,
      "flos": 20483374999680.0,
      "grad_norm": 1.7869277150268115,
      "language_loss": 0.8668676,
      "learning_rate": 3.6152217474522527e-07,
      "loss": 0.88862932,
      "num_input_tokens_seen": 145647345,
      "step": 6748,
      "time_per_iteration": 2.4633281230926514
    },
    {
      "auxiliary_loss_clip": 0.01150552,
      "auxiliary_loss_mlp": 0.01027627,
      "balance_loss_clip": 1.04738283,
      "balance_loss_mlp": 1.02104044,
      "epoch": 0.8115192689232249,
      "flos": 24901141656960.0,
      "grad_norm": 1.782953547231123,
      "language_loss": 0.72669113,
      "learning_rate": 3.6107559590838975e-07,
      "loss": 0.74847293,
      "num_input_tokens_seen": 145666330,
      "step": 6749,
      "time_per_iteration": 2.5125668048858643
    },
    {
      "auxiliary_loss_clip": 0.01091151,
      "auxiliary_loss_mlp": 0.01024067,
      "balance_loss_clip": 1.04053617,
      "balance_loss_mlp": 1.01686072,
      "epoch": 0.811639511813864,
      "flos": 24057204635520.0,
      "grad_norm": 2.306753225631151,
      "language_loss": 0.66318452,
      "learning_rate": 3.606292656950822e-07,
      "loss": 0.68433666,
      "num_input_tokens_seen": 145684740,
      "step": 6750,
      "time_per_iteration": 2.596280574798584
    },
    {
      "auxiliary_loss_clip": 0.01133294,
      "auxiliary_loss_mlp": 0.01021933,
      "balance_loss_clip": 1.04214907,
      "balance_loss_mlp": 1.01464307,
      "epoch": 0.8117597547045031,
      "flos": 23185150243200.0,
      "grad_norm": 2.020430092515925,
      "language_loss": 0.86557746,
      "learning_rate": 3.601831841730121e-07,
      "loss": 0.88712972,
      "num_input_tokens_seen": 145702660,
      "step": 6751,
      "time_per_iteration": 2.458836317062378
    },
    {
      "auxiliary_loss_clip": 0.01151578,
      "auxiliary_loss_mlp": 0.01022982,
      "balance_loss_clip": 1.04663992,
      "balance_loss_mlp": 1.01578462,
      "epoch": 0.8118799975951422,
      "flos": 23040250778880.0,
      "grad_norm": 1.9722918012160227,
      "language_loss": 0.72822905,
      "learning_rate": 3.5973735140984916e-07,
      "loss": 0.74997461,
      "num_input_tokens_seen": 145722830,
      "step": 6752,
      "time_per_iteration": 2.4769294261932373
    },
    {
      "auxiliary_loss_clip": 0.01103634,
      "auxiliary_loss_mlp": 0.00760804,
      "balance_loss_clip": 1.03789997,
      "balance_loss_mlp": 1.0003264,
      "epoch": 0.8120002404857812,
      "flos": 24639962889600.0,
      "grad_norm": 1.9830176021582018,
      "language_loss": 0.79507232,
      "learning_rate": 3.5929176747322607e-07,
      "loss": 0.81371665,
      "num_input_tokens_seen": 145741935,
      "step": 6753,
      "time_per_iteration": 2.562243700027466
    },
    {
      "auxiliary_loss_clip": 0.01046647,
      "auxiliary_loss_mlp": 0.01000779,
      "balance_loss_clip": 1.01001108,
      "balance_loss_mlp": 0.9998495,
      "epoch": 0.8121204833764204,
      "flos": 57415742156160.0,
      "grad_norm": 0.8115165272066597,
      "language_loss": 0.56236714,
      "learning_rate": 3.588464324307372e-07,
      "loss": 0.5828414,
      "num_input_tokens_seen": 145805560,
      "step": 6754,
      "time_per_iteration": 3.1068928241729736
    },
    {
      "auxiliary_loss_clip": 0.01151819,
      "auxiliary_loss_mlp": 0.01023164,
      "balance_loss_clip": 1.04366028,
      "balance_loss_mlp": 1.01618767,
      "epoch": 0.8122407262670595,
      "flos": 19464589549440.0,
      "grad_norm": 2.0230850209949782,
      "language_loss": 0.7553066,
      "learning_rate": 3.584013463499391e-07,
      "loss": 0.77705646,
      "num_input_tokens_seen": 145824180,
      "step": 6755,
      "time_per_iteration": 2.4217872619628906
    },
    {
      "auxiliary_loss_clip": 0.01045,
      "auxiliary_loss_mlp": 0.01002071,
      "balance_loss_clip": 1.01142633,
      "balance_loss_mlp": 1.00114107,
      "epoch": 0.8123609691576985,
      "flos": 56425325472000.0,
      "grad_norm": 0.7364622901970289,
      "language_loss": 0.64498335,
      "learning_rate": 3.579565092983521e-07,
      "loss": 0.66545409,
      "num_input_tokens_seen": 145885300,
      "step": 6756,
      "time_per_iteration": 2.9235947132110596
    },
    {
      "auxiliary_loss_clip": 0.01167596,
      "auxiliary_loss_mlp": 0.0103167,
      "balance_loss_clip": 1.04898143,
      "balance_loss_mlp": 1.02456522,
      "epoch": 0.8124812120483377,
      "flos": 20631973564800.0,
      "grad_norm": 2.057129140048583,
      "language_loss": 0.83786952,
      "learning_rate": 3.575119213434565e-07,
      "loss": 0.85986215,
      "num_input_tokens_seen": 145903815,
      "step": 6757,
      "time_per_iteration": 2.4005305767059326
    },
    {
      "auxiliary_loss_clip": 0.01148256,
      "auxiliary_loss_mlp": 0.01020378,
      "balance_loss_clip": 1.04555666,
      "balance_loss_mlp": 1.01345491,
      "epoch": 0.8126014549389767,
      "flos": 22492397566080.0,
      "grad_norm": 1.7362842786458825,
      "language_loss": 0.81990516,
      "learning_rate": 3.5706758255269765e-07,
      "loss": 0.84159148,
      "num_input_tokens_seen": 145922270,
      "step": 6758,
      "time_per_iteration": 2.451983690261841
    },
    {
      "auxiliary_loss_clip": 0.01140172,
      "auxiliary_loss_mlp": 0.01025422,
      "balance_loss_clip": 1.04530013,
      "balance_loss_mlp": 1.01828694,
      "epoch": 0.8127216978296158,
      "flos": 23287961946240.0,
      "grad_norm": 1.484807185471433,
      "language_loss": 0.69738537,
      "learning_rate": 3.566234929934795e-07,
      "loss": 0.71904129,
      "num_input_tokens_seen": 145941470,
      "step": 6759,
      "time_per_iteration": 2.4744327068328857
    },
    {
      "auxiliary_loss_clip": 0.01151942,
      "auxiliary_loss_mlp": 0.01028017,
      "balance_loss_clip": 1.04967868,
      "balance_loss_mlp": 1.02092731,
      "epoch": 0.812841940720255,
      "flos": 25154994049920.0,
      "grad_norm": 1.467336948036511,
      "language_loss": 0.717875,
      "learning_rate": 3.561796527331706e-07,
      "loss": 0.73967457,
      "num_input_tokens_seen": 145963145,
      "step": 6760,
      "time_per_iteration": 2.4990997314453125
    },
    {
      "auxiliary_loss_clip": 0.0112695,
      "auxiliary_loss_mlp": 0.01021798,
      "balance_loss_clip": 1.04342651,
      "balance_loss_mlp": 1.01450837,
      "epoch": 0.812962183610894,
      "flos": 26648446752000.0,
      "grad_norm": 3.316365297541003,
      "language_loss": 0.77720737,
      "learning_rate": 3.5573606183910163e-07,
      "loss": 0.79869485,
      "num_input_tokens_seen": 145983150,
      "step": 6761,
      "time_per_iteration": 3.4013988971710205
    },
    {
      "auxiliary_loss_clip": 0.01156931,
      "auxiliary_loss_mlp": 0.01023663,
      "balance_loss_clip": 1.04515123,
      "balance_loss_mlp": 1.0164094,
      "epoch": 0.8130824265015331,
      "flos": 24966965329920.0,
      "grad_norm": 3.1749222682202984,
      "language_loss": 0.7878952,
      "learning_rate": 3.5529272037856493e-07,
      "loss": 0.80970114,
      "num_input_tokens_seen": 146001365,
      "step": 6762,
      "time_per_iteration": 2.455958843231201
    },
    {
      "auxiliary_loss_clip": 0.01019128,
      "auxiliary_loss_mlp": 0.01001795,
      "balance_loss_clip": 1.01098633,
      "balance_loss_mlp": 1.00054884,
      "epoch": 0.8132026693921722,
      "flos": 67622918175360.0,
      "grad_norm": 0.7079936112692861,
      "language_loss": 0.5393635,
      "learning_rate": 3.548496284188149e-07,
      "loss": 0.5595727,
      "num_input_tokens_seen": 146061570,
      "step": 6763,
      "time_per_iteration": 4.004713773727417
    },
    {
      "auxiliary_loss_clip": 0.01106041,
      "auxiliary_loss_mlp": 0.01022403,
      "balance_loss_clip": 1.04592276,
      "balance_loss_mlp": 1.01542282,
      "epoch": 0.8133229122828113,
      "flos": 19495149045120.0,
      "grad_norm": 1.788848618826129,
      "language_loss": 0.7925967,
      "learning_rate": 3.544067860270681e-07,
      "loss": 0.81388116,
      "num_input_tokens_seen": 146079145,
      "step": 6764,
      "time_per_iteration": 2.506537437438965
    },
    {
      "auxiliary_loss_clip": 0.011261,
      "auxiliary_loss_mlp": 0.01022283,
      "balance_loss_clip": 1.04371929,
      "balance_loss_mlp": 1.01499963,
      "epoch": 0.8134431551734503,
      "flos": 20668135582080.0,
      "grad_norm": 1.6238590636023333,
      "language_loss": 0.71366942,
      "learning_rate": 3.539641932705029e-07,
      "loss": 0.7351532,
      "num_input_tokens_seen": 146097625,
      "step": 6765,
      "time_per_iteration": 3.4392738342285156
    },
    {
      "auxiliary_loss_clip": 0.01168683,
      "auxiliary_loss_mlp": 0.01022638,
      "balance_loss_clip": 1.0468924,
      "balance_loss_mlp": 1.01481187,
      "epoch": 0.8135633980640895,
      "flos": 21507332008320.0,
      "grad_norm": 2.0457383455938745,
      "language_loss": 0.76971227,
      "learning_rate": 3.53521850216262e-07,
      "loss": 0.7916255,
      "num_input_tokens_seen": 146117195,
      "step": 6766,
      "time_per_iteration": 2.4210317134857178
    },
    {
      "auxiliary_loss_clip": 0.0116716,
      "auxiliary_loss_mlp": 0.01025439,
      "balance_loss_clip": 1.04795361,
      "balance_loss_mlp": 1.01800609,
      "epoch": 0.8136836409547286,
      "flos": 20554442058240.0,
      "grad_norm": 1.8248419244985346,
      "language_loss": 0.76607561,
      "learning_rate": 3.530797569314461e-07,
      "loss": 0.78800166,
      "num_input_tokens_seen": 146136220,
      "step": 6767,
      "time_per_iteration": 2.449932813644409
    },
    {
      "auxiliary_loss_clip": 0.0116627,
      "auxiliary_loss_mlp": 0.01021435,
      "balance_loss_clip": 1.04790521,
      "balance_loss_mlp": 1.01436305,
      "epoch": 0.8138038838453676,
      "flos": 20299045380480.0,
      "grad_norm": 2.0417898437070305,
      "language_loss": 0.77742028,
      "learning_rate": 3.5263791348312235e-07,
      "loss": 0.79929733,
      "num_input_tokens_seen": 146155415,
      "step": 6768,
      "time_per_iteration": 3.1844124794006348
    },
    {
      "auxiliary_loss_clip": 0.01135365,
      "auxiliary_loss_mlp": 0.01020437,
      "balance_loss_clip": 1.04314303,
      "balance_loss_mlp": 1.01336169,
      "epoch": 0.8139241267360068,
      "flos": 29789840551680.0,
      "grad_norm": 1.835123071558431,
      "language_loss": 0.70644391,
      "learning_rate": 3.521963199383171e-07,
      "loss": 0.72800195,
      "num_input_tokens_seen": 146178370,
      "step": 6769,
      "time_per_iteration": 2.521782398223877
    },
    {
      "auxiliary_loss_clip": 0.01111653,
      "auxiliary_loss_mlp": 0.01024076,
      "balance_loss_clip": 1.04306459,
      "balance_loss_mlp": 1.01651525,
      "epoch": 0.8140443696266458,
      "flos": 19713270384000.0,
      "grad_norm": 1.8564443512224955,
      "language_loss": 0.77004403,
      "learning_rate": 3.517549763640197e-07,
      "loss": 0.79140127,
      "num_input_tokens_seen": 146196010,
      "step": 6770,
      "time_per_iteration": 2.511802911758423
    },
    {
      "auxiliary_loss_clip": 0.01150091,
      "auxiliary_loss_mlp": 0.00761422,
      "balance_loss_clip": 1.04841042,
      "balance_loss_mlp": 1.0003798,
      "epoch": 0.8141646125172849,
      "flos": 27160568910720.0,
      "grad_norm": 1.8615189306943924,
      "language_loss": 0.71163589,
      "learning_rate": 3.513138828271829e-07,
      "loss": 0.73075098,
      "num_input_tokens_seen": 146215880,
      "step": 6771,
      "time_per_iteration": 2.478868246078491
    },
    {
      "auxiliary_loss_clip": 0.01121097,
      "auxiliary_loss_mlp": 0.0102805,
      "balance_loss_clip": 1.04437292,
      "balance_loss_mlp": 1.02095151,
      "epoch": 0.8142848554079241,
      "flos": 39673102700160.0,
      "grad_norm": 2.1408818788341955,
      "language_loss": 0.70136356,
      "learning_rate": 3.508730393947179e-07,
      "loss": 0.72285497,
      "num_input_tokens_seen": 146239135,
      "step": 6772,
      "time_per_iteration": 2.6523797512054443
    },
    {
      "auxiliary_loss_clip": 0.01124148,
      "auxiliary_loss_mlp": 0.01025436,
      "balance_loss_clip": 1.04441214,
      "balance_loss_mlp": 1.01801252,
      "epoch": 0.8144050982985631,
      "flos": 22237288197120.0,
      "grad_norm": 1.5977252217901663,
      "language_loss": 0.72052252,
      "learning_rate": 3.504324461335024e-07,
      "loss": 0.74201846,
      "num_input_tokens_seen": 146259245,
      "step": 6773,
      "time_per_iteration": 2.5373635292053223
    },
    {
      "auxiliary_loss_clip": 0.01101859,
      "auxiliary_loss_mlp": 0.01025827,
      "balance_loss_clip": 1.04027641,
      "balance_loss_mlp": 1.01764369,
      "epoch": 0.8145253411892022,
      "flos": 23038239617280.0,
      "grad_norm": 1.9788502437136732,
      "language_loss": 0.88268816,
      "learning_rate": 3.499921031103732e-07,
      "loss": 0.90396506,
      "num_input_tokens_seen": 146280015,
      "step": 6774,
      "time_per_iteration": 2.567244052886963
    },
    {
      "auxiliary_loss_clip": 0.01133096,
      "auxiliary_loss_mlp": 0.01021849,
      "balance_loss_clip": 1.04251802,
      "balance_loss_mlp": 1.01452994,
      "epoch": 0.8146455840798413,
      "flos": 24827668387200.0,
      "grad_norm": 1.7630653603384026,
      "language_loss": 0.78407025,
      "learning_rate": 3.4955201039212987e-07,
      "loss": 0.80561972,
      "num_input_tokens_seen": 146300935,
      "step": 6775,
      "time_per_iteration": 2.546619415283203
    },
    {
      "auxiliary_loss_clip": 0.01158709,
      "auxiliary_loss_mlp": 0.0102378,
      "balance_loss_clip": 1.04821849,
      "balance_loss_mlp": 1.01678216,
      "epoch": 0.8147658269704804,
      "flos": 19974520978560.0,
      "grad_norm": 1.8649612895370975,
      "language_loss": 0.65722191,
      "learning_rate": 3.4911216804553465e-07,
      "loss": 0.67904675,
      "num_input_tokens_seen": 146319835,
      "step": 6776,
      "time_per_iteration": 2.4294660091400146
    },
    {
      "auxiliary_loss_clip": 0.01137693,
      "auxiliary_loss_mlp": 0.01028213,
      "balance_loss_clip": 1.04436111,
      "balance_loss_mlp": 1.02024341,
      "epoch": 0.8148860698611194,
      "flos": 21178031097600.0,
      "grad_norm": 2.32663602890063,
      "language_loss": 0.70550597,
      "learning_rate": 3.4867257613731017e-07,
      "loss": 0.72716498,
      "num_input_tokens_seen": 146339030,
      "step": 6777,
      "time_per_iteration": 2.457915782928467
    },
    {
      "auxiliary_loss_clip": 0.0113917,
      "auxiliary_loss_mlp": 0.01028638,
      "balance_loss_clip": 1.0445354,
      "balance_loss_mlp": 1.02190268,
      "epoch": 0.8150063127517585,
      "flos": 19606903234560.0,
      "grad_norm": 1.7156260156982601,
      "language_loss": 0.85830152,
      "learning_rate": 3.4823323473414343e-07,
      "loss": 0.87997961,
      "num_input_tokens_seen": 146358550,
      "step": 6778,
      "time_per_iteration": 2.468317985534668
    },
    {
      "auxiliary_loss_clip": 0.01129731,
      "auxiliary_loss_mlp": 0.01028186,
      "balance_loss_clip": 1.04342556,
      "balance_loss_mlp": 1.02039576,
      "epoch": 0.8151265556423977,
      "flos": 22638374438400.0,
      "grad_norm": 1.983731193480427,
      "language_loss": 0.76251066,
      "learning_rate": 3.477941439026812e-07,
      "loss": 0.7840898,
      "num_input_tokens_seen": 146376770,
      "step": 6779,
      "time_per_iteration": 2.494454860687256
    },
    {
      "auxiliary_loss_clip": 0.01138949,
      "auxiliary_loss_mlp": 0.01020383,
      "balance_loss_clip": 1.0453769,
      "balance_loss_mlp": 1.01381469,
      "epoch": 0.8152467985330367,
      "flos": 17968048277760.0,
      "grad_norm": 1.7503836485970063,
      "language_loss": 0.7342881,
      "learning_rate": 3.473553037095349e-07,
      "loss": 0.75588149,
      "num_input_tokens_seen": 146395795,
      "step": 6780,
      "time_per_iteration": 2.452880382537842
    },
    {
      "auxiliary_loss_clip": 0.01131214,
      "auxiliary_loss_mlp": 0.01023514,
      "balance_loss_clip": 1.04388928,
      "balance_loss_mlp": 1.01690102,
      "epoch": 0.8153670414236758,
      "flos": 24969012405120.0,
      "grad_norm": 1.9062589312524958,
      "language_loss": 0.83227015,
      "learning_rate": 3.469167142212743e-07,
      "loss": 0.85381746,
      "num_input_tokens_seen": 146417640,
      "step": 6781,
      "time_per_iteration": 2.5235676765441895
    },
    {
      "auxiliary_loss_clip": 0.01152717,
      "auxiliary_loss_mlp": 0.01021006,
      "balance_loss_clip": 1.04702604,
      "balance_loss_mlp": 1.0134542,
      "epoch": 0.8154872843143149,
      "flos": 31066069754880.0,
      "grad_norm": 2.3758032009090173,
      "language_loss": 0.63515741,
      "learning_rate": 3.4647837550443337e-07,
      "loss": 0.65689468,
      "num_input_tokens_seen": 146436205,
      "step": 6782,
      "time_per_iteration": 2.5068750381469727
    },
    {
      "auxiliary_loss_clip": 0.01125705,
      "auxiliary_loss_mlp": 0.01024466,
      "balance_loss_clip": 1.04395533,
      "balance_loss_mlp": 1.01763213,
      "epoch": 0.815607527204954,
      "flos": 19391654983680.0,
      "grad_norm": 1.7211574302120567,
      "language_loss": 0.74506408,
      "learning_rate": 3.460402876255086e-07,
      "loss": 0.7665658,
      "num_input_tokens_seen": 146453595,
      "step": 6783,
      "time_per_iteration": 2.486147403717041
    },
    {
      "auxiliary_loss_clip": 0.01153723,
      "auxiliary_loss_mlp": 0.01021259,
      "balance_loss_clip": 1.04567194,
      "balance_loss_mlp": 1.01410079,
      "epoch": 0.815727770095593,
      "flos": 26140418743680.0,
      "grad_norm": 2.3238753950209503,
      "language_loss": 0.72175866,
      "learning_rate": 3.456024506509574e-07,
      "loss": 0.74350846,
      "num_input_tokens_seen": 146474515,
      "step": 6784,
      "time_per_iteration": 2.4733428955078125
    },
    {
      "auxiliary_loss_clip": 0.01152921,
      "auxiliary_loss_mlp": 0.00762019,
      "balance_loss_clip": 1.04905045,
      "balance_loss_mlp": 1.00027728,
      "epoch": 0.8158480129862322,
      "flos": 25337527989120.0,
      "grad_norm": 1.5051325928858246,
      "language_loss": 0.74103093,
      "learning_rate": 3.4516486464719873e-07,
      "loss": 0.76018029,
      "num_input_tokens_seen": 146493905,
      "step": 6785,
      "time_per_iteration": 2.4823248386383057
    },
    {
      "auxiliary_loss_clip": 0.01106708,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 1.04218411,
      "balance_loss_mlp": 1.01806688,
      "epoch": 0.8159682558768713,
      "flos": 34423645559040.0,
      "grad_norm": 2.8210384125043633,
      "language_loss": 0.61852992,
      "learning_rate": 3.4472752968061445e-07,
      "loss": 0.6398524,
      "num_input_tokens_seen": 146518335,
      "step": 6786,
      "time_per_iteration": 2.6731882095336914
    },
    {
      "auxiliary_loss_clip": 0.01150896,
      "auxiliary_loss_mlp": 0.01022862,
      "balance_loss_clip": 1.04534078,
      "balance_loss_mlp": 1.01609671,
      "epoch": 0.8160884987675103,
      "flos": 18653223185280.0,
      "grad_norm": 1.8253541527372343,
      "language_loss": 0.73594785,
      "learning_rate": 3.442904458175475e-07,
      "loss": 0.75768542,
      "num_input_tokens_seen": 146535655,
      "step": 6787,
      "time_per_iteration": 2.4440877437591553
    },
    {
      "auxiliary_loss_clip": 0.01149909,
      "auxiliary_loss_mlp": 0.01022029,
      "balance_loss_clip": 1.04453826,
      "balance_loss_mlp": 1.01494253,
      "epoch": 0.8162087416581495,
      "flos": 31430527102080.0,
      "grad_norm": 1.5866742510091092,
      "language_loss": 0.75886804,
      "learning_rate": 3.438536131243044e-07,
      "loss": 0.78058738,
      "num_input_tokens_seen": 146556815,
      "step": 6788,
      "time_per_iteration": 3.3892970085144043
    },
    {
      "auxiliary_loss_clip": 0.01141134,
      "auxiliary_loss_mlp": 0.01021383,
      "balance_loss_clip": 1.04529071,
      "balance_loss_mlp": 1.01405191,
      "epoch": 0.8163289845487885,
      "flos": 37593910915200.0,
      "grad_norm": 2.0582690814130373,
      "language_loss": 0.62014854,
      "learning_rate": 3.434170316671503e-07,
      "loss": 0.6417737,
      "num_input_tokens_seen": 146581845,
      "step": 6789,
      "time_per_iteration": 2.616269826889038
    },
    {
      "auxiliary_loss_clip": 0.01119483,
      "auxiliary_loss_mlp": 0.01021749,
      "balance_loss_clip": 1.04565001,
      "balance_loss_mlp": 1.01456952,
      "epoch": 0.8164492274394276,
      "flos": 13953989554560.0,
      "grad_norm": 2.5681373642359984,
      "language_loss": 0.89551032,
      "learning_rate": 3.4298070151231583e-07,
      "loss": 0.91692269,
      "num_input_tokens_seen": 146597245,
      "step": 6790,
      "time_per_iteration": 4.1213014125823975
    },
    {
      "auxiliary_loss_clip": 0.01141093,
      "auxiliary_loss_mlp": 0.01023911,
      "balance_loss_clip": 1.04423189,
      "balance_loss_mlp": 1.01686323,
      "epoch": 0.8165694703300668,
      "flos": 28986554747520.0,
      "grad_norm": 1.807408001445995,
      "language_loss": 0.59984934,
      "learning_rate": 3.425446227259916e-07,
      "loss": 0.62149942,
      "num_input_tokens_seen": 146618210,
      "step": 6791,
      "time_per_iteration": 2.5338826179504395
    },
    {
      "auxiliary_loss_clip": 0.01138134,
      "auxiliary_loss_mlp": 0.01022011,
      "balance_loss_clip": 1.04407978,
      "balance_loss_mlp": 1.0158453,
      "epoch": 0.8166897132207058,
      "flos": 25118365155840.0,
      "grad_norm": 1.8671361287771309,
      "language_loss": 0.82407534,
      "learning_rate": 3.421087953743296e-07,
      "loss": 0.84567678,
      "num_input_tokens_seen": 146637975,
      "step": 6792,
      "time_per_iteration": 2.5040810108184814
    },
    {
      "auxiliary_loss_clip": 0.01151004,
      "auxiliary_loss_mlp": 0.0102189,
      "balance_loss_clip": 1.04346251,
      "balance_loss_mlp": 1.01463604,
      "epoch": 0.8168099561113449,
      "flos": 23148593176320.0,
      "grad_norm": 2.553035375556881,
      "language_loss": 0.80015707,
      "learning_rate": 3.416732195234464e-07,
      "loss": 0.82188606,
      "num_input_tokens_seen": 146658030,
      "step": 6793,
      "time_per_iteration": 2.4484047889709473
    },
    {
      "auxiliary_loss_clip": 0.01154799,
      "auxiliary_loss_mlp": 0.01019627,
      "balance_loss_clip": 1.04622722,
      "balance_loss_mlp": 1.01309741,
      "epoch": 0.816930199001984,
      "flos": 18407666833920.0,
      "grad_norm": 1.5536434911464005,
      "language_loss": 0.79532695,
      "learning_rate": 3.4123789523941613e-07,
      "loss": 0.81707126,
      "num_input_tokens_seen": 146677855,
      "step": 6794,
      "time_per_iteration": 3.206885814666748
    },
    {
      "auxiliary_loss_clip": 0.01146089,
      "auxiliary_loss_mlp": 0.0102228,
      "balance_loss_clip": 1.04367006,
      "balance_loss_mlp": 1.01458549,
      "epoch": 0.8170504418926231,
      "flos": 21251324799360.0,
      "grad_norm": 1.4107944683858575,
      "language_loss": 0.63365006,
      "learning_rate": 3.4080282258827884e-07,
      "loss": 0.65533376,
      "num_input_tokens_seen": 146696230,
      "step": 6795,
      "time_per_iteration": 2.4373180866241455
    },
    {
      "auxiliary_loss_clip": 0.01152875,
      "auxiliary_loss_mlp": 0.01024394,
      "balance_loss_clip": 1.04489505,
      "balance_loss_mlp": 1.01735151,
      "epoch": 0.8171706847832622,
      "flos": 19099234362240.0,
      "grad_norm": 2.140463471777983,
      "language_loss": 0.72583759,
      "learning_rate": 3.403680016360342e-07,
      "loss": 0.74761027,
      "num_input_tokens_seen": 146714835,
      "step": 6796,
      "time_per_iteration": 2.417269229888916
    },
    {
      "auxiliary_loss_clip": 0.011466,
      "auxiliary_loss_mlp": 0.01028384,
      "balance_loss_clip": 1.04709888,
      "balance_loss_mlp": 1.02022433,
      "epoch": 0.8172909276739013,
      "flos": 21470128496640.0,
      "grad_norm": 1.710010476710125,
      "language_loss": 0.67913902,
      "learning_rate": 3.3993343244864403e-07,
      "loss": 0.70088887,
      "num_input_tokens_seen": 146734425,
      "step": 6797,
      "time_per_iteration": 2.448167562484741
    },
    {
      "auxiliary_loss_clip": 0.01150581,
      "auxiliary_loss_mlp": 0.01023064,
      "balance_loss_clip": 1.04670596,
      "balance_loss_mlp": 1.01618886,
      "epoch": 0.8174111705645404,
      "flos": 27599792417280.0,
      "grad_norm": 1.5355567119877302,
      "language_loss": 0.72985637,
      "learning_rate": 3.394991150920323e-07,
      "loss": 0.75159276,
      "num_input_tokens_seen": 146757545,
      "step": 6798,
      "time_per_iteration": 2.4922478199005127
    },
    {
      "auxiliary_loss_clip": 0.01112965,
      "auxiliary_loss_mlp": 0.00762497,
      "balance_loss_clip": 1.04283333,
      "balance_loss_mlp": 1.00042295,
      "epoch": 0.8175314134551794,
      "flos": 14064594508800.0,
      "grad_norm": 2.1401578866203996,
      "language_loss": 0.74430454,
      "learning_rate": 3.3906504963208396e-07,
      "loss": 0.76305914,
      "num_input_tokens_seen": 146774240,
      "step": 6799,
      "time_per_iteration": 2.5008223056793213
    },
    {
      "auxiliary_loss_clip": 0.01105796,
      "auxiliary_loss_mlp": 0.01020806,
      "balance_loss_clip": 1.04337239,
      "balance_loss_mlp": 1.01371288,
      "epoch": 0.8176516563458186,
      "flos": 22708076780160.0,
      "grad_norm": 1.8264213608802478,
      "language_loss": 0.66216028,
      "learning_rate": 3.3863123613464774e-07,
      "loss": 0.68342632,
      "num_input_tokens_seen": 146793140,
      "step": 6800,
      "time_per_iteration": 2.5437674522399902
    },
    {
      "auxiliary_loss_clip": 0.01137908,
      "auxiliary_loss_mlp": 0.01024504,
      "balance_loss_clip": 1.04049945,
      "balance_loss_mlp": 1.0177331,
      "epoch": 0.8177718992364577,
      "flos": 21945406279680.0,
      "grad_norm": 1.7232300717056792,
      "language_loss": 0.75027788,
      "learning_rate": 3.381976746655317e-07,
      "loss": 0.77190197,
      "num_input_tokens_seen": 146812895,
      "step": 6801,
      "time_per_iteration": 2.479572296142578
    },
    {
      "auxiliary_loss_clip": 0.01102541,
      "auxiliary_loss_mlp": 0.01024681,
      "balance_loss_clip": 1.04279757,
      "balance_loss_mlp": 1.01776099,
      "epoch": 0.8178921421270967,
      "flos": 22017443005440.0,
      "grad_norm": 2.240674159579414,
      "language_loss": 0.67516673,
      "learning_rate": 3.3776436529050756e-07,
      "loss": 0.69643903,
      "num_input_tokens_seen": 146832445,
      "step": 6802,
      "time_per_iteration": 2.5210154056549072
    },
    {
      "auxiliary_loss_clip": 0.01161293,
      "auxiliary_loss_mlp": 0.01024136,
      "balance_loss_clip": 1.04499745,
      "balance_loss_mlp": 1.01685262,
      "epoch": 0.8180123850177359,
      "flos": 33183111496320.0,
      "grad_norm": 1.5574127030815785,
      "language_loss": 0.72516805,
      "learning_rate": 3.373313080753073e-07,
      "loss": 0.74702239,
      "num_input_tokens_seen": 146856505,
      "step": 6803,
      "time_per_iteration": 2.507780075073242
    },
    {
      "auxiliary_loss_clip": 0.01146004,
      "auxiliary_loss_mlp": 0.01026121,
      "balance_loss_clip": 1.0437026,
      "balance_loss_mlp": 1.01911998,
      "epoch": 0.8181326279083749,
      "flos": 22091167670400.0,
      "grad_norm": 1.5935360346107537,
      "language_loss": 0.77332461,
      "learning_rate": 3.3689850308562527e-07,
      "loss": 0.79504585,
      "num_input_tokens_seen": 146876950,
      "step": 6804,
      "time_per_iteration": 2.45294451713562
    },
    {
      "auxiliary_loss_clip": 0.01102351,
      "auxiliary_loss_mlp": 0.01025249,
      "balance_loss_clip": 1.0438962,
      "balance_loss_mlp": 1.01847172,
      "epoch": 0.818252870799014,
      "flos": 15705747936000.0,
      "grad_norm": 1.988547393998068,
      "language_loss": 0.7751469,
      "learning_rate": 3.364659503871183e-07,
      "loss": 0.79642284,
      "num_input_tokens_seen": 146894885,
      "step": 6805,
      "time_per_iteration": 2.517223834991455
    },
    {
      "auxiliary_loss_clip": 0.01121308,
      "auxiliary_loss_mlp": 0.01022799,
      "balance_loss_clip": 1.04175854,
      "balance_loss_mlp": 1.01660275,
      "epoch": 0.8183731136896532,
      "flos": 18770687637120.0,
      "grad_norm": 1.8962312875875167,
      "language_loss": 0.83957648,
      "learning_rate": 3.3603365004540417e-07,
      "loss": 0.86101753,
      "num_input_tokens_seen": 146913180,
      "step": 6806,
      "time_per_iteration": 2.4995391368865967
    },
    {
      "auxiliary_loss_clip": 0.01164914,
      "auxiliary_loss_mlp": 0.01025446,
      "balance_loss_clip": 1.04882169,
      "balance_loss_mlp": 1.01820111,
      "epoch": 0.8184933565802922,
      "flos": 26541792293760.0,
      "grad_norm": 4.5038079164222955,
      "language_loss": 0.77001929,
      "learning_rate": 3.356016021260624e-07,
      "loss": 0.79192287,
      "num_input_tokens_seen": 146933510,
      "step": 6807,
      "time_per_iteration": 2.434100389480591
    },
    {
      "auxiliary_loss_clip": 0.01152504,
      "auxiliary_loss_mlp": 0.01023411,
      "balance_loss_clip": 1.04692888,
      "balance_loss_mlp": 1.01617527,
      "epoch": 0.8186135994709313,
      "flos": 17530117660800.0,
      "grad_norm": 4.464870280662668,
      "language_loss": 0.65455735,
      "learning_rate": 3.35169806694634e-07,
      "loss": 0.6763165,
      "num_input_tokens_seen": 146951760,
      "step": 6808,
      "time_per_iteration": 2.418520212173462
    },
    {
      "auxiliary_loss_clip": 0.01031117,
      "auxiliary_loss_mlp": 0.01001583,
      "balance_loss_clip": 1.01141548,
      "balance_loss_mlp": 1.00076663,
      "epoch": 0.8187338423615703,
      "flos": 63480300675840.0,
      "grad_norm": 0.7969511332793651,
      "language_loss": 0.60679591,
      "learning_rate": 3.3473826381662186e-07,
      "loss": 0.627123,
      "num_input_tokens_seen": 147022900,
      "step": 6809,
      "time_per_iteration": 3.213564872741699
    },
    {
      "auxiliary_loss_clip": 0.0114486,
      "auxiliary_loss_mlp": 0.01022537,
      "balance_loss_clip": 1.04562509,
      "balance_loss_mlp": 1.01572168,
      "epoch": 0.8188540852522095,
      "flos": 17529974006400.0,
      "grad_norm": 1.808331669833302,
      "language_loss": 0.8150667,
      "learning_rate": 3.3430697355749216e-07,
      "loss": 0.83674073,
      "num_input_tokens_seen": 147040590,
      "step": 6810,
      "time_per_iteration": 2.410327434539795
    },
    {
      "auxiliary_loss_clip": 0.01105508,
      "auxiliary_loss_mlp": 0.01023655,
      "balance_loss_clip": 1.04036248,
      "balance_loss_mlp": 1.01603508,
      "epoch": 0.8189743281428485,
      "flos": 14392530702720.0,
      "grad_norm": 1.8819289404063788,
      "language_loss": 0.75460494,
      "learning_rate": 3.3387593598266907e-07,
      "loss": 0.77589655,
      "num_input_tokens_seen": 147057200,
      "step": 6811,
      "time_per_iteration": 2.4785306453704834
    },
    {
      "auxiliary_loss_clip": 0.01114251,
      "auxiliary_loss_mlp": 0.01021904,
      "balance_loss_clip": 1.04026806,
      "balance_loss_mlp": 1.01505256,
      "epoch": 0.8190945710334876,
      "flos": 25080479285760.0,
      "grad_norm": 2.59803691005206,
      "language_loss": 0.78205597,
      "learning_rate": 3.3344515115754225e-07,
      "loss": 0.80341756,
      "num_input_tokens_seen": 147076180,
      "step": 6812,
      "time_per_iteration": 2.528937816619873
    },
    {
      "auxiliary_loss_clip": 0.01127628,
      "auxiliary_loss_mlp": 0.0102061,
      "balance_loss_clip": 1.04191816,
      "balance_loss_mlp": 1.01353741,
      "epoch": 0.8192148139241268,
      "flos": 21507152440320.0,
      "grad_norm": 3.966273827817656,
      "language_loss": 0.80260217,
      "learning_rate": 3.33014619147461e-07,
      "loss": 0.82408452,
      "num_input_tokens_seen": 147094205,
      "step": 6813,
      "time_per_iteration": 2.5006282329559326
    },
    {
      "auxiliary_loss_clip": 0.01138862,
      "auxiliary_loss_mlp": 0.01026467,
      "balance_loss_clip": 1.04736912,
      "balance_loss_mlp": 1.01959419,
      "epoch": 0.8193350568147658,
      "flos": 23952166289280.0,
      "grad_norm": 2.9015100002381757,
      "language_loss": 0.71450073,
      "learning_rate": 3.325843400177362e-07,
      "loss": 0.73615396,
      "num_input_tokens_seen": 147115545,
      "step": 6814,
      "time_per_iteration": 3.3452374935150146
    },
    {
      "auxiliary_loss_clip": 0.01155299,
      "auxiliary_loss_mlp": 0.00761855,
      "balance_loss_clip": 1.04669714,
      "balance_loss_mlp": 1.00044894,
      "epoch": 0.8194552997054049,
      "flos": 20559469962240.0,
      "grad_norm": 1.8868272918197915,
      "language_loss": 0.73585165,
      "learning_rate": 3.32154313833642e-07,
      "loss": 0.75502312,
      "num_input_tokens_seen": 147135700,
      "step": 6815,
      "time_per_iteration": 2.52557635307312
    },
    {
      "auxiliary_loss_clip": 0.01167123,
      "auxiliary_loss_mlp": 0.01025334,
      "balance_loss_clip": 1.04735839,
      "balance_loss_mlp": 1.01802087,
      "epoch": 0.819575542596044,
      "flos": 26031753123840.0,
      "grad_norm": 2.701960319744234,
      "language_loss": 0.59457517,
      "learning_rate": 3.3172454066041164e-07,
      "loss": 0.61649978,
      "num_input_tokens_seen": 147155205,
      "step": 6816,
      "time_per_iteration": 3.199972629547119
    },
    {
      "auxiliary_loss_clip": 0.01096637,
      "auxiliary_loss_mlp": 0.00761462,
      "balance_loss_clip": 1.04412985,
      "balance_loss_mlp": 1.00040317,
      "epoch": 0.8196957854866831,
      "flos": 29096944220160.0,
      "grad_norm": 1.8209534521393016,
      "language_loss": 0.75782466,
      "learning_rate": 3.3129502056324234e-07,
      "loss": 0.77640569,
      "num_input_tokens_seen": 147176570,
      "step": 6817,
      "time_per_iteration": 3.434685707092285
    },
    {
      "auxiliary_loss_clip": 0.01005078,
      "auxiliary_loss_mlp": 0.01001706,
      "balance_loss_clip": 1.01192069,
      "balance_loss_mlp": 1.00066268,
      "epoch": 0.8198160283773221,
      "flos": 69033631898880.0,
      "grad_norm": 0.8014186045695255,
      "language_loss": 0.59753567,
      "learning_rate": 3.3086575360729165e-07,
      "loss": 0.61760348,
      "num_input_tokens_seen": 147234105,
      "step": 6818,
      "time_per_iteration": 3.1724495887756348
    },
    {
      "auxiliary_loss_clip": 0.011364,
      "auxiliary_loss_mlp": 0.01025633,
      "balance_loss_clip": 1.04478908,
      "balance_loss_mlp": 1.0182184,
      "epoch": 0.8199362712679613,
      "flos": 16618058496000.0,
      "grad_norm": 1.6375991749784842,
      "language_loss": 0.71092522,
      "learning_rate": 3.3043673985767906e-07,
      "loss": 0.73254561,
      "num_input_tokens_seen": 147253170,
      "step": 6819,
      "time_per_iteration": 2.771613121032715
    },
    {
      "auxiliary_loss_clip": 0.0111398,
      "auxiliary_loss_mlp": 0.01026777,
      "balance_loss_clip": 1.03930044,
      "balance_loss_mlp": 1.01935029,
      "epoch": 0.8200565141586004,
      "flos": 21757664868480.0,
      "grad_norm": 1.836627459465846,
      "language_loss": 0.77541542,
      "learning_rate": 3.3000797937948564e-07,
      "loss": 0.79682302,
      "num_input_tokens_seen": 147271465,
      "step": 6820,
      "time_per_iteration": 2.5191915035247803
    },
    {
      "auxiliary_loss_clip": 0.01031906,
      "auxiliary_loss_mlp": 0.01001433,
      "balance_loss_clip": 1.00739598,
      "balance_loss_mlp": 1.00044334,
      "epoch": 0.8201767570492394,
      "flos": 69807112392960.0,
      "grad_norm": 0.9465417163068718,
      "language_loss": 0.65043789,
      "learning_rate": 3.295794722377534e-07,
      "loss": 0.6707713,
      "num_input_tokens_seen": 147335070,
      "step": 6821,
      "time_per_iteration": 4.202096939086914
    },
    {
      "auxiliary_loss_clip": 0.01161817,
      "auxiliary_loss_mlp": 0.01024813,
      "balance_loss_clip": 1.0443958,
      "balance_loss_mlp": 1.01835465,
      "epoch": 0.8202969999398786,
      "flos": 23111892455040.0,
      "grad_norm": 2.0523362890499146,
      "language_loss": 0.7999754,
      "learning_rate": 3.291512184974876e-07,
      "loss": 0.82184172,
      "num_input_tokens_seen": 147355460,
      "step": 6822,
      "time_per_iteration": 2.4443702697753906
    },
    {
      "auxiliary_loss_clip": 0.01134688,
      "auxiliary_loss_mlp": 0.010225,
      "balance_loss_clip": 1.04119492,
      "balance_loss_mlp": 1.01518691,
      "epoch": 0.8204172428305176,
      "flos": 28220616109440.0,
      "grad_norm": 1.6446217423910388,
      "language_loss": 0.66581541,
      "learning_rate": 3.2872321822365346e-07,
      "loss": 0.68738729,
      "num_input_tokens_seen": 147375675,
      "step": 6823,
      "time_per_iteration": 2.5599071979522705
    },
    {
      "auxiliary_loss_clip": 0.01150777,
      "auxiliary_loss_mlp": 0.01020447,
      "balance_loss_clip": 1.04727721,
      "balance_loss_mlp": 1.01334214,
      "epoch": 0.8205374857211567,
      "flos": 20887011106560.0,
      "grad_norm": 1.8260951084729837,
      "language_loss": 0.73330677,
      "learning_rate": 3.282954714811783e-07,
      "loss": 0.75501901,
      "num_input_tokens_seen": 147394580,
      "step": 6824,
      "time_per_iteration": 2.4431612491607666
    },
    {
      "auxiliary_loss_clip": 0.01124304,
      "auxiliary_loss_mlp": 0.01023651,
      "balance_loss_clip": 1.04019284,
      "balance_loss_mlp": 1.01628447,
      "epoch": 0.8206577286117959,
      "flos": 13152140294400.0,
      "grad_norm": 2.3441282185655514,
      "language_loss": 0.70823205,
      "learning_rate": 3.2786797833495093e-07,
      "loss": 0.72971153,
      "num_input_tokens_seen": 147409935,
      "step": 6825,
      "time_per_iteration": 2.463106155395508
    },
    {
      "auxiliary_loss_clip": 0.01163014,
      "auxiliary_loss_mlp": 0.01024655,
      "balance_loss_clip": 1.04668975,
      "balance_loss_mlp": 1.01823604,
      "epoch": 0.8207779715024349,
      "flos": 25265634917760.0,
      "grad_norm": 1.7494974601576794,
      "language_loss": 0.7275176,
      "learning_rate": 3.274407388498213e-07,
      "loss": 0.7493943,
      "num_input_tokens_seen": 147428065,
      "step": 6826,
      "time_per_iteration": 2.440802812576294
    },
    {
      "auxiliary_loss_clip": 0.01118471,
      "auxiliary_loss_mlp": 0.01027381,
      "balance_loss_clip": 1.04158759,
      "balance_loss_mlp": 1.02046967,
      "epoch": 0.820898214393074,
      "flos": 19610243199360.0,
      "grad_norm": 5.975114684799946,
      "language_loss": 0.73896289,
      "learning_rate": 3.270137530906021e-07,
      "loss": 0.7604214,
      "num_input_tokens_seen": 147447300,
      "step": 6827,
      "time_per_iteration": 2.5104928016662598
    },
    {
      "auxiliary_loss_clip": 0.01102647,
      "auxiliary_loss_mlp": 0.01023184,
      "balance_loss_clip": 1.04478908,
      "balance_loss_mlp": 1.01661253,
      "epoch": 0.8210184572837131,
      "flos": 15596615439360.0,
      "grad_norm": 1.7635218594643283,
      "language_loss": 0.83393168,
      "learning_rate": 3.265870211220665e-07,
      "loss": 0.85519004,
      "num_input_tokens_seen": 147465135,
      "step": 6828,
      "time_per_iteration": 2.555386781692505
    },
    {
      "auxiliary_loss_clip": 0.01121022,
      "auxiliary_loss_mlp": 0.01028747,
      "balance_loss_clip": 1.04428482,
      "balance_loss_mlp": 1.02113509,
      "epoch": 0.8211387001743522,
      "flos": 20813932886400.0,
      "grad_norm": 1.9315952694056193,
      "language_loss": 0.81658369,
      "learning_rate": 3.2616054300894934e-07,
      "loss": 0.83808136,
      "num_input_tokens_seen": 147484585,
      "step": 6829,
      "time_per_iteration": 2.6306464672088623
    },
    {
      "auxiliary_loss_clip": 0.01128758,
      "auxiliary_loss_mlp": 0.01021882,
      "balance_loss_clip": 1.04499853,
      "balance_loss_mlp": 1.01465201,
      "epoch": 0.8212589430649913,
      "flos": 27704579368320.0,
      "grad_norm": 1.952161659050395,
      "language_loss": 0.84430063,
      "learning_rate": 3.2573431881594693e-07,
      "loss": 0.865807,
      "num_input_tokens_seen": 147504130,
      "step": 6830,
      "time_per_iteration": 2.5522513389587402
    },
    {
      "auxiliary_loss_clip": 0.01094959,
      "auxiliary_loss_mlp": 0.01022258,
      "balance_loss_clip": 1.03914845,
      "balance_loss_mlp": 1.01516461,
      "epoch": 0.8213791859556304,
      "flos": 22455625017600.0,
      "grad_norm": 2.0103338968504167,
      "language_loss": 0.65954649,
      "learning_rate": 3.2530834860771663e-07,
      "loss": 0.68071866,
      "num_input_tokens_seen": 147523510,
      "step": 6831,
      "time_per_iteration": 2.6042492389678955
    },
    {
      "auxiliary_loss_clip": 0.01151151,
      "auxiliary_loss_mlp": 0.01023988,
      "balance_loss_clip": 1.04412436,
      "balance_loss_mlp": 1.01647198,
      "epoch": 0.8214994288462695,
      "flos": 16654471908480.0,
      "grad_norm": 1.9991066861645834,
      "language_loss": 0.74314129,
      "learning_rate": 3.248826324488794e-07,
      "loss": 0.7648927,
      "num_input_tokens_seen": 147540805,
      "step": 6832,
      "time_per_iteration": 2.421898365020752
    },
    {
      "auxiliary_loss_clip": 0.01166024,
      "auxiliary_loss_mlp": 0.01026505,
      "balance_loss_clip": 1.05060315,
      "balance_loss_mlp": 1.0197432,
      "epoch": 0.8216196717369085,
      "flos": 25221787390080.0,
      "grad_norm": 1.6731329061506115,
      "language_loss": 0.8793807,
      "learning_rate": 3.244571704040138e-07,
      "loss": 0.90130591,
      "num_input_tokens_seen": 147560965,
      "step": 6833,
      "time_per_iteration": 2.4621951580047607
    },
    {
      "auxiliary_loss_clip": 0.01148958,
      "auxiliary_loss_mlp": 0.01026551,
      "balance_loss_clip": 1.04374933,
      "balance_loss_mlp": 1.01836705,
      "epoch": 0.8217399146275477,
      "flos": 25371930240000.0,
      "grad_norm": 2.232277605495166,
      "language_loss": 0.73934895,
      "learning_rate": 3.2403196253766374e-07,
      "loss": 0.76110405,
      "num_input_tokens_seen": 147580045,
      "step": 6834,
      "time_per_iteration": 2.4883499145507812
    },
    {
      "auxiliary_loss_clip": 0.01148492,
      "auxiliary_loss_mlp": 0.01025578,
      "balance_loss_clip": 1.04524648,
      "balance_loss_mlp": 1.01779974,
      "epoch": 0.8218601575181868,
      "flos": 25629625388160.0,
      "grad_norm": 2.4671913219894672,
      "language_loss": 0.78968859,
      "learning_rate": 3.2360700891433254e-07,
      "loss": 0.81142926,
      "num_input_tokens_seen": 147599070,
      "step": 6835,
      "time_per_iteration": 2.4703404903411865
    },
    {
      "auxiliary_loss_clip": 0.01023107,
      "auxiliary_loss_mlp": 0.01002827,
      "balance_loss_clip": 1.01077461,
      "balance_loss_mlp": 1.0017724,
      "epoch": 0.8219804004088258,
      "flos": 67660229427840.0,
      "grad_norm": 0.799394554818687,
      "language_loss": 0.57317376,
      "learning_rate": 3.231823095984847e-07,
      "loss": 0.59343314,
      "num_input_tokens_seen": 147653710,
      "step": 6836,
      "time_per_iteration": 3.033067226409912
    },
    {
      "auxiliary_loss_clip": 0.01137238,
      "auxiliary_loss_mlp": 0.01022,
      "balance_loss_clip": 1.04527628,
      "balance_loss_mlp": 1.01504469,
      "epoch": 0.822100643299465,
      "flos": 19464266327040.0,
      "grad_norm": 2.012688134441855,
      "language_loss": 0.76158744,
      "learning_rate": 3.2275786465454814e-07,
      "loss": 0.78317982,
      "num_input_tokens_seen": 147670360,
      "step": 6837,
      "time_per_iteration": 2.4612369537353516
    },
    {
      "auxiliary_loss_clip": 0.01120838,
      "auxiliary_loss_mlp": 0.01023007,
      "balance_loss_clip": 1.04266298,
      "balance_loss_mlp": 1.01639128,
      "epoch": 0.822220886190104,
      "flos": 24681368292480.0,
      "grad_norm": 2.137336169612773,
      "language_loss": 0.75605828,
      "learning_rate": 3.2233367414690917e-07,
      "loss": 0.7774967,
      "num_input_tokens_seen": 147692550,
      "step": 6838,
      "time_per_iteration": 2.548297643661499
    },
    {
      "auxiliary_loss_clip": 0.01118648,
      "auxiliary_loss_mlp": 0.01022626,
      "balance_loss_clip": 1.04009938,
      "balance_loss_mlp": 1.01574206,
      "epoch": 0.8223411290807431,
      "flos": 27819062991360.0,
      "grad_norm": 2.6293256821619257,
      "language_loss": 0.8494792,
      "learning_rate": 3.219097381399183e-07,
      "loss": 0.87089193,
      "num_input_tokens_seen": 147709725,
      "step": 6839,
      "time_per_iteration": 2.5535924434661865
    },
    {
      "auxiliary_loss_clip": 0.01143055,
      "auxiliary_loss_mlp": 0.01024934,
      "balance_loss_clip": 1.04465055,
      "balance_loss_mlp": 1.01835024,
      "epoch": 0.8224613719713821,
      "flos": 23218546913280.0,
      "grad_norm": 2.384893230331127,
      "language_loss": 0.80890906,
      "learning_rate": 3.2148605669788584e-07,
      "loss": 0.83058888,
      "num_input_tokens_seen": 147729615,
      "step": 6840,
      "time_per_iteration": 3.3277125358581543
    },
    {
      "auxiliary_loss_clip": 0.01138229,
      "auxiliary_loss_mlp": 0.0102417,
      "balance_loss_clip": 1.04555702,
      "balance_loss_mlp": 1.01680923,
      "epoch": 0.8225816148620213,
      "flos": 15706250726400.0,
      "grad_norm": 2.6911656078117816,
      "language_loss": 0.77285266,
      "learning_rate": 3.2106262988508405e-07,
      "loss": 0.79447669,
      "num_input_tokens_seen": 147747665,
      "step": 6841,
      "time_per_iteration": 2.462916851043701
    },
    {
      "auxiliary_loss_clip": 0.01139259,
      "auxiliary_loss_mlp": 0.0102207,
      "balance_loss_clip": 1.04456878,
      "balance_loss_mlp": 1.01505494,
      "epoch": 0.8227018577526604,
      "flos": 18515111391360.0,
      "grad_norm": 1.7893112431219729,
      "language_loss": 0.74198681,
      "learning_rate": 3.206394577657465e-07,
      "loss": 0.76360011,
      "num_input_tokens_seen": 147765445,
      "step": 6842,
      "time_per_iteration": 2.455817222595215
    },
    {
      "auxiliary_loss_clip": 0.01156358,
      "auxiliary_loss_mlp": 0.01028684,
      "balance_loss_clip": 1.04799604,
      "balance_loss_mlp": 1.02120948,
      "epoch": 0.8228221006432994,
      "flos": 22236785406720.0,
      "grad_norm": 2.408334106447516,
      "language_loss": 0.72702813,
      "learning_rate": 3.202165404040675e-07,
      "loss": 0.74887854,
      "num_input_tokens_seen": 147783365,
      "step": 6843,
      "time_per_iteration": 3.2742409706115723
    },
    {
      "auxiliary_loss_clip": 0.01097748,
      "auxiliary_loss_mlp": 0.0102869,
      "balance_loss_clip": 1.0427283,
      "balance_loss_mlp": 1.02111745,
      "epoch": 0.8229423435339386,
      "flos": 24097532630400.0,
      "grad_norm": 2.8885698183824884,
      "language_loss": 0.74479699,
      "learning_rate": 3.1979387786420396e-07,
      "loss": 0.76606143,
      "num_input_tokens_seen": 147803605,
      "step": 6844,
      "time_per_iteration": 2.601386308670044
    },
    {
      "auxiliary_loss_clip": 0.01138487,
      "auxiliary_loss_mlp": 0.01018616,
      "balance_loss_clip": 1.04223073,
      "balance_loss_mlp": 1.01184177,
      "epoch": 0.8230625864245776,
      "flos": 23878549365120.0,
      "grad_norm": 1.7502135431720856,
      "language_loss": 0.82327712,
      "learning_rate": 3.1937147021027346e-07,
      "loss": 0.84484816,
      "num_input_tokens_seen": 147822060,
      "step": 6845,
      "time_per_iteration": 2.4975786209106445
    },
    {
      "auxiliary_loss_clip": 0.01148717,
      "auxiliary_loss_mlp": 0.01021213,
      "balance_loss_clip": 1.04495025,
      "balance_loss_mlp": 1.01484394,
      "epoch": 0.8231828293152167,
      "flos": 16581106379520.0,
      "grad_norm": 2.4845316230889787,
      "language_loss": 0.7667774,
      "learning_rate": 3.189493175063547e-07,
      "loss": 0.78847677,
      "num_input_tokens_seen": 147839295,
      "step": 6846,
      "time_per_iteration": 2.4254555702209473
    },
    {
      "auxiliary_loss_clip": 0.0113708,
      "auxiliary_loss_mlp": 0.01024559,
      "balance_loss_clip": 1.04483533,
      "balance_loss_mlp": 1.01731062,
      "epoch": 0.8233030722058559,
      "flos": 18880071528960.0,
      "grad_norm": 1.8202686101939696,
      "language_loss": 0.67626613,
      "learning_rate": 3.1852741981648776e-07,
      "loss": 0.69788247,
      "num_input_tokens_seen": 147857945,
      "step": 6847,
      "time_per_iteration": 2.458911418914795
    },
    {
      "auxiliary_loss_clip": 0.01112208,
      "auxiliary_loss_mlp": 0.01029079,
      "balance_loss_clip": 1.04267943,
      "balance_loss_mlp": 1.0218966,
      "epoch": 0.8234233150964949,
      "flos": 28439024757120.0,
      "grad_norm": 2.729353710444371,
      "language_loss": 0.69912225,
      "learning_rate": 3.1810577720467404e-07,
      "loss": 0.7205351,
      "num_input_tokens_seen": 147879675,
      "step": 6848,
      "time_per_iteration": 3.3158977031707764
    },
    {
      "auxiliary_loss_clip": 0.01138887,
      "auxiliary_loss_mlp": 0.01021392,
      "balance_loss_clip": 1.04517186,
      "balance_loss_mlp": 1.01412606,
      "epoch": 0.823543557987134,
      "flos": 33765941577600.0,
      "grad_norm": 10.5492874034828,
      "language_loss": 0.56379616,
      "learning_rate": 3.176843897348769e-07,
      "loss": 0.58539897,
      "num_input_tokens_seen": 147902870,
      "step": 6849,
      "time_per_iteration": 2.606759548187256
    },
    {
      "auxiliary_loss_clip": 0.011339,
      "auxiliary_loss_mlp": 0.01023834,
      "balance_loss_clip": 1.04406989,
      "balance_loss_mlp": 1.01666057,
      "epoch": 0.8236638008777731,
      "flos": 17092366611840.0,
      "grad_norm": 2.457616355858711,
      "language_loss": 0.75882953,
      "learning_rate": 3.1726325747102034e-07,
      "loss": 0.78040683,
      "num_input_tokens_seen": 147921245,
      "step": 6850,
      "time_per_iteration": 2.4447011947631836
    },
    {
      "auxiliary_loss_clip": 0.01100901,
      "auxiliary_loss_mlp": 0.01024609,
      "balance_loss_clip": 1.03642058,
      "balance_loss_mlp": 1.01736712,
      "epoch": 0.8237840437684122,
      "flos": 61639982334720.0,
      "grad_norm": 1.4723497310501072,
      "language_loss": 0.64271373,
      "learning_rate": 3.1684238047698974e-07,
      "loss": 0.66396886,
      "num_input_tokens_seen": 147949515,
      "step": 6851,
      "time_per_iteration": 2.912355661392212
    },
    {
      "auxiliary_loss_clip": 0.01140571,
      "auxiliary_loss_mlp": 0.01026097,
      "balance_loss_clip": 1.04617202,
      "balance_loss_mlp": 1.01889682,
      "epoch": 0.8239042866590512,
      "flos": 27309023821440.0,
      "grad_norm": 2.1365551628339623,
      "language_loss": 0.53172874,
      "learning_rate": 3.1642175881663155e-07,
      "loss": 0.55339539,
      "num_input_tokens_seen": 147969245,
      "step": 6852,
      "time_per_iteration": 2.53395938873291
    },
    {
      "auxiliary_loss_clip": 0.01162797,
      "auxiliary_loss_mlp": 0.01021581,
      "balance_loss_clip": 1.04615116,
      "balance_loss_mlp": 1.01508379,
      "epoch": 0.8240245295496904,
      "flos": 21726351187200.0,
      "grad_norm": 2.023049428612565,
      "language_loss": 0.83929038,
      "learning_rate": 3.160013925537537e-07,
      "loss": 0.86113411,
      "num_input_tokens_seen": 147990080,
      "step": 6853,
      "time_per_iteration": 2.416192054748535
    },
    {
      "auxiliary_loss_clip": 0.01123604,
      "auxiliary_loss_mlp": 0.01024113,
      "balance_loss_clip": 1.04085815,
      "balance_loss_mlp": 1.01683879,
      "epoch": 0.8241447724403295,
      "flos": 20009318279040.0,
      "grad_norm": 7.111965539277004,
      "language_loss": 0.75757122,
      "learning_rate": 3.155812817521266e-07,
      "loss": 0.77904838,
      "num_input_tokens_seen": 148010455,
      "step": 6854,
      "time_per_iteration": 2.501901388168335
    },
    {
      "auxiliary_loss_clip": 0.01138439,
      "auxiliary_loss_mlp": 0.0102572,
      "balance_loss_clip": 1.04483306,
      "balance_loss_mlp": 1.01877356,
      "epoch": 0.8242650153309685,
      "flos": 22272983337600.0,
      "grad_norm": 2.017624505195568,
      "language_loss": 0.78218204,
      "learning_rate": 3.151614264754787e-07,
      "loss": 0.80382371,
      "num_input_tokens_seen": 148028400,
      "step": 6855,
      "time_per_iteration": 2.4778809547424316
    },
    {
      "auxiliary_loss_clip": 0.01164009,
      "auxiliary_loss_mlp": 0.01024597,
      "balance_loss_clip": 1.04513669,
      "balance_loss_mlp": 1.01758194,
      "epoch": 0.8243852582216077,
      "flos": 22309971367680.0,
      "grad_norm": 2.4675703312436514,
      "language_loss": 0.79649591,
      "learning_rate": 3.147418267875035e-07,
      "loss": 0.81838197,
      "num_input_tokens_seen": 148046530,
      "step": 6856,
      "time_per_iteration": 2.4160687923431396
    },
    {
      "auxiliary_loss_clip": 0.01091824,
      "auxiliary_loss_mlp": 0.00761465,
      "balance_loss_clip": 1.03800154,
      "balance_loss_mlp": 1.00042939,
      "epoch": 0.8245055011122467,
      "flos": 24645421756800.0,
      "grad_norm": 2.167277393445466,
      "language_loss": 0.65251005,
      "learning_rate": 3.1432248275185315e-07,
      "loss": 0.67104286,
      "num_input_tokens_seen": 148067040,
      "step": 6857,
      "time_per_iteration": 2.6238596439361572
    },
    {
      "auxiliary_loss_clip": 0.01150927,
      "auxiliary_loss_mlp": 0.01023067,
      "balance_loss_clip": 1.04731131,
      "balance_loss_mlp": 1.01618898,
      "epoch": 0.8246257440028858,
      "flos": 17487275713920.0,
      "grad_norm": 2.049562297499048,
      "language_loss": 0.76828903,
      "learning_rate": 3.139033944321412e-07,
      "loss": 0.79002899,
      "num_input_tokens_seen": 148084400,
      "step": 6858,
      "time_per_iteration": 2.4194962978363037
    },
    {
      "auxiliary_loss_clip": 0.01152845,
      "auxiliary_loss_mlp": 0.01021198,
      "balance_loss_clip": 1.04540741,
      "balance_loss_mlp": 1.01443911,
      "epoch": 0.824745986893525,
      "flos": 25010130499200.0,
      "grad_norm": 1.639518671723465,
      "language_loss": 0.79117405,
      "learning_rate": 3.1348456189194507e-07,
      "loss": 0.81291449,
      "num_input_tokens_seen": 148104860,
      "step": 6859,
      "time_per_iteration": 2.4707069396972656
    },
    {
      "auxiliary_loss_clip": 0.01112755,
      "auxiliary_loss_mlp": 0.01021608,
      "balance_loss_clip": 1.03932667,
      "balance_loss_mlp": 1.01434231,
      "epoch": 0.824866229784164,
      "flos": 18772698798720.0,
      "grad_norm": 1.7934963059501927,
      "language_loss": 0.83034438,
      "learning_rate": 3.1306598519479876e-07,
      "loss": 0.85168803,
      "num_input_tokens_seen": 148124680,
      "step": 6860,
      "time_per_iteration": 2.5230376720428467
    },
    {
      "auxiliary_loss_clip": 0.01131201,
      "auxiliary_loss_mlp": 0.0102251,
      "balance_loss_clip": 1.04248166,
      "balance_loss_mlp": 1.01595318,
      "epoch": 0.8249864726748031,
      "flos": 23842171866240.0,
      "grad_norm": 3.4669129118230835,
      "language_loss": 0.78268087,
      "learning_rate": 3.1264766440420177e-07,
      "loss": 0.80421793,
      "num_input_tokens_seen": 148147150,
      "step": 6861,
      "time_per_iteration": 2.5132832527160645
    },
    {
      "auxiliary_loss_clip": 0.01149082,
      "auxiliary_loss_mlp": 0.01022166,
      "balance_loss_clip": 1.04645252,
      "balance_loss_mlp": 1.01529336,
      "epoch": 0.8251067155654422,
      "flos": 20303103617280.0,
      "grad_norm": 2.0856019670899046,
      "language_loss": 0.69264448,
      "learning_rate": 3.122295995836124e-07,
      "loss": 0.71435696,
      "num_input_tokens_seen": 148167020,
      "step": 6862,
      "time_per_iteration": 2.4347610473632812
    },
    {
      "auxiliary_loss_clip": 0.01154201,
      "auxiliary_loss_mlp": 0.01021424,
      "balance_loss_clip": 1.04321921,
      "balance_loss_mlp": 1.01381302,
      "epoch": 0.8252269584560813,
      "flos": 25009699536000.0,
      "grad_norm": 1.7768888128862097,
      "language_loss": 0.77480811,
      "learning_rate": 3.118117907964508e-07,
      "loss": 0.7965644,
      "num_input_tokens_seen": 148188965,
      "step": 6863,
      "time_per_iteration": 2.5595972537994385
    },
    {
      "auxiliary_loss_clip": 0.01129612,
      "auxiliary_loss_mlp": 0.01024955,
      "balance_loss_clip": 1.04345632,
      "balance_loss_mlp": 1.01807952,
      "epoch": 0.8253472013467203,
      "flos": 17128564542720.0,
      "grad_norm": 1.8906899471746743,
      "language_loss": 0.80131149,
      "learning_rate": 3.1139423810609856e-07,
      "loss": 0.82285714,
      "num_input_tokens_seen": 148205660,
      "step": 6864,
      "time_per_iteration": 2.5707077980041504
    },
    {
      "auxiliary_loss_clip": 0.01162816,
      "auxiliary_loss_mlp": 0.01025638,
      "balance_loss_clip": 1.04380798,
      "balance_loss_mlp": 1.01819897,
      "epoch": 0.8254674442373595,
      "flos": 22414794232320.0,
      "grad_norm": 1.8437331759961366,
      "language_loss": 0.75427389,
      "learning_rate": 3.1097694157589714e-07,
      "loss": 0.77615833,
      "num_input_tokens_seen": 148225545,
      "step": 6865,
      "time_per_iteration": 2.445305585861206
    },
    {
      "auxiliary_loss_clip": 0.01148257,
      "auxiliary_loss_mlp": 0.01027076,
      "balance_loss_clip": 1.04637527,
      "balance_loss_mlp": 1.01984286,
      "epoch": 0.8255876871279986,
      "flos": 24786765774720.0,
      "grad_norm": 2.964809227991256,
      "language_loss": 0.76846826,
      "learning_rate": 3.105599012691511e-07,
      "loss": 0.79022163,
      "num_input_tokens_seen": 148243975,
      "step": 6866,
      "time_per_iteration": 2.467745780944824
    },
    {
      "auxiliary_loss_clip": 0.01147489,
      "auxiliary_loss_mlp": 0.01023943,
      "balance_loss_clip": 1.04500008,
      "balance_loss_mlp": 1.01710033,
      "epoch": 0.8257079300186376,
      "flos": 27455431656960.0,
      "grad_norm": 1.4302936193903149,
      "language_loss": 0.82487613,
      "learning_rate": 3.101431172491249e-07,
      "loss": 0.84659046,
      "num_input_tokens_seen": 148265520,
      "step": 6867,
      "time_per_iteration": 3.3579022884368896
    },
    {
      "auxiliary_loss_clip": 0.011261,
      "auxiliary_loss_mlp": 0.00761875,
      "balance_loss_clip": 1.04161596,
      "balance_loss_mlp": 1.00042987,
      "epoch": 0.8258281729092768,
      "flos": 16471866142080.0,
      "grad_norm": 2.126919550353325,
      "language_loss": 0.71970677,
      "learning_rate": 3.097265895790444e-07,
      "loss": 0.73858654,
      "num_input_tokens_seen": 148283730,
      "step": 6868,
      "time_per_iteration": 2.537630558013916
    },
    {
      "auxiliary_loss_clip": 0.01123445,
      "auxiliary_loss_mlp": 0.01021809,
      "balance_loss_clip": 1.0421474,
      "balance_loss_mlp": 1.01530576,
      "epoch": 0.8259484157999158,
      "flos": 21433822824960.0,
      "grad_norm": 2.618214932771576,
      "language_loss": 0.83413196,
      "learning_rate": 3.093103183220962e-07,
      "loss": 0.85558456,
      "num_input_tokens_seen": 148303775,
      "step": 6869,
      "time_per_iteration": 3.3767173290252686
    },
    {
      "auxiliary_loss_clip": 0.01053424,
      "auxiliary_loss_mlp": 0.01000874,
      "balance_loss_clip": 1.00961125,
      "balance_loss_mlp": 0.99995059,
      "epoch": 0.8260686586905549,
      "flos": 58322342453760.0,
      "grad_norm": 0.8200754910891466,
      "language_loss": 0.5940001,
      "learning_rate": 3.0889430354142796e-07,
      "loss": 0.61454308,
      "num_input_tokens_seen": 148365285,
      "step": 6870,
      "time_per_iteration": 3.014164924621582
    },
    {
      "auxiliary_loss_clip": 0.01126464,
      "auxiliary_loss_mlp": 0.01024728,
      "balance_loss_clip": 1.04191077,
      "balance_loss_mlp": 1.01742935,
      "epoch": 0.826188901581194,
      "flos": 27527288814720.0,
      "grad_norm": 1.9459485876219063,
      "language_loss": 0.69888842,
      "learning_rate": 3.084785453001497e-07,
      "loss": 0.72040039,
      "num_input_tokens_seen": 148386200,
      "step": 6871,
      "time_per_iteration": 2.5698933601379395
    },
    {
      "auxiliary_loss_clip": 0.01137556,
      "auxiliary_loss_mlp": 0.00761863,
      "balance_loss_clip": 1.04683053,
      "balance_loss_mlp": 1.00045538,
      "epoch": 0.8263091444718331,
      "flos": 23696051339520.0,
      "grad_norm": 2.13352818553947,
      "language_loss": 0.82282352,
      "learning_rate": 3.080630436613314e-07,
      "loss": 0.84181774,
      "num_input_tokens_seen": 148403970,
      "step": 6872,
      "time_per_iteration": 2.4855942726135254
    },
    {
      "auxiliary_loss_clip": 0.0114296,
      "auxiliary_loss_mlp": 0.01028623,
      "balance_loss_clip": 1.04358542,
      "balance_loss_mlp": 1.02146769,
      "epoch": 0.8264293873624722,
      "flos": 17165157523200.0,
      "grad_norm": 2.547302559084706,
      "language_loss": 0.86318392,
      "learning_rate": 3.076477986880039e-07,
      "loss": 0.88489974,
      "num_input_tokens_seen": 148421765,
      "step": 6873,
      "time_per_iteration": 2.4098455905914307
    },
    {
      "auxiliary_loss_clip": 0.01135802,
      "auxiliary_loss_mlp": 0.01021283,
      "balance_loss_clip": 1.04510796,
      "balance_loss_mlp": 1.01429129,
      "epoch": 0.8265496302531112,
      "flos": 24098645952000.0,
      "grad_norm": 2.3262202204578513,
      "language_loss": 0.69370931,
      "learning_rate": 3.0723281044315986e-07,
      "loss": 0.71528018,
      "num_input_tokens_seen": 148443720,
      "step": 6874,
      "time_per_iteration": 3.2196807861328125
    },
    {
      "auxiliary_loss_clip": 0.01158421,
      "auxiliary_loss_mlp": 0.01023775,
      "balance_loss_clip": 1.04345608,
      "balance_loss_mlp": 1.01721573,
      "epoch": 0.8266698731437504,
      "flos": 14099894599680.0,
      "grad_norm": 2.0246735928259816,
      "language_loss": 0.76180756,
      "learning_rate": 3.068180789897521e-07,
      "loss": 0.78362954,
      "num_input_tokens_seen": 148462130,
      "step": 6875,
      "time_per_iteration": 2.3761699199676514
    },
    {
      "auxiliary_loss_clip": 0.01154858,
      "auxiliary_loss_mlp": 0.01022623,
      "balance_loss_clip": 1.04526854,
      "balance_loss_mlp": 1.01517797,
      "epoch": 0.8267901160343895,
      "flos": 30777563715840.0,
      "grad_norm": 1.4361712172831118,
      "language_loss": 0.81562686,
      "learning_rate": 3.064036043906966e-07,
      "loss": 0.83740163,
      "num_input_tokens_seen": 148485570,
      "step": 6876,
      "time_per_iteration": 2.5402393341064453
    },
    {
      "auxiliary_loss_clip": 0.01129314,
      "auxiliary_loss_mlp": 0.01025423,
      "balance_loss_clip": 1.04240394,
      "balance_loss_mlp": 1.01782906,
      "epoch": 0.8269103589250285,
      "flos": 40624915242240.0,
      "grad_norm": 2.376346023107187,
      "language_loss": 0.68122613,
      "learning_rate": 3.059893867088668e-07,
      "loss": 0.70277345,
      "num_input_tokens_seen": 148509715,
      "step": 6877,
      "time_per_iteration": 2.6738810539245605
    },
    {
      "auxiliary_loss_clip": 0.01147215,
      "auxiliary_loss_mlp": 0.01024467,
      "balance_loss_clip": 1.04483867,
      "balance_loss_mlp": 1.01801217,
      "epoch": 0.8270306018156677,
      "flos": 30263645877120.0,
      "grad_norm": 2.17552021079681,
      "language_loss": 0.66741371,
      "learning_rate": 3.055754260071004e-07,
      "loss": 0.68913054,
      "num_input_tokens_seen": 148532010,
      "step": 6878,
      "time_per_iteration": 2.506004571914673
    },
    {
      "auxiliary_loss_clip": 0.01150462,
      "auxiliary_loss_mlp": 0.0102653,
      "balance_loss_clip": 1.04599977,
      "balance_loss_mlp": 1.01986313,
      "epoch": 0.8271508447063067,
      "flos": 25226599812480.0,
      "grad_norm": 4.579153493549264,
      "language_loss": 0.7360543,
      "learning_rate": 3.051617223481948e-07,
      "loss": 0.75782418,
      "num_input_tokens_seen": 148553330,
      "step": 6879,
      "time_per_iteration": 2.4655911922454834
    },
    {
      "auxiliary_loss_clip": 0.01134737,
      "auxiliary_loss_mlp": 0.01029523,
      "balance_loss_clip": 1.04430628,
      "balance_loss_mlp": 1.02197146,
      "epoch": 0.8272710875969458,
      "flos": 17566602900480.0,
      "grad_norm": 2.055779408794126,
      "language_loss": 0.75496978,
      "learning_rate": 3.047482757949078e-07,
      "loss": 0.7766124,
      "num_input_tokens_seen": 148570960,
      "step": 6880,
      "time_per_iteration": 2.485534429550171
    },
    {
      "auxiliary_loss_clip": 0.01116602,
      "auxiliary_loss_mlp": 0.00761293,
      "balance_loss_clip": 1.03960729,
      "balance_loss_mlp": 1.00044405,
      "epoch": 0.827391330487585,
      "flos": 19755465886080.0,
      "grad_norm": 2.143221753535972,
      "language_loss": 0.85831439,
      "learning_rate": 3.043350864099605e-07,
      "loss": 0.87709337,
      "num_input_tokens_seen": 148589520,
      "step": 6881,
      "time_per_iteration": 2.491403102874756
    },
    {
      "auxiliary_loss_clip": 0.01152137,
      "auxiliary_loss_mlp": 0.01022256,
      "balance_loss_clip": 1.04414189,
      "balance_loss_mlp": 1.01538384,
      "epoch": 0.827511573378224,
      "flos": 16835174254080.0,
      "grad_norm": 2.165174932197664,
      "language_loss": 0.8078407,
      "learning_rate": 3.039221542560315e-07,
      "loss": 0.82958466,
      "num_input_tokens_seen": 148606085,
      "step": 6882,
      "time_per_iteration": 2.398483991622925
    },
    {
      "auxiliary_loss_clip": 0.0115022,
      "auxiliary_loss_mlp": 0.01021935,
      "balance_loss_clip": 1.04638052,
      "balance_loss_mlp": 1.0149138,
      "epoch": 0.8276318162688631,
      "flos": 18369242259840.0,
      "grad_norm": 2.210022440403164,
      "language_loss": 0.73260939,
      "learning_rate": 3.0350947939576356e-07,
      "loss": 0.75433099,
      "num_input_tokens_seen": 148625240,
      "step": 6883,
      "time_per_iteration": 2.4215855598449707
    },
    {
      "auxiliary_loss_clip": 0.01156976,
      "auxiliary_loss_mlp": 0.01027053,
      "balance_loss_clip": 1.04678643,
      "balance_loss_mlp": 1.01920927,
      "epoch": 0.8277520591595022,
      "flos": 19352691705600.0,
      "grad_norm": 1.5453395721510128,
      "language_loss": 0.72249901,
      "learning_rate": 3.0309706189175876e-07,
      "loss": 0.74433929,
      "num_input_tokens_seen": 148645075,
      "step": 6884,
      "time_per_iteration": 2.438347339630127
    },
    {
      "auxiliary_loss_clip": 0.01044694,
      "auxiliary_loss_mlp": 0.01001846,
      "balance_loss_clip": 1.0100615,
      "balance_loss_mlp": 1.00084484,
      "epoch": 0.8278723020501413,
      "flos": 67918858329600.0,
      "grad_norm": 0.7623672521857662,
      "language_loss": 0.57401317,
      "learning_rate": 3.0268490180658045e-07,
      "loss": 0.59447861,
      "num_input_tokens_seen": 148707855,
      "step": 6885,
      "time_per_iteration": 3.0473792552948
    },
    {
      "auxiliary_loss_clip": 0.01169552,
      "auxiliary_loss_mlp": 0.01022055,
      "balance_loss_clip": 1.04985881,
      "balance_loss_mlp": 1.01508081,
      "epoch": 0.8279925449407803,
      "flos": 18185738653440.0,
      "grad_norm": 2.4562757559457586,
      "language_loss": 0.7903136,
      "learning_rate": 3.0227299920275305e-07,
      "loss": 0.81222963,
      "num_input_tokens_seen": 148724170,
      "step": 6886,
      "time_per_iteration": 2.370415210723877
    },
    {
      "auxiliary_loss_clip": 0.01128666,
      "auxiliary_loss_mlp": 0.01028614,
      "balance_loss_clip": 1.04585683,
      "balance_loss_mlp": 1.02101135,
      "epoch": 0.8281127878314195,
      "flos": 20631434860800.0,
      "grad_norm": 1.8884400687405891,
      "language_loss": 0.85513437,
      "learning_rate": 3.018613541427613e-07,
      "loss": 0.87670708,
      "num_input_tokens_seen": 148743690,
      "step": 6887,
      "time_per_iteration": 2.5087547302246094
    },
    {
      "auxiliary_loss_clip": 0.0116309,
      "auxiliary_loss_mlp": 0.01024937,
      "balance_loss_clip": 1.04531491,
      "balance_loss_mlp": 1.01805305,
      "epoch": 0.8282330307220586,
      "flos": 18004282122240.0,
      "grad_norm": 1.6714680723685145,
      "language_loss": 0.73599643,
      "learning_rate": 3.0144996668905243e-07,
      "loss": 0.75787669,
      "num_input_tokens_seen": 148761070,
      "step": 6888,
      "time_per_iteration": 2.4022560119628906
    },
    {
      "auxiliary_loss_clip": 0.01097036,
      "auxiliary_loss_mlp": 0.0076154,
      "balance_loss_clip": 1.03731155,
      "balance_loss_mlp": 1.00039434,
      "epoch": 0.8283532736126976,
      "flos": 20084120352000.0,
      "grad_norm": 3.183612837149206,
      "language_loss": 0.82138085,
      "learning_rate": 3.010388369040331e-07,
      "loss": 0.83996665,
      "num_input_tokens_seen": 148779730,
      "step": 6889,
      "time_per_iteration": 2.60994291305542
    },
    {
      "auxiliary_loss_clip": 0.01152733,
      "auxiliary_loss_mlp": 0.01024846,
      "balance_loss_clip": 1.04653656,
      "balance_loss_mlp": 1.01799452,
      "epoch": 0.8284735165033368,
      "flos": 31868421805440.0,
      "grad_norm": 1.6621977960126766,
      "language_loss": 0.82844639,
      "learning_rate": 3.0062796485007156e-07,
      "loss": 0.85022223,
      "num_input_tokens_seen": 148800670,
      "step": 6890,
      "time_per_iteration": 2.541858196258545
    },
    {
      "auxiliary_loss_clip": 0.01164281,
      "auxiliary_loss_mlp": 0.00761771,
      "balance_loss_clip": 1.04595208,
      "balance_loss_mlp": 1.00039065,
      "epoch": 0.8285937593939758,
      "flos": 26651319840000.0,
      "grad_norm": 3.8447576595839563,
      "language_loss": 0.65936232,
      "learning_rate": 3.002173505894965e-07,
      "loss": 0.67862284,
      "num_input_tokens_seen": 148819820,
      "step": 6891,
      "time_per_iteration": 2.4771533012390137
    },
    {
      "auxiliary_loss_clip": 0.01154038,
      "auxiliary_loss_mlp": 0.01023016,
      "balance_loss_clip": 1.04346776,
      "balance_loss_mlp": 1.01531482,
      "epoch": 0.8287140022846149,
      "flos": 20193683811840.0,
      "grad_norm": 9.186926026801611,
      "language_loss": 0.62150168,
      "learning_rate": 2.998069941845973e-07,
      "loss": 0.64327222,
      "num_input_tokens_seen": 148838890,
      "step": 6892,
      "time_per_iteration": 2.45940899848938
    },
    {
      "auxiliary_loss_clip": 0.01062982,
      "auxiliary_loss_mlp": 0.01001451,
      "balance_loss_clip": 1.00901604,
      "balance_loss_mlp": 1.00048542,
      "epoch": 0.8288342451752541,
      "flos": 70755980019840.0,
      "grad_norm": 0.7093223707026493,
      "language_loss": 0.57449889,
      "learning_rate": 2.993968956976258e-07,
      "loss": 0.5951432,
      "num_input_tokens_seen": 148906635,
      "step": 6893,
      "time_per_iteration": 3.120269775390625
    },
    {
      "auxiliary_loss_clip": 0.01170626,
      "auxiliary_loss_mlp": 0.01023406,
      "balance_loss_clip": 1.04820871,
      "balance_loss_mlp": 1.01537752,
      "epoch": 0.8289544880658931,
      "flos": 24572235795840.0,
      "grad_norm": 1.7797692418218232,
      "language_loss": 0.70151687,
      "learning_rate": 2.9898705519079313e-07,
      "loss": 0.72345722,
      "num_input_tokens_seen": 148925740,
      "step": 6894,
      "time_per_iteration": 3.3278324604034424
    },
    {
      "auxiliary_loss_clip": 0.01130049,
      "auxiliary_loss_mlp": 0.01021247,
      "balance_loss_clip": 1.04206598,
      "balance_loss_mlp": 1.01422596,
      "epoch": 0.8290747309565322,
      "flos": 22273378387200.0,
      "grad_norm": 1.8173374357746075,
      "language_loss": 0.74415332,
      "learning_rate": 2.985774727262715e-07,
      "loss": 0.76566625,
      "num_input_tokens_seen": 148944585,
      "step": 6895,
      "time_per_iteration": 2.473827838897705
    },
    {
      "auxiliary_loss_clip": 0.01163261,
      "auxiliary_loss_mlp": 0.01023308,
      "balance_loss_clip": 1.04610658,
      "balance_loss_mlp": 1.01673973,
      "epoch": 0.8291949738471713,
      "flos": 23255570856960.0,
      "grad_norm": 1.8623617111640176,
      "language_loss": 0.81484956,
      "learning_rate": 2.981681483661949e-07,
      "loss": 0.83671522,
      "num_input_tokens_seen": 148964170,
      "step": 6896,
      "time_per_iteration": 3.122509717941284
    },
    {
      "auxiliary_loss_clip": 0.01152999,
      "auxiliary_loss_mlp": 0.01031486,
      "balance_loss_clip": 1.04818177,
      "balance_loss_mlp": 1.02474427,
      "epoch": 0.8293152167378104,
      "flos": 52555768185600.0,
      "grad_norm": 1.7380999878035768,
      "language_loss": 0.71114653,
      "learning_rate": 2.9775908217265633e-07,
      "loss": 0.7329914,
      "num_input_tokens_seen": 148989405,
      "step": 6897,
      "time_per_iteration": 3.413006544113159
    },
    {
      "auxiliary_loss_clip": 0.01008657,
      "auxiliary_loss_mlp": 0.01002649,
      "balance_loss_clip": 1.00801134,
      "balance_loss_mlp": 1.00151622,
      "epoch": 0.8294354596284494,
      "flos": 63356156294400.0,
      "grad_norm": 0.8240186140067517,
      "language_loss": 0.50400698,
      "learning_rate": 2.9735027420771253e-07,
      "loss": 0.52412009,
      "num_input_tokens_seen": 149049740,
      "step": 6898,
      "time_per_iteration": 3.0846452713012695
    },
    {
      "auxiliary_loss_clip": 0.01133749,
      "auxiliary_loss_mlp": 0.01028298,
      "balance_loss_clip": 1.04777813,
      "balance_loss_mlp": 1.02184868,
      "epoch": 0.8295557025190886,
      "flos": 24827021942400.0,
      "grad_norm": 1.7845273763816505,
      "language_loss": 0.71218121,
      "learning_rate": 2.969417245333774e-07,
      "loss": 0.73380172,
      "num_input_tokens_seen": 149069120,
      "step": 6899,
      "time_per_iteration": 2.501249074935913
    },
    {
      "auxiliary_loss_clip": 0.01121774,
      "auxiliary_loss_mlp": 0.0102371,
      "balance_loss_clip": 1.04458678,
      "balance_loss_mlp": 1.01688242,
      "epoch": 0.8296759454097277,
      "flos": 25118580637440.0,
      "grad_norm": 2.0046030871539307,
      "language_loss": 0.77855927,
      "learning_rate": 2.9653343321162915e-07,
      "loss": 0.80001408,
      "num_input_tokens_seen": 149088630,
      "step": 6900,
      "time_per_iteration": 2.515753984451294
    },
    {
      "auxiliary_loss_clip": 0.01125178,
      "auxiliary_loss_mlp": 0.01021405,
      "balance_loss_clip": 1.04566622,
      "balance_loss_mlp": 1.01363289,
      "epoch": 0.8297961883003667,
      "flos": 24132581326080.0,
      "grad_norm": 4.005801740255351,
      "language_loss": 0.65104586,
      "learning_rate": 2.9612540030440446e-07,
      "loss": 0.67251164,
      "num_input_tokens_seen": 149109175,
      "step": 6901,
      "time_per_iteration": 3.291018486022949
    },
    {
      "auxiliary_loss_clip": 0.01042717,
      "auxiliary_loss_mlp": 0.01001102,
      "balance_loss_clip": 1.00893545,
      "balance_loss_mlp": 1.00011218,
      "epoch": 0.8299164311910058,
      "flos": 67446561375360.0,
      "grad_norm": 0.8973955168334014,
      "language_loss": 0.64099193,
      "learning_rate": 2.9571762587360206e-07,
      "loss": 0.66143012,
      "num_input_tokens_seen": 149165560,
      "step": 6902,
      "time_per_iteration": 3.005819320678711
    },
    {
      "auxiliary_loss_clip": 0.01103939,
      "auxiliary_loss_mlp": 0.01024169,
      "balance_loss_clip": 1.03537178,
      "balance_loss_mlp": 1.01749635,
      "epoch": 0.8300366740816449,
      "flos": 25228682801280.0,
      "grad_norm": 1.796975921297616,
      "language_loss": 0.74119967,
      "learning_rate": 2.953101099810806e-07,
      "loss": 0.76248074,
      "num_input_tokens_seen": 149185165,
      "step": 6903,
      "time_per_iteration": 2.5984020233154297
    },
    {
      "auxiliary_loss_clip": 0.01145991,
      "auxiliary_loss_mlp": 0.01025332,
      "balance_loss_clip": 1.046229,
      "balance_loss_mlp": 1.01837027,
      "epoch": 0.830156916972284,
      "flos": 18041018757120.0,
      "grad_norm": 2.4255223080633463,
      "language_loss": 0.82811439,
      "learning_rate": 2.9490285268865965e-07,
      "loss": 0.84982765,
      "num_input_tokens_seen": 149202655,
      "step": 6904,
      "time_per_iteration": 2.437838077545166
    },
    {
      "auxiliary_loss_clip": 0.01156831,
      "auxiliary_loss_mlp": 0.01019836,
      "balance_loss_clip": 1.04906797,
      "balance_loss_mlp": 1.01275253,
      "epoch": 0.830277159862923,
      "flos": 26322485806080.0,
      "grad_norm": 2.039131686130138,
      "language_loss": 0.79322606,
      "learning_rate": 2.9449585405812085e-07,
      "loss": 0.81499279,
      "num_input_tokens_seen": 149220035,
      "step": 6905,
      "time_per_iteration": 2.4953598976135254
    },
    {
      "auxiliary_loss_clip": 0.01129367,
      "auxiliary_loss_mlp": 0.01019925,
      "balance_loss_clip": 1.04520488,
      "balance_loss_mlp": 1.01289141,
      "epoch": 0.8303974027535622,
      "flos": 19938861751680.0,
      "grad_norm": 1.9171193094958274,
      "language_loss": 0.73835111,
      "learning_rate": 2.940891141512043e-07,
      "loss": 0.75984395,
      "num_input_tokens_seen": 149238055,
      "step": 6906,
      "time_per_iteration": 2.5078468322753906
    },
    {
      "auxiliary_loss_clip": 0.01136621,
      "auxiliary_loss_mlp": 0.01026169,
      "balance_loss_clip": 1.04362547,
      "balance_loss_mlp": 1.01843262,
      "epoch": 0.8305176456442013,
      "flos": 17165552572800.0,
      "grad_norm": 3.06073748522985,
      "language_loss": 0.71972716,
      "learning_rate": 2.9368263302961385e-07,
      "loss": 0.74135506,
      "num_input_tokens_seen": 149256755,
      "step": 6907,
      "time_per_iteration": 2.4663076400756836
    },
    {
      "auxiliary_loss_clip": 0.01096808,
      "auxiliary_loss_mlp": 0.01019424,
      "balance_loss_clip": 1.03937244,
      "balance_loss_mlp": 1.01244736,
      "epoch": 0.8306378885348403,
      "flos": 25627614226560.0,
      "grad_norm": 1.9844074649284424,
      "language_loss": 0.80027926,
      "learning_rate": 2.9327641075501075e-07,
      "loss": 0.82144165,
      "num_input_tokens_seen": 149275745,
      "step": 6908,
      "time_per_iteration": 2.6545732021331787
    },
    {
      "auxiliary_loss_clip": 0.01129501,
      "auxiliary_loss_mlp": 0.01030973,
      "balance_loss_clip": 1.04019618,
      "balance_loss_mlp": 1.0232538,
      "epoch": 0.8307581314254795,
      "flos": 33947864985600.0,
      "grad_norm": 2.362089000669139,
      "language_loss": 0.66148579,
      "learning_rate": 2.9287044738901866e-07,
      "loss": 0.68309051,
      "num_input_tokens_seen": 149293730,
      "step": 6909,
      "time_per_iteration": 2.5797693729400635
    },
    {
      "auxiliary_loss_clip": 0.01152017,
      "auxiliary_loss_mlp": 0.00761515,
      "balance_loss_clip": 1.04486823,
      "balance_loss_mlp": 1.00038576,
      "epoch": 0.8308783743161186,
      "flos": 17562724231680.0,
      "grad_norm": 1.860933637122064,
      "language_loss": 0.90652752,
      "learning_rate": 2.9246474299322274e-07,
      "loss": 0.92566288,
      "num_input_tokens_seen": 149309290,
      "step": 6910,
      "time_per_iteration": 2.4368836879730225
    },
    {
      "auxiliary_loss_clip": 0.0102841,
      "auxiliary_loss_mlp": 0.01002087,
      "balance_loss_clip": 1.0083518,
      "balance_loss_mlp": 1.00103843,
      "epoch": 0.8309986172067576,
      "flos": 69412885649280.0,
      "grad_norm": 0.8961285096441381,
      "language_loss": 0.63182837,
      "learning_rate": 2.920592976291678e-07,
      "loss": 0.65213335,
      "num_input_tokens_seen": 149366620,
      "step": 6911,
      "time_per_iteration": 3.0481698513031006
    },
    {
      "auxiliary_loss_clip": 0.01150029,
      "auxiliary_loss_mlp": 0.01027453,
      "balance_loss_clip": 1.04512262,
      "balance_loss_mlp": 1.020208,
      "epoch": 0.8311188600973968,
      "flos": 22309755886080.0,
      "grad_norm": 1.8766682038346312,
      "language_loss": 0.80626452,
      "learning_rate": 2.916541113583595e-07,
      "loss": 0.82803929,
      "num_input_tokens_seen": 149385120,
      "step": 6912,
      "time_per_iteration": 2.4464707374572754
    },
    {
      "auxiliary_loss_clip": 0.01129246,
      "auxiliary_loss_mlp": 0.01023349,
      "balance_loss_clip": 1.04649973,
      "balance_loss_mlp": 1.01607394,
      "epoch": 0.8312391029880358,
      "flos": 18770077105920.0,
      "grad_norm": 2.5345308066752987,
      "language_loss": 0.66431165,
      "learning_rate": 2.912491842422642e-07,
      "loss": 0.68583763,
      "num_input_tokens_seen": 149402825,
      "step": 6913,
      "time_per_iteration": 2.5723748207092285
    },
    {
      "auxiliary_loss_clip": 0.01152995,
      "auxiliary_loss_mlp": 0.01023288,
      "balance_loss_clip": 1.04673028,
      "balance_loss_mlp": 1.01606369,
      "epoch": 0.8313593458786749,
      "flos": 20376648714240.0,
      "grad_norm": 1.7016825346540787,
      "language_loss": 0.70998251,
      "learning_rate": 2.9084451634230857e-07,
      "loss": 0.73174536,
      "num_input_tokens_seen": 149422125,
      "step": 6914,
      "time_per_iteration": 2.456085205078125
    },
    {
      "auxiliary_loss_clip": 0.01121059,
      "auxiliary_loss_mlp": 0.01027056,
      "balance_loss_clip": 1.04170632,
      "balance_loss_mlp": 1.0198977,
      "epoch": 0.831479588769314,
      "flos": 32124069878400.0,
      "grad_norm": 2.970555828884649,
      "language_loss": 0.71587288,
      "learning_rate": 2.9044010771988125e-07,
      "loss": 0.73735398,
      "num_input_tokens_seen": 149441940,
      "step": 6915,
      "time_per_iteration": 2.627363443374634
    },
    {
      "auxiliary_loss_clip": 0.01130238,
      "auxiliary_loss_mlp": 0.01025398,
      "balance_loss_clip": 1.04355609,
      "balance_loss_mlp": 1.01846337,
      "epoch": 0.8315998316599531,
      "flos": 45185929338240.0,
      "grad_norm": 1.709565035641112,
      "language_loss": 0.72039658,
      "learning_rate": 2.900359584363303e-07,
      "loss": 0.7419529,
      "num_input_tokens_seen": 149465045,
      "step": 6916,
      "time_per_iteration": 2.6957600116729736
    },
    {
      "auxiliary_loss_clip": 0.01106916,
      "auxiliary_loss_mlp": 0.01031515,
      "balance_loss_clip": 1.04366982,
      "balance_loss_mlp": 1.02343297,
      "epoch": 0.8317200745505922,
      "flos": 18363747479040.0,
      "grad_norm": 2.0739871062153847,
      "language_loss": 0.84478104,
      "learning_rate": 2.8963206855296494e-07,
      "loss": 0.86616528,
      "num_input_tokens_seen": 149481285,
      "step": 6917,
      "time_per_iteration": 2.498197078704834
    },
    {
      "auxiliary_loss_clip": 0.01151923,
      "auxiliary_loss_mlp": 0.01028665,
      "balance_loss_clip": 1.04496694,
      "balance_loss_mlp": 1.02136922,
      "epoch": 0.8318403174412313,
      "flos": 24206557386240.0,
      "grad_norm": 2.164780494814873,
      "language_loss": 0.7699967,
      "learning_rate": 2.892284381310548e-07,
      "loss": 0.79180259,
      "num_input_tokens_seen": 149502700,
      "step": 6918,
      "time_per_iteration": 2.459580898284912
    },
    {
      "auxiliary_loss_clip": 0.01131532,
      "auxiliary_loss_mlp": 0.01025444,
      "balance_loss_clip": 1.04203439,
      "balance_loss_mlp": 1.01781154,
      "epoch": 0.8319605603318704,
      "flos": 22418780641920.0,
      "grad_norm": 2.5551873929896742,
      "language_loss": 0.72564185,
      "learning_rate": 2.888250672318302e-07,
      "loss": 0.74721158,
      "num_input_tokens_seen": 149520100,
      "step": 6919,
      "time_per_iteration": 2.4520742893218994
    },
    {
      "auxiliary_loss_clip": 0.01169981,
      "auxiliary_loss_mlp": 0.0102842,
      "balance_loss_clip": 1.05039489,
      "balance_loss_mlp": 1.02141023,
      "epoch": 0.8320808032225094,
      "flos": 37414501459200.0,
      "grad_norm": 1.7787985402648339,
      "language_loss": 0.6867522,
      "learning_rate": 2.884219559164831e-07,
      "loss": 0.70873618,
      "num_input_tokens_seen": 149543245,
      "step": 6920,
      "time_per_iteration": 2.5364813804626465
    },
    {
      "auxiliary_loss_clip": 0.01150751,
      "auxiliary_loss_mlp": 0.01022845,
      "balance_loss_clip": 1.04656553,
      "balance_loss_mlp": 1.01604974,
      "epoch": 0.8322010461131486,
      "flos": 12787395638400.0,
      "grad_norm": 1.9305490846526931,
      "language_loss": 0.81247079,
      "learning_rate": 2.880191042461635e-07,
      "loss": 0.83420676,
      "num_input_tokens_seen": 149559185,
      "step": 6921,
      "time_per_iteration": 3.244818925857544
    },
    {
      "auxiliary_loss_clip": 0.01113086,
      "auxiliary_loss_mlp": 0.01022549,
      "balance_loss_clip": 1.04098594,
      "balance_loss_mlp": 1.01606703,
      "epoch": 0.8323212890037877,
      "flos": 15815455050240.0,
      "grad_norm": 1.64007199583088,
      "language_loss": 0.79980624,
      "learning_rate": 2.876165122819849e-07,
      "loss": 0.82116258,
      "num_input_tokens_seen": 149577165,
      "step": 6922,
      "time_per_iteration": 2.516418933868408
    },
    {
      "auxiliary_loss_clip": 0.01163232,
      "auxiliary_loss_mlp": 0.0101988,
      "balance_loss_clip": 1.04628515,
      "balance_loss_mlp": 1.0132308,
      "epoch": 0.8324415318944267,
      "flos": 21719276208000.0,
      "grad_norm": 1.7273990896534759,
      "language_loss": 0.79559994,
      "learning_rate": 2.872141800850201e-07,
      "loss": 0.81743103,
      "num_input_tokens_seen": 149594340,
      "step": 6923,
      "time_per_iteration": 3.223858118057251
    },
    {
      "auxiliary_loss_clip": 0.01164221,
      "auxiliary_loss_mlp": 0.01023564,
      "balance_loss_clip": 1.04692876,
      "balance_loss_mlp": 1.01726723,
      "epoch": 0.8325617747850659,
      "flos": 34198700636160.0,
      "grad_norm": 2.0780732159364854,
      "language_loss": 0.73479724,
      "learning_rate": 2.868121077163024e-07,
      "loss": 0.75667512,
      "num_input_tokens_seen": 149613895,
      "step": 6924,
      "time_per_iteration": 3.296491861343384
    },
    {
      "auxiliary_loss_clip": 0.01154852,
      "auxiliary_loss_mlp": 0.01028769,
      "balance_loss_clip": 1.0455606,
      "balance_loss_mlp": 1.02177751,
      "epoch": 0.8326820176757049,
      "flos": 18369457741440.0,
      "grad_norm": 1.7263277654853793,
      "language_loss": 0.72476411,
      "learning_rate": 2.864102952368257e-07,
      "loss": 0.74660033,
      "num_input_tokens_seen": 149631820,
      "step": 6925,
      "time_per_iteration": 2.4176251888275146
    },
    {
      "auxiliary_loss_clip": 0.0109671,
      "auxiliary_loss_mlp": 0.01025103,
      "balance_loss_clip": 1.0358516,
      "balance_loss_mlp": 1.01763415,
      "epoch": 0.832802260566344,
      "flos": 35991325716480.0,
      "grad_norm": 1.32661437481317,
      "language_loss": 0.59605002,
      "learning_rate": 2.860087427075444e-07,
      "loss": 0.61726815,
      "num_input_tokens_seen": 149656070,
      "step": 6926,
      "time_per_iteration": 2.6803042888641357
    },
    {
      "auxiliary_loss_clip": 0.01131479,
      "auxiliary_loss_mlp": 0.01027585,
      "balance_loss_clip": 1.0436573,
      "balance_loss_mlp": 1.02074826,
      "epoch": 0.8329225034569832,
      "flos": 14244434928000.0,
      "grad_norm": 2.473242672375076,
      "language_loss": 0.86173701,
      "learning_rate": 2.856074501893744e-07,
      "loss": 0.8833276,
      "num_input_tokens_seen": 149671270,
      "step": 6927,
      "time_per_iteration": 2.436058521270752
    },
    {
      "auxiliary_loss_clip": 0.01155315,
      "auxiliary_loss_mlp": 0.01023799,
      "balance_loss_clip": 1.04841924,
      "balance_loss_mlp": 1.01699221,
      "epoch": 0.8330427463476222,
      "flos": 18077468083200.0,
      "grad_norm": 1.6385088483415946,
      "language_loss": 0.81599665,
      "learning_rate": 2.8520641774319054e-07,
      "loss": 0.83778775,
      "num_input_tokens_seen": 149689360,
      "step": 6928,
      "time_per_iteration": 3.199218511581421
    },
    {
      "auxiliary_loss_clip": 0.01137855,
      "auxiliary_loss_mlp": 0.01024733,
      "balance_loss_clip": 1.04070127,
      "balance_loss_mlp": 1.01763415,
      "epoch": 0.8331629892382613,
      "flos": 18040839189120.0,
      "grad_norm": 2.275202050732897,
      "language_loss": 0.75433743,
      "learning_rate": 2.848056454298309e-07,
      "loss": 0.77596325,
      "num_input_tokens_seen": 149706685,
      "step": 6929,
      "time_per_iteration": 2.4604151248931885
    },
    {
      "auxiliary_loss_clip": 0.0113524,
      "auxiliary_loss_mlp": 0.0102128,
      "balance_loss_clip": 1.04547763,
      "balance_loss_mlp": 1.0141964,
      "epoch": 0.8332832321289004,
      "flos": 17457398576640.0,
      "grad_norm": 1.8827614930271386,
      "language_loss": 0.65574348,
      "learning_rate": 2.844051333100905e-07,
      "loss": 0.67730868,
      "num_input_tokens_seen": 149724230,
      "step": 6930,
      "time_per_iteration": 2.449049711227417
    },
    {
      "auxiliary_loss_clip": 0.01138384,
      "auxiliary_loss_mlp": 0.01021326,
      "balance_loss_clip": 1.04648757,
      "balance_loss_mlp": 1.01484096,
      "epoch": 0.8334034750195395,
      "flos": 15084852416640.0,
      "grad_norm": 1.9847170586925396,
      "language_loss": 0.8394497,
      "learning_rate": 2.840048814447269e-07,
      "loss": 0.86104679,
      "num_input_tokens_seen": 149742395,
      "step": 6931,
      "time_per_iteration": 2.453190565109253
    },
    {
      "auxiliary_loss_clip": 0.01131392,
      "auxiliary_loss_mlp": 0.01021846,
      "balance_loss_clip": 1.04281545,
      "balance_loss_mlp": 1.0149827,
      "epoch": 0.8335237179101785,
      "flos": 19427170556160.0,
      "grad_norm": 2.315069979658486,
      "language_loss": 0.739308,
      "learning_rate": 2.836048898944587e-07,
      "loss": 0.76084042,
      "num_input_tokens_seen": 149760820,
      "step": 6932,
      "time_per_iteration": 2.454880952835083
    },
    {
      "auxiliary_loss_clip": 0.0113615,
      "auxiliary_loss_mlp": 0.01024557,
      "balance_loss_clip": 1.04366779,
      "balance_loss_mlp": 1.01809025,
      "epoch": 0.8336439608008177,
      "flos": 21762046327680.0,
      "grad_norm": 2.6239200687743787,
      "language_loss": 0.72031128,
      "learning_rate": 2.832051587199642e-07,
      "loss": 0.74191839,
      "num_input_tokens_seen": 149778075,
      "step": 6933,
      "time_per_iteration": 2.458258867263794
    },
    {
      "auxiliary_loss_clip": 0.01053453,
      "auxiliary_loss_mlp": 0.01002175,
      "balance_loss_clip": 1.00882685,
      "balance_loss_mlp": 1.00120354,
      "epoch": 0.8337642036914568,
      "flos": 59702783990400.0,
      "grad_norm": 0.8098193649508391,
      "language_loss": 0.57789481,
      "learning_rate": 2.828056879818821e-07,
      "loss": 0.59845108,
      "num_input_tokens_seen": 149837150,
      "step": 6934,
      "time_per_iteration": 2.9578049182891846
    },
    {
      "auxiliary_loss_clip": 0.01121406,
      "auxiliary_loss_mlp": 0.01024675,
      "balance_loss_clip": 1.03947902,
      "balance_loss_mlp": 1.01855052,
      "epoch": 0.8338844465820958,
      "flos": 27162185022720.0,
      "grad_norm": 1.729355543817932,
      "language_loss": 0.83444381,
      "learning_rate": 2.824064777408117e-07,
      "loss": 0.85590464,
      "num_input_tokens_seen": 149856940,
      "step": 6935,
      "time_per_iteration": 2.5528767108917236
    },
    {
      "auxiliary_loss_clip": 0.01150406,
      "auxiliary_loss_mlp": 0.01024755,
      "balance_loss_clip": 1.04656959,
      "balance_loss_mlp": 1.01760292,
      "epoch": 0.8340046894727349,
      "flos": 30481264425600.0,
      "grad_norm": 1.8801422958469023,
      "language_loss": 0.75670588,
      "learning_rate": 2.8200752805731263e-07,
      "loss": 0.77845752,
      "num_input_tokens_seen": 149879930,
      "step": 6936,
      "time_per_iteration": 2.507885456085205
    },
    {
      "auxiliary_loss_clip": 0.01149698,
      "auxiliary_loss_mlp": 0.01023174,
      "balance_loss_clip": 1.04594362,
      "balance_loss_mlp": 1.01621246,
      "epoch": 0.834124932363374,
      "flos": 27126166659840.0,
      "grad_norm": 1.6709066521178355,
      "language_loss": 0.81128395,
      "learning_rate": 2.8160883899190625e-07,
      "loss": 0.8330127,
      "num_input_tokens_seen": 149903200,
      "step": 6937,
      "time_per_iteration": 2.4954211711883545
    },
    {
      "auxiliary_loss_clip": 0.01116148,
      "auxiliary_loss_mlp": 0.01024616,
      "balance_loss_clip": 1.04419136,
      "balance_loss_mlp": 1.0178299,
      "epoch": 0.8342451752540131,
      "flos": 24569865498240.0,
      "grad_norm": 2.1986581566204837,
      "language_loss": 0.72984147,
      "learning_rate": 2.8121041060507234e-07,
      "loss": 0.75124907,
      "num_input_tokens_seen": 149922230,
      "step": 6938,
      "time_per_iteration": 2.5200958251953125
    },
    {
      "auxiliary_loss_clip": 0.01155553,
      "auxiliary_loss_mlp": 0.01021479,
      "balance_loss_clip": 1.0454793,
      "balance_loss_mlp": 1.01440072,
      "epoch": 0.8343654181446521,
      "flos": 26615085995520.0,
      "grad_norm": 1.5648783740553736,
      "language_loss": 0.71559078,
      "learning_rate": 2.808122429572528e-07,
      "loss": 0.73736107,
      "num_input_tokens_seen": 149942435,
      "step": 6939,
      "time_per_iteration": 2.477778911590576
    },
    {
      "auxiliary_loss_clip": 0.01130513,
      "auxiliary_loss_mlp": 0.01025887,
      "balance_loss_clip": 1.04283655,
      "balance_loss_mlp": 1.0185318,
      "epoch": 0.8344856610352913,
      "flos": 20777268078720.0,
      "grad_norm": 2.8486585117066476,
      "language_loss": 0.76067793,
      "learning_rate": 2.804143361088489e-07,
      "loss": 0.78224194,
      "num_input_tokens_seen": 149961615,
      "step": 6940,
      "time_per_iteration": 2.499452829360962
    },
    {
      "auxiliary_loss_clip": 0.0113024,
      "auxiliary_loss_mlp": 0.01029388,
      "balance_loss_clip": 1.04214132,
      "balance_loss_mlp": 1.0218333,
      "epoch": 0.8346059039259304,
      "flos": 26095960684800.0,
      "grad_norm": 2.387089435526413,
      "language_loss": 0.78118169,
      "learning_rate": 2.8001669012022277e-07,
      "loss": 0.80277801,
      "num_input_tokens_seen": 149979585,
      "step": 6941,
      "time_per_iteration": 2.493180274963379
    },
    {
      "auxiliary_loss_clip": 0.01151673,
      "auxiliary_loss_mlp": 0.01026672,
      "balance_loss_clip": 1.04898906,
      "balance_loss_mlp": 1.01939154,
      "epoch": 0.8347261468165694,
      "flos": 29027708755200.0,
      "grad_norm": 1.7584457250076992,
      "language_loss": 0.69360065,
      "learning_rate": 2.7961930505169795e-07,
      "loss": 0.71538413,
      "num_input_tokens_seen": 150003830,
      "step": 6942,
      "time_per_iteration": 2.5236563682556152
    },
    {
      "auxiliary_loss_clip": 0.01153725,
      "auxiliary_loss_mlp": 0.00761738,
      "balance_loss_clip": 1.04787278,
      "balance_loss_mlp": 1.00048041,
      "epoch": 0.8348463897072086,
      "flos": 26396461866240.0,
      "grad_norm": 2.015232040819183,
      "language_loss": 0.76177752,
      "learning_rate": 2.792221809635558e-07,
      "loss": 0.78093219,
      "num_input_tokens_seen": 150024460,
      "step": 6943,
      "time_per_iteration": 2.494431257247925
    },
    {
      "auxiliary_loss_clip": 0.01086584,
      "auxiliary_loss_mlp": 0.01023602,
      "balance_loss_clip": 1.04176188,
      "balance_loss_mlp": 1.01638675,
      "epoch": 0.8349666325978476,
      "flos": 23367720096000.0,
      "grad_norm": 2.1210041043345345,
      "language_loss": 0.7506578,
      "learning_rate": 2.788253179160411e-07,
      "loss": 0.77175963,
      "num_input_tokens_seen": 150045620,
      "step": 6944,
      "time_per_iteration": 2.7207767963409424
    },
    {
      "auxiliary_loss_clip": 0.01136251,
      "auxiliary_loss_mlp": 0.0102763,
      "balance_loss_clip": 1.04426312,
      "balance_loss_mlp": 1.02097559,
      "epoch": 0.8350868754884867,
      "flos": 12896528135040.0,
      "grad_norm": 1.8673273006479607,
      "language_loss": 0.64921761,
      "learning_rate": 2.7842871596935725e-07,
      "loss": 0.67085636,
      "num_input_tokens_seen": 150064135,
      "step": 6945,
      "time_per_iteration": 2.9518702030181885
    },
    {
      "auxiliary_loss_clip": 0.01154998,
      "auxiliary_loss_mlp": 0.0101947,
      "balance_loss_clip": 1.04560351,
      "balance_loss_mlp": 1.01249576,
      "epoch": 0.8352071183791259,
      "flos": 26505522535680.0,
      "grad_norm": 1.6611386414730471,
      "language_loss": 0.68976617,
      "learning_rate": 2.780323751836682e-07,
      "loss": 0.71151078,
      "num_input_tokens_seen": 150085350,
      "step": 6946,
      "time_per_iteration": 2.495692729949951
    },
    {
      "auxiliary_loss_clip": 0.01134376,
      "auxiliary_loss_mlp": 0.00761209,
      "balance_loss_clip": 1.04079604,
      "balance_loss_mlp": 1.00044143,
      "epoch": 0.8353273612697649,
      "flos": 20668063754880.0,
      "grad_norm": 1.4186230174993504,
      "language_loss": 0.78450453,
      "learning_rate": 2.7763629561909876e-07,
      "loss": 0.80346036,
      "num_input_tokens_seen": 150106180,
      "step": 6947,
      "time_per_iteration": 2.4927756786346436
    },
    {
      "auxiliary_loss_clip": 0.01161122,
      "auxiliary_loss_mlp": 0.01020948,
      "balance_loss_clip": 1.04444301,
      "balance_loss_mlp": 1.01373017,
      "epoch": 0.835447604160404,
      "flos": 19754137082880.0,
      "grad_norm": 1.901840883099076,
      "language_loss": 0.76970482,
      "learning_rate": 2.772404773357335e-07,
      "loss": 0.79152548,
      "num_input_tokens_seen": 150125585,
      "step": 6948,
      "time_per_iteration": 3.2674615383148193
    },
    {
      "auxiliary_loss_clip": 0.01117828,
      "auxiliary_loss_mlp": 0.01024899,
      "balance_loss_clip": 1.04270077,
      "balance_loss_mlp": 1.01764548,
      "epoch": 0.8355678470510431,
      "flos": 23435842239360.0,
      "grad_norm": 1.755933792338582,
      "language_loss": 0.78216481,
      "learning_rate": 2.7684492039361853e-07,
      "loss": 0.80359209,
      "num_input_tokens_seen": 150144810,
      "step": 6949,
      "time_per_iteration": 3.301015853881836
    },
    {
      "auxiliary_loss_clip": 0.01166449,
      "auxiliary_loss_mlp": 0.01024814,
      "balance_loss_clip": 1.04882836,
      "balance_loss_mlp": 1.01816261,
      "epoch": 0.8356880899416822,
      "flos": 21214588164480.0,
      "grad_norm": 2.087516574864285,
      "language_loss": 0.83766705,
      "learning_rate": 2.764496248527586e-07,
      "loss": 0.85957968,
      "num_input_tokens_seen": 150163785,
      "step": 6950,
      "time_per_iteration": 3.195256471633911
    },
    {
      "auxiliary_loss_clip": 0.0113307,
      "auxiliary_loss_mlp": 0.01024228,
      "balance_loss_clip": 1.0437808,
      "balance_loss_mlp": 1.0173614,
      "epoch": 0.8358083328323213,
      "flos": 28037543466240.0,
      "grad_norm": 2.3510273278985987,
      "language_loss": 0.78784251,
      "learning_rate": 2.760545907731211e-07,
      "loss": 0.80941546,
      "num_input_tokens_seen": 150184360,
      "step": 6951,
      "time_per_iteration": 2.5752291679382324
    },
    {
      "auxiliary_loss_clip": 0.01151516,
      "auxiliary_loss_mlp": 0.0102019,
      "balance_loss_clip": 1.04373634,
      "balance_loss_mlp": 1.01315629,
      "epoch": 0.8359285757229604,
      "flos": 27783655159680.0,
      "grad_norm": 1.70605508551502,
      "language_loss": 0.6787008,
      "learning_rate": 2.75659818214631e-07,
      "loss": 0.70041788,
      "num_input_tokens_seen": 150205465,
      "step": 6952,
      "time_per_iteration": 2.4936938285827637
    },
    {
      "auxiliary_loss_clip": 0.01138847,
      "auxiliary_loss_mlp": 0.01021823,
      "balance_loss_clip": 1.04327345,
      "balance_loss_mlp": 1.0147959,
      "epoch": 0.8360488186135995,
      "flos": 21435115714560.0,
      "grad_norm": 2.805824680037939,
      "language_loss": 0.78170407,
      "learning_rate": 2.752653072371749e-07,
      "loss": 0.80331075,
      "num_input_tokens_seen": 150224900,
      "step": 6953,
      "time_per_iteration": 2.4774389266967773
    },
    {
      "auxiliary_loss_clip": 0.0111829,
      "auxiliary_loss_mlp": 0.01025847,
      "balance_loss_clip": 1.04366326,
      "balance_loss_mlp": 1.01947212,
      "epoch": 0.8361690615042385,
      "flos": 27632327160960.0,
      "grad_norm": 1.7937690686683287,
      "language_loss": 0.7484467,
      "learning_rate": 2.7487105790060105e-07,
      "loss": 0.7698881,
      "num_input_tokens_seen": 150244310,
      "step": 6954,
      "time_per_iteration": 2.5428404808044434
    },
    {
      "auxiliary_loss_clip": 0.01152706,
      "auxiliary_loss_mlp": 0.01024512,
      "balance_loss_clip": 1.04540277,
      "balance_loss_mlp": 1.01797664,
      "epoch": 0.8362893043948777,
      "flos": 39202529598720.0,
      "grad_norm": 1.9532298233738328,
      "language_loss": 0.69287205,
      "learning_rate": 2.7447707026471587e-07,
      "loss": 0.71464431,
      "num_input_tokens_seen": 150267285,
      "step": 6955,
      "time_per_iteration": 3.3278300762176514
    },
    {
      "auxiliary_loss_clip": 0.01122584,
      "auxiliary_loss_mlp": 0.0102251,
      "balance_loss_clip": 1.04156768,
      "balance_loss_mlp": 1.01582241,
      "epoch": 0.8364095472855168,
      "flos": 24785329230720.0,
      "grad_norm": 1.9375372219850042,
      "language_loss": 0.79851127,
      "learning_rate": 2.740833443892874e-07,
      "loss": 0.81996214,
      "num_input_tokens_seen": 150285455,
      "step": 6956,
      "time_per_iteration": 2.5371198654174805
    },
    {
      "auxiliary_loss_clip": 0.01135907,
      "auxiliary_loss_mlp": 0.01021642,
      "balance_loss_clip": 1.04385018,
      "balance_loss_mlp": 1.01459134,
      "epoch": 0.8365297901761558,
      "flos": 22743412784640.0,
      "grad_norm": 1.9392602941511379,
      "language_loss": 0.79682028,
      "learning_rate": 2.7368988033404327e-07,
      "loss": 0.81839573,
      "num_input_tokens_seen": 150302970,
      "step": 6957,
      "time_per_iteration": 2.468261957168579
    },
    {
      "auxiliary_loss_clip": 0.01125615,
      "auxiliary_loss_mlp": 0.01024318,
      "balance_loss_clip": 1.04395747,
      "balance_loss_mlp": 1.01799953,
      "epoch": 0.836650033066795,
      "flos": 28396003242240.0,
      "grad_norm": 1.5531683819212478,
      "language_loss": 0.84720874,
      "learning_rate": 2.732966781586712e-07,
      "loss": 0.86870801,
      "num_input_tokens_seen": 150322715,
      "step": 6958,
      "time_per_iteration": 2.5641021728515625
    },
    {
      "auxiliary_loss_clip": 0.01144791,
      "auxiliary_loss_mlp": 0.0102156,
      "balance_loss_clip": 1.04266,
      "balance_loss_mlp": 1.01514602,
      "epoch": 0.836770275957434,
      "flos": 22236857233920.0,
      "grad_norm": 1.7216225154039906,
      "language_loss": 0.66908062,
      "learning_rate": 2.729037379228205e-07,
      "loss": 0.69074416,
      "num_input_tokens_seen": 150342900,
      "step": 6959,
      "time_per_iteration": 2.4576852321624756
    },
    {
      "auxiliary_loss_clip": 0.01137675,
      "auxiliary_loss_mlp": 0.01026606,
      "balance_loss_clip": 1.04780185,
      "balance_loss_mlp": 1.01967692,
      "epoch": 0.8368905188480731,
      "flos": 22491930689280.0,
      "grad_norm": 1.4476338864757565,
      "language_loss": 0.8069762,
      "learning_rate": 2.725110596860998e-07,
      "loss": 0.828619,
      "num_input_tokens_seen": 150363580,
      "step": 6960,
      "time_per_iteration": 2.4842004776000977
    },
    {
      "auxiliary_loss_clip": 0.01104386,
      "auxiliary_loss_mlp": 0.01022301,
      "balance_loss_clip": 1.04132843,
      "balance_loss_mlp": 1.01566148,
      "epoch": 0.8370107617387123,
      "flos": 13370405287680.0,
      "grad_norm": 2.1321467620516,
      "language_loss": 0.70102775,
      "learning_rate": 2.7211864350807776e-07,
      "loss": 0.72229463,
      "num_input_tokens_seen": 150381780,
      "step": 6961,
      "time_per_iteration": 2.521544933319092
    },
    {
      "auxiliary_loss_clip": 0.01163386,
      "auxiliary_loss_mlp": 0.01026392,
      "balance_loss_clip": 1.04560983,
      "balance_loss_mlp": 1.01908493,
      "epoch": 0.8371310046293513,
      "flos": 25261289372160.0,
      "grad_norm": 1.703790546895152,
      "language_loss": 0.74247926,
      "learning_rate": 2.717264894482836e-07,
      "loss": 0.76437712,
      "num_input_tokens_seen": 150402120,
      "step": 6962,
      "time_per_iteration": 2.464632749557495
    },
    {
      "auxiliary_loss_clip": 0.01154373,
      "auxiliary_loss_mlp": 0.01023766,
      "balance_loss_clip": 1.04802001,
      "balance_loss_mlp": 1.01622856,
      "epoch": 0.8372512475199904,
      "flos": 19792705311360.0,
      "grad_norm": 2.007787186534202,
      "language_loss": 0.81070566,
      "learning_rate": 2.7133459756620646e-07,
      "loss": 0.83248711,
      "num_input_tokens_seen": 150419315,
      "step": 6963,
      "time_per_iteration": 2.468907594680786
    },
    {
      "auxiliary_loss_clip": 0.01145278,
      "auxiliary_loss_mlp": 0.01030272,
      "balance_loss_clip": 1.0443536,
      "balance_loss_mlp": 1.02338147,
      "epoch": 0.8373714904106295,
      "flos": 19391224020480.0,
      "grad_norm": 1.7195082167261235,
      "language_loss": 0.73618448,
      "learning_rate": 2.7094296792129733e-07,
      "loss": 0.75793993,
      "num_input_tokens_seen": 150438915,
      "step": 6964,
      "time_per_iteration": 2.430236577987671
    },
    {
      "auxiliary_loss_clip": 0.01150345,
      "auxiliary_loss_mlp": 0.01022109,
      "balance_loss_clip": 1.04481459,
      "balance_loss_mlp": 1.01544523,
      "epoch": 0.8374917333012686,
      "flos": 14975935401600.0,
      "grad_norm": 1.7689790264549623,
      "language_loss": 0.75638026,
      "learning_rate": 2.7055160057296424e-07,
      "loss": 0.77810478,
      "num_input_tokens_seen": 150456155,
      "step": 6965,
      "time_per_iteration": 2.5920419692993164
    },
    {
      "auxiliary_loss_clip": 0.01121272,
      "auxiliary_loss_mlp": 0.01023584,
      "balance_loss_clip": 1.04243863,
      "balance_loss_mlp": 1.01647019,
      "epoch": 0.8376119761919076,
      "flos": 30331839847680.0,
      "grad_norm": 1.5715509710924365,
      "language_loss": 0.72631294,
      "learning_rate": 2.7016049558057896e-07,
      "loss": 0.74776149,
      "num_input_tokens_seen": 150478115,
      "step": 6966,
      "time_per_iteration": 2.5752642154693604
    },
    {
      "auxiliary_loss_clip": 0.01149372,
      "auxiliary_loss_mlp": 0.0102895,
      "balance_loss_clip": 1.04579222,
      "balance_loss_mlp": 1.02147007,
      "epoch": 0.8377322190825467,
      "flos": 29423336129280.0,
      "grad_norm": 1.6801702603297883,
      "language_loss": 0.70689237,
      "learning_rate": 2.6976965300347074e-07,
      "loss": 0.72867548,
      "num_input_tokens_seen": 150500725,
      "step": 6967,
      "time_per_iteration": 2.5245361328125
    },
    {
      "auxiliary_loss_clip": 0.01132521,
      "auxiliary_loss_mlp": 0.01019616,
      "balance_loss_clip": 1.04240608,
      "balance_loss_mlp": 1.01227856,
      "epoch": 0.8378524619731859,
      "flos": 26687086807680.0,
      "grad_norm": 2.9911795981002314,
      "language_loss": 0.69328952,
      "learning_rate": 2.693790729009309e-07,
      "loss": 0.71481085,
      "num_input_tokens_seen": 150522335,
      "step": 6968,
      "time_per_iteration": 2.5130228996276855
    },
    {
      "auxiliary_loss_clip": 0.01135986,
      "auxiliary_loss_mlp": 0.01022417,
      "balance_loss_clip": 1.04358733,
      "balance_loss_mlp": 1.01564646,
      "epoch": 0.8379727048638249,
      "flos": 20703866636160.0,
      "grad_norm": 1.808333112762289,
      "language_loss": 0.8842935,
      "learning_rate": 2.6898875533220946e-07,
      "loss": 0.90587747,
      "num_input_tokens_seen": 150541640,
      "step": 6969,
      "time_per_iteration": 2.480952739715576
    },
    {
      "auxiliary_loss_clip": 0.01160835,
      "auxiliary_loss_mlp": 0.01021923,
      "balance_loss_clip": 1.04758787,
      "balance_loss_mlp": 1.01572394,
      "epoch": 0.838092947754464,
      "flos": 20084084438400.0,
      "grad_norm": 1.9256385222617913,
      "language_loss": 0.81550467,
      "learning_rate": 2.685987003565171e-07,
      "loss": 0.83733219,
      "num_input_tokens_seen": 150559680,
      "step": 6970,
      "time_per_iteration": 2.430691957473755
    },
    {
      "auxiliary_loss_clip": 0.0112071,
      "auxiliary_loss_mlp": 0.01023381,
      "balance_loss_clip": 1.04733825,
      "balance_loss_mlp": 1.01575792,
      "epoch": 0.8382131906451031,
      "flos": 18113270964480.0,
      "grad_norm": 3.0809953122499074,
      "language_loss": 0.74431747,
      "learning_rate": 2.6820890803302566e-07,
      "loss": 0.7657584,
      "num_input_tokens_seen": 150575205,
      "step": 6971,
      "time_per_iteration": 2.490973711013794
    },
    {
      "auxiliary_loss_clip": 0.01137266,
      "auxiliary_loss_mlp": 0.01021454,
      "balance_loss_clip": 1.04685569,
      "balance_loss_mlp": 1.01451862,
      "epoch": 0.8383334335357422,
      "flos": 17092653920640.0,
      "grad_norm": 2.0855543562827252,
      "language_loss": 0.81963855,
      "learning_rate": 2.6781937842086557e-07,
      "loss": 0.84122574,
      "num_input_tokens_seen": 150593995,
      "step": 6972,
      "time_per_iteration": 2.4419753551483154
    },
    {
      "auxiliary_loss_clip": 0.01152122,
      "auxiliary_loss_mlp": 0.0102735,
      "balance_loss_clip": 1.0447402,
      "balance_loss_mlp": 1.02017021,
      "epoch": 0.8384536764263812,
      "flos": 20704728562560.0,
      "grad_norm": 2.1998285428165913,
      "language_loss": 0.67823023,
      "learning_rate": 2.6743011157912933e-07,
      "loss": 0.70002496,
      "num_input_tokens_seen": 150613715,
      "step": 6973,
      "time_per_iteration": 2.4497125148773193
    },
    {
      "auxiliary_loss_clip": 0.01106161,
      "auxiliary_loss_mlp": 0.01023144,
      "balance_loss_clip": 1.03709149,
      "balance_loss_mlp": 1.01575351,
      "epoch": 0.8385739193170204,
      "flos": 28986842056320.0,
      "grad_norm": 1.7955662608864091,
      "language_loss": 0.65369266,
      "learning_rate": 2.6704110756686725e-07,
      "loss": 0.67498571,
      "num_input_tokens_seen": 150634540,
      "step": 6974,
      "time_per_iteration": 2.6011180877685547
    },
    {
      "auxiliary_loss_clip": 0.01133066,
      "auxiliary_loss_mlp": 0.00761537,
      "balance_loss_clip": 1.04210615,
      "balance_loss_mlp": 1.00039816,
      "epoch": 0.8386941622076595,
      "flos": 23438068882560.0,
      "grad_norm": 1.6424947887458081,
      "language_loss": 0.83731735,
      "learning_rate": 2.6665236644309085e-07,
      "loss": 0.8562634,
      "num_input_tokens_seen": 150654850,
      "step": 6975,
      "time_per_iteration": 3.338249444961548
    },
    {
      "auxiliary_loss_clip": 0.01149939,
      "auxiliary_loss_mlp": 0.01022066,
      "balance_loss_clip": 1.04336858,
      "balance_loss_mlp": 1.01553047,
      "epoch": 0.8388144050982985,
      "flos": 23002724044800.0,
      "grad_norm": 1.8162602908230185,
      "language_loss": 0.79731917,
      "learning_rate": 2.662638882667727e-07,
      "loss": 0.81903923,
      "num_input_tokens_seen": 150673790,
      "step": 6976,
      "time_per_iteration": 4.033307790756226
    },
    {
      "auxiliary_loss_clip": 0.01167135,
      "auxiliary_loss_mlp": 0.01025097,
      "balance_loss_clip": 1.0469172,
      "balance_loss_mlp": 1.0177809,
      "epoch": 0.8389346479889377,
      "flos": 24280353878400.0,
      "grad_norm": 2.0341288116991194,
      "language_loss": 0.72796947,
      "learning_rate": 2.658756730968443e-07,
      "loss": 0.74989176,
      "num_input_tokens_seen": 150692255,
      "step": 6977,
      "time_per_iteration": 2.4284560680389404
    },
    {
      "auxiliary_loss_clip": 0.01142641,
      "auxiliary_loss_mlp": 0.01021912,
      "balance_loss_clip": 1.048944,
      "balance_loss_mlp": 1.01503658,
      "epoch": 0.8390548908795767,
      "flos": 21215019127680.0,
      "grad_norm": 2.2504662227208594,
      "language_loss": 0.88496709,
      "learning_rate": 2.654877209921975e-07,
      "loss": 0.90661263,
      "num_input_tokens_seen": 150709790,
      "step": 6978,
      "time_per_iteration": 2.4695749282836914
    },
    {
      "auxiliary_loss_clip": 0.01114414,
      "auxiliary_loss_mlp": 0.01027919,
      "balance_loss_clip": 1.04012942,
      "balance_loss_mlp": 1.01961565,
      "epoch": 0.8391751337702158,
      "flos": 35627299332480.0,
      "grad_norm": 4.455165390242117,
      "language_loss": 0.63079774,
      "learning_rate": 2.651000320116843e-07,
      "loss": 0.65222108,
      "num_input_tokens_seen": 150730675,
      "step": 6979,
      "time_per_iteration": 2.661766290664673
    },
    {
      "auxiliary_loss_clip": 0.01118862,
      "auxiliary_loss_mlp": 0.00762102,
      "balance_loss_clip": 1.04148579,
      "balance_loss_mlp": 1.00054991,
      "epoch": 0.839295376660855,
      "flos": 21325229032320.0,
      "grad_norm": 2.0538298703010174,
      "language_loss": 0.7569896,
      "learning_rate": 2.647126062141163e-07,
      "loss": 0.77579921,
      "num_input_tokens_seen": 150749750,
      "step": 6980,
      "time_per_iteration": 2.5189621448516846
    },
    {
      "auxiliary_loss_clip": 0.01139241,
      "auxiliary_loss_mlp": 0.0102409,
      "balance_loss_clip": 1.04163551,
      "balance_loss_mlp": 1.01731288,
      "epoch": 0.839415619551494,
      "flos": 18442535961600.0,
      "grad_norm": 1.668233801528303,
      "language_loss": 0.83698547,
      "learning_rate": 2.643254436582669e-07,
      "loss": 0.85861874,
      "num_input_tokens_seen": 150769240,
      "step": 6981,
      "time_per_iteration": 2.4516139030456543
    },
    {
      "auxiliary_loss_clip": 0.01109935,
      "auxiliary_loss_mlp": 0.01022436,
      "balance_loss_clip": 1.04181814,
      "balance_loss_mlp": 1.01553988,
      "epoch": 0.8395358624421331,
      "flos": 23221958705280.0,
      "grad_norm": 1.783339782362761,
      "language_loss": 0.82419789,
      "learning_rate": 2.6393854440286743e-07,
      "loss": 0.84552157,
      "num_input_tokens_seen": 150788410,
      "step": 6982,
      "time_per_iteration": 3.3319625854492188
    },
    {
      "auxiliary_loss_clip": 0.01165053,
      "auxiliary_loss_mlp": 0.01023904,
      "balance_loss_clip": 1.04849601,
      "balance_loss_mlp": 1.0171212,
      "epoch": 0.8396561053327722,
      "flos": 24381657210240.0,
      "grad_norm": 2.569599960357759,
      "language_loss": 0.70636833,
      "learning_rate": 2.6355190850661045e-07,
      "loss": 0.72825789,
      "num_input_tokens_seen": 150805245,
      "step": 6983,
      "time_per_iteration": 2.415895700454712
    },
    {
      "auxiliary_loss_clip": 0.01138006,
      "auxiliary_loss_mlp": 0.0102481,
      "balance_loss_clip": 1.04539049,
      "balance_loss_mlp": 1.01755309,
      "epoch": 0.8397763482234113,
      "flos": 22237755073920.0,
      "grad_norm": 1.5791957367098992,
      "language_loss": 0.86302984,
      "learning_rate": 2.631655360281486e-07,
      "loss": 0.88465798,
      "num_input_tokens_seen": 150824920,
      "step": 6984,
      "time_per_iteration": 2.4833171367645264
    },
    {
      "auxiliary_loss_clip": 0.01154089,
      "auxiliary_loss_mlp": 0.00761754,
      "balance_loss_clip": 1.04390144,
      "balance_loss_mlp": 1.0005095,
      "epoch": 0.8398965911140504,
      "flos": 22163743100160.0,
      "grad_norm": 1.7952765267680695,
      "language_loss": 0.65966392,
      "learning_rate": 2.6277942702609323e-07,
      "loss": 0.67882234,
      "num_input_tokens_seen": 150844400,
      "step": 6985,
      "time_per_iteration": 2.4489188194274902
    },
    {
      "auxiliary_loss_clip": 0.01124673,
      "auxiliary_loss_mlp": 0.01026844,
      "balance_loss_clip": 1.04439592,
      "balance_loss_mlp": 1.01978111,
      "epoch": 0.8400168340046895,
      "flos": 21542775753600.0,
      "grad_norm": 1.962985614768301,
      "language_loss": 0.8769843,
      "learning_rate": 2.623935815590186e-07,
      "loss": 0.89849943,
      "num_input_tokens_seen": 150862780,
      "step": 6986,
      "time_per_iteration": 2.5191588401794434
    },
    {
      "auxiliary_loss_clip": 0.01139752,
      "auxiliary_loss_mlp": 0.01022126,
      "balance_loss_clip": 1.04759479,
      "balance_loss_mlp": 1.01508653,
      "epoch": 0.8401370768953286,
      "flos": 22491966602880.0,
      "grad_norm": 1.793309077995339,
      "language_loss": 0.80801427,
      "learning_rate": 2.6200799968545516e-07,
      "loss": 0.829633,
      "num_input_tokens_seen": 150883075,
      "step": 6987,
      "time_per_iteration": 2.4853620529174805
    },
    {
      "auxiliary_loss_clip": 0.01038587,
      "auxiliary_loss_mlp": 0.0100071,
      "balance_loss_clip": 1.01141524,
      "balance_loss_mlp": 0.99992907,
      "epoch": 0.8402573197859676,
      "flos": 59238890818560.0,
      "grad_norm": 0.7862854362295917,
      "language_loss": 0.56427372,
      "learning_rate": 2.616226814638969e-07,
      "loss": 0.58466667,
      "num_input_tokens_seen": 150948180,
      "step": 6988,
      "time_per_iteration": 3.092822551727295
    },
    {
      "auxiliary_loss_clip": 0.01138336,
      "auxiliary_loss_mlp": 0.0102241,
      "balance_loss_clip": 1.04511523,
      "balance_loss_mlp": 1.0157733,
      "epoch": 0.8403775626766068,
      "flos": 22674608282880.0,
      "grad_norm": 1.905849159551543,
      "language_loss": 0.77137804,
      "learning_rate": 2.612376269527954e-07,
      "loss": 0.79298556,
      "num_input_tokens_seen": 150967885,
      "step": 6989,
      "time_per_iteration": 2.5085322856903076
    },
    {
      "auxiliary_loss_clip": 0.01133763,
      "auxiliary_loss_mlp": 0.01025864,
      "balance_loss_clip": 1.04457855,
      "balance_loss_mlp": 1.01916468,
      "epoch": 0.8404978055672458,
      "flos": 19609704495360.0,
      "grad_norm": 1.669140990877128,
      "language_loss": 0.67639565,
      "learning_rate": 2.608528362105635e-07,
      "loss": 0.69799197,
      "num_input_tokens_seen": 150987255,
      "step": 6990,
      "time_per_iteration": 2.461534023284912
    },
    {
      "auxiliary_loss_clip": 0.01122708,
      "auxiliary_loss_mlp": 0.01022297,
      "balance_loss_clip": 1.03927422,
      "balance_loss_mlp": 1.01555538,
      "epoch": 0.8406180484578849,
      "flos": 27526929678720.0,
      "grad_norm": 1.9221059056010676,
      "language_loss": 0.7348206,
      "learning_rate": 2.6046830929557374e-07,
      "loss": 0.75627065,
      "num_input_tokens_seen": 151006905,
      "step": 6991,
      "time_per_iteration": 2.5702147483825684
    },
    {
      "auxiliary_loss_clip": 0.01119833,
      "auxiliary_loss_mlp": 0.01023608,
      "balance_loss_clip": 1.04361022,
      "balance_loss_mlp": 1.01688504,
      "epoch": 0.8407382913485241,
      "flos": 22127473342080.0,
      "grad_norm": 1.9428695315506979,
      "language_loss": 0.84759438,
      "learning_rate": 2.6008404626615776e-07,
      "loss": 0.86902881,
      "num_input_tokens_seen": 151025405,
      "step": 6992,
      "time_per_iteration": 2.5197558403015137
    },
    {
      "auxiliary_loss_clip": 0.01157517,
      "auxiliary_loss_mlp": 0.01023965,
      "balance_loss_clip": 1.04906511,
      "balance_loss_mlp": 1.01738441,
      "epoch": 0.8408585342391631,
      "flos": 13918473982080.0,
      "grad_norm": 3.084258260778641,
      "language_loss": 0.74324363,
      "learning_rate": 2.597000471806092e-07,
      "loss": 0.7650584,
      "num_input_tokens_seen": 151041970,
      "step": 6993,
      "time_per_iteration": 2.442943572998047
    },
    {
      "auxiliary_loss_clip": 0.01135269,
      "auxiliary_loss_mlp": 0.01023159,
      "balance_loss_clip": 1.04818904,
      "balance_loss_mlp": 1.01504421,
      "epoch": 0.8409787771298022,
      "flos": 20187865808640.0,
      "grad_norm": 1.9098399148648195,
      "language_loss": 0.73390204,
      "learning_rate": 2.593163120971793e-07,
      "loss": 0.75548637,
      "num_input_tokens_seen": 151060835,
      "step": 6994,
      "time_per_iteration": 2.439973831176758
    },
    {
      "auxiliary_loss_clip": 0.01099618,
      "auxiliary_loss_mlp": 0.010228,
      "balance_loss_clip": 1.03676105,
      "balance_loss_mlp": 1.01613069,
      "epoch": 0.8410990200204413,
      "flos": 23142523777920.0,
      "grad_norm": 1.7711748046407754,
      "language_loss": 0.68708283,
      "learning_rate": 2.5893284107408165e-07,
      "loss": 0.70830703,
      "num_input_tokens_seen": 151078205,
      "step": 6995,
      "time_per_iteration": 2.5352261066436768
    },
    {
      "auxiliary_loss_clip": 0.01111214,
      "auxiliary_loss_mlp": 0.01027742,
      "balance_loss_clip": 1.0441314,
      "balance_loss_mlp": 1.02020264,
      "epoch": 0.8412192629110804,
      "flos": 24027219757440.0,
      "grad_norm": 1.7909777825080915,
      "language_loss": 0.77824235,
      "learning_rate": 2.5854963416948726e-07,
      "loss": 0.79963195,
      "num_input_tokens_seen": 151100470,
      "step": 6996,
      "time_per_iteration": 2.6048130989074707
    },
    {
      "auxiliary_loss_clip": 0.01107499,
      "auxiliary_loss_mlp": 0.01024388,
      "balance_loss_clip": 1.03611398,
      "balance_loss_mlp": 1.01739049,
      "epoch": 0.8413395058017195,
      "flos": 25591703604480.0,
      "grad_norm": 2.029241840102886,
      "language_loss": 0.69332278,
      "learning_rate": 2.5816669144152816e-07,
      "loss": 0.71464163,
      "num_input_tokens_seen": 151121650,
      "step": 6997,
      "time_per_iteration": 2.5674712657928467
    },
    {
      "auxiliary_loss_clip": 0.01061934,
      "auxiliary_loss_mlp": 0.01000901,
      "balance_loss_clip": 1.00829625,
      "balance_loss_mlp": 1.00005424,
      "epoch": 0.8414597486923585,
      "flos": 63635396624640.0,
      "grad_norm": 0.8665706395173894,
      "language_loss": 0.66371548,
      "learning_rate": 2.5778401294829777e-07,
      "loss": 0.68434381,
      "num_input_tokens_seen": 151180390,
      "step": 6998,
      "time_per_iteration": 3.0523874759674072
    },
    {
      "auxiliary_loss_clip": 0.0114817,
      "auxiliary_loss_mlp": 0.00761816,
      "balance_loss_clip": 1.04551339,
      "balance_loss_mlp": 1.00046432,
      "epoch": 0.8415799915829977,
      "flos": 19098731571840.0,
      "grad_norm": 1.6629609593982126,
      "language_loss": 0.65053612,
      "learning_rate": 2.574015987478473e-07,
      "loss": 0.66963595,
      "num_input_tokens_seen": 151198520,
      "step": 6999,
      "time_per_iteration": 2.4193525314331055
    },
    {
      "auxiliary_loss_clip": 0.01142237,
      "auxiliary_loss_mlp": 0.01024528,
      "balance_loss_clip": 1.04505825,
      "balance_loss_mlp": 1.01735151,
      "epoch": 0.8417002344736367,
      "flos": 19821612781440.0,
      "grad_norm": 2.4083744176707147,
      "language_loss": 0.86991733,
      "learning_rate": 2.570194488981887e-07,
      "loss": 0.89158493,
      "num_input_tokens_seen": 151215065,
      "step": 7000,
      "time_per_iteration": 2.569504499435425
    },
    {
      "auxiliary_loss_clip": 0.01062398,
      "auxiliary_loss_mlp": 0.0100188,
      "balance_loss_clip": 1.00882101,
      "balance_loss_mlp": 1.00096226,
      "epoch": 0.8418204773642758,
      "flos": 62161516834560.0,
      "grad_norm": 0.8395303638495538,
      "language_loss": 0.60327309,
      "learning_rate": 2.566375634572939e-07,
      "loss": 0.62391591,
      "num_input_tokens_seen": 151275705,
      "step": 7001,
      "time_per_iteration": 3.7731032371520996
    },
    {
      "auxiliary_loss_clip": 0.01127754,
      "auxiliary_loss_mlp": 0.01026222,
      "balance_loss_clip": 1.04184294,
      "balance_loss_mlp": 1.0197494,
      "epoch": 0.841940720254915,
      "flos": 17092905315840.0,
      "grad_norm": 2.1404324055956803,
      "language_loss": 0.76612818,
      "learning_rate": 2.562559424830943e-07,
      "loss": 0.78766793,
      "num_input_tokens_seen": 151293665,
      "step": 7002,
      "time_per_iteration": 3.2008044719696045
    },
    {
      "auxiliary_loss_clip": 0.01135183,
      "auxiliary_loss_mlp": 0.01021252,
      "balance_loss_clip": 1.04324269,
      "balance_loss_mlp": 1.01389098,
      "epoch": 0.842060963145554,
      "flos": 16283586026880.0,
      "grad_norm": 2.29952970873097,
      "language_loss": 0.70264083,
      "learning_rate": 2.5587458603348256e-07,
      "loss": 0.7242052,
      "num_input_tokens_seen": 151310955,
      "step": 7003,
      "time_per_iteration": 3.311286211013794
    },
    {
      "auxiliary_loss_clip": 0.01117077,
      "auxiliary_loss_mlp": 0.0102657,
      "balance_loss_clip": 1.04225576,
      "balance_loss_mlp": 1.0195303,
      "epoch": 0.8421812060361931,
      "flos": 21908238681600.0,
      "grad_norm": 1.9825496680862764,
      "language_loss": 0.84104604,
      "learning_rate": 2.554934941663085e-07,
      "loss": 0.86248249,
      "num_input_tokens_seen": 151328490,
      "step": 7004,
      "time_per_iteration": 2.5046706199645996
    },
    {
      "auxiliary_loss_clip": 0.01121421,
      "auxiliary_loss_mlp": 0.01025504,
      "balance_loss_clip": 1.04192698,
      "balance_loss_mlp": 1.01776981,
      "epoch": 0.8423014489268322,
      "flos": 27777693502080.0,
      "grad_norm": 2.7630898728600113,
      "language_loss": 0.73586118,
      "learning_rate": 2.5511266693938484e-07,
      "loss": 0.75733042,
      "num_input_tokens_seen": 151346950,
      "step": 7005,
      "time_per_iteration": 2.54984974861145
    },
    {
      "auxiliary_loss_clip": 0.01135532,
      "auxiliary_loss_mlp": 0.01022897,
      "balance_loss_clip": 1.04614401,
      "balance_loss_mlp": 1.01546133,
      "epoch": 0.8424216918174713,
      "flos": 25117610970240.0,
      "grad_norm": 1.6079876458492246,
      "language_loss": 0.77764171,
      "learning_rate": 2.547321044104822e-07,
      "loss": 0.79922599,
      "num_input_tokens_seen": 151368445,
      "step": 7006,
      "time_per_iteration": 2.513831615447998
    },
    {
      "auxiliary_loss_clip": 0.01167217,
      "auxiliary_loss_mlp": 0.01023001,
      "balance_loss_clip": 1.04758072,
      "balance_loss_mlp": 1.01549375,
      "epoch": 0.8425419347081103,
      "flos": 24748448941440.0,
      "grad_norm": 1.634285320447925,
      "language_loss": 0.76497817,
      "learning_rate": 2.5435180663733113e-07,
      "loss": 0.78688037,
      "num_input_tokens_seen": 151388745,
      "step": 7007,
      "time_per_iteration": 2.4596002101898193
    },
    {
      "auxiliary_loss_clip": 0.01117529,
      "auxiliary_loss_mlp": 0.01026392,
      "balance_loss_clip": 1.04068685,
      "balance_loss_mlp": 1.01949883,
      "epoch": 0.8426621775987495,
      "flos": 24820916630400.0,
      "grad_norm": 2.883054714080116,
      "language_loss": 0.71693021,
      "learning_rate": 2.539717736776241e-07,
      "loss": 0.73836941,
      "num_input_tokens_seen": 151404970,
      "step": 7008,
      "time_per_iteration": 2.551286220550537
    },
    {
      "auxiliary_loss_clip": 0.01148071,
      "auxiliary_loss_mlp": 0.01020657,
      "balance_loss_clip": 1.04632854,
      "balance_loss_mlp": 1.01380873,
      "epoch": 0.8427824204893886,
      "flos": 23550074467200.0,
      "grad_norm": 1.3407034826020992,
      "language_loss": 0.76278132,
      "learning_rate": 2.535920055890097e-07,
      "loss": 0.78446853,
      "num_input_tokens_seen": 151426265,
      "step": 7009,
      "time_per_iteration": 3.2268190383911133
    },
    {
      "auxiliary_loss_clip": 0.0110609,
      "auxiliary_loss_mlp": 0.01024158,
      "balance_loss_clip": 1.04093981,
      "balance_loss_mlp": 1.0167129,
      "epoch": 0.8429026633800276,
      "flos": 16143858120960.0,
      "grad_norm": 2.1513441011988768,
      "language_loss": 0.64647067,
      "learning_rate": 2.5321250242910006e-07,
      "loss": 0.66777313,
      "num_input_tokens_seen": 151444180,
      "step": 7010,
      "time_per_iteration": 2.519245147705078
    },
    {
      "auxiliary_loss_clip": 0.01164749,
      "auxiliary_loss_mlp": 0.01021121,
      "balance_loss_clip": 1.04798377,
      "balance_loss_mlp": 1.01430249,
      "epoch": 0.8430229062706668,
      "flos": 22198540400640.0,
      "grad_norm": 1.7210399338278748,
      "language_loss": 0.8663035,
      "learning_rate": 2.5283326425546493e-07,
      "loss": 0.8881622,
      "num_input_tokens_seen": 151463290,
      "step": 7011,
      "time_per_iteration": 2.4217967987060547
    },
    {
      "auxiliary_loss_clip": 0.01118338,
      "auxiliary_loss_mlp": 0.01021769,
      "balance_loss_clip": 1.04444456,
      "balance_loss_mlp": 1.01512289,
      "epoch": 0.8431431491613058,
      "flos": 35330317683840.0,
      "grad_norm": 2.4405993873849874,
      "language_loss": 0.6960088,
      "learning_rate": 2.5245429112563443e-07,
      "loss": 0.71740985,
      "num_input_tokens_seen": 151483965,
      "step": 7012,
      "time_per_iteration": 2.660486936569214
    },
    {
      "auxiliary_loss_clip": 0.01151365,
      "auxiliary_loss_mlp": 0.01025005,
      "balance_loss_clip": 1.04739499,
      "balance_loss_mlp": 1.01815689,
      "epoch": 0.8432633920519449,
      "flos": 25812374808960.0,
      "grad_norm": 2.11706414987177,
      "language_loss": 0.81914449,
      "learning_rate": 2.5207558309709865e-07,
      "loss": 0.84090823,
      "num_input_tokens_seen": 151503700,
      "step": 7013,
      "time_per_iteration": 2.4803364276885986
    },
    {
      "auxiliary_loss_clip": 0.01036122,
      "auxiliary_loss_mlp": 0.00752945,
      "balance_loss_clip": 1.00835383,
      "balance_loss_mlp": 0.99995768,
      "epoch": 0.8433836349425841,
      "flos": 64959531592320.0,
      "grad_norm": 0.6855319897730692,
      "language_loss": 0.56307161,
      "learning_rate": 2.516971402273065e-07,
      "loss": 0.58096224,
      "num_input_tokens_seen": 151569765,
      "step": 7014,
      "time_per_iteration": 3.092280864715576
    },
    {
      "auxiliary_loss_clip": 0.01138566,
      "auxiliary_loss_mlp": 0.0102131,
      "balance_loss_clip": 1.04304528,
      "balance_loss_mlp": 1.01442587,
      "epoch": 0.8435038778332231,
      "flos": 20229989483520.0,
      "grad_norm": 2.0057004313342417,
      "language_loss": 0.67897224,
      "learning_rate": 2.513189625736687e-07,
      "loss": 0.70057106,
      "num_input_tokens_seen": 151586660,
      "step": 7015,
      "time_per_iteration": 2.4535529613494873
    },
    {
      "auxiliary_loss_clip": 0.01127464,
      "auxiliary_loss_mlp": 0.01029711,
      "balance_loss_clip": 1.04397988,
      "balance_loss_mlp": 1.02226651,
      "epoch": 0.8436241207238622,
      "flos": 20992229020800.0,
      "grad_norm": 2.3479606485779536,
      "language_loss": 0.71756548,
      "learning_rate": 2.509410501935534e-07,
      "loss": 0.73913723,
      "num_input_tokens_seen": 151602295,
      "step": 7016,
      "time_per_iteration": 2.494269847869873
    },
    {
      "auxiliary_loss_clip": 0.01137849,
      "auxiliary_loss_mlp": 0.01025129,
      "balance_loss_clip": 1.04366112,
      "balance_loss_mlp": 1.01775014,
      "epoch": 0.8437443636145013,
      "flos": 14682257804160.0,
      "grad_norm": 2.8334619111776687,
      "language_loss": 0.75318938,
      "learning_rate": 2.5056340314429116e-07,
      "loss": 0.77481914,
      "num_input_tokens_seen": 151619760,
      "step": 7017,
      "time_per_iteration": 2.4375545978546143
    },
    {
      "auxiliary_loss_clip": 0.01112279,
      "auxiliary_loss_mlp": 0.01026009,
      "balance_loss_clip": 1.04012799,
      "balance_loss_mlp": 1.01819444,
      "epoch": 0.8438646065051404,
      "flos": 21608814908160.0,
      "grad_norm": 2.265978582369029,
      "language_loss": 0.80613446,
      "learning_rate": 2.5018602148316904e-07,
      "loss": 0.82751739,
      "num_input_tokens_seen": 151635795,
      "step": 7018,
      "time_per_iteration": 2.542299509048462
    },
    {
      "auxiliary_loss_clip": 0.01164011,
      "auxiliary_loss_mlp": 0.01024838,
      "balance_loss_clip": 1.04841876,
      "balance_loss_mlp": 1.01825452,
      "epoch": 0.8439848493957794,
      "flos": 23289937194240.0,
      "grad_norm": 2.1239860431179127,
      "language_loss": 0.79964215,
      "learning_rate": 2.498089052674359e-07,
      "loss": 0.82153064,
      "num_input_tokens_seen": 151653770,
      "step": 7019,
      "time_per_iteration": 2.4089088439941406
    },
    {
      "auxiliary_loss_clip": 0.01151191,
      "auxiliary_loss_mlp": 0.01031505,
      "balance_loss_clip": 1.04663754,
      "balance_loss_mlp": 1.02433455,
      "epoch": 0.8441050922864186,
      "flos": 19719339782400.0,
      "grad_norm": 4.8655346410299085,
      "language_loss": 0.75239706,
      "learning_rate": 2.494320545543007e-07,
      "loss": 0.77422404,
      "num_input_tokens_seen": 151673340,
      "step": 7020,
      "time_per_iteration": 2.429816246032715
    },
    {
      "auxiliary_loss_clip": 0.01168795,
      "auxiliary_loss_mlp": 0.01025901,
      "balance_loss_clip": 1.04787779,
      "balance_loss_mlp": 1.01792574,
      "epoch": 0.8442253351770577,
      "flos": 21835268202240.0,
      "grad_norm": 2.9438224088060516,
      "language_loss": 0.66626304,
      "learning_rate": 2.490554694009308e-07,
      "loss": 0.68821001,
      "num_input_tokens_seen": 151694205,
      "step": 7021,
      "time_per_iteration": 2.442040205001831
    },
    {
      "auxiliary_loss_clip": 0.01154134,
      "auxiliary_loss_mlp": 0.01028043,
      "balance_loss_clip": 1.04513526,
      "balance_loss_mlp": 1.02133143,
      "epoch": 0.8443455780676967,
      "flos": 34346365447680.0,
      "grad_norm": 1.690566373744392,
      "language_loss": 0.78432953,
      "learning_rate": 2.4867914986445426e-07,
      "loss": 0.80615127,
      "num_input_tokens_seen": 151716595,
      "step": 7022,
      "time_per_iteration": 2.5730972290039062
    },
    {
      "auxiliary_loss_clip": 0.01139241,
      "auxiliary_loss_mlp": 0.01024833,
      "balance_loss_clip": 1.04228497,
      "balance_loss_mlp": 1.01833272,
      "epoch": 0.8444658209583359,
      "flos": 48214599281280.0,
      "grad_norm": 1.8828106901096475,
      "language_loss": 0.71129698,
      "learning_rate": 2.483030960019581e-07,
      "loss": 0.73293775,
      "num_input_tokens_seen": 151740525,
      "step": 7023,
      "time_per_iteration": 2.7226295471191406
    },
    {
      "auxiliary_loss_clip": 0.01018381,
      "auxiliary_loss_mlp": 0.01003795,
      "balance_loss_clip": 1.0081439,
      "balance_loss_mlp": 1.00277555,
      "epoch": 0.8445860638489749,
      "flos": 68484773105280.0,
      "grad_norm": 0.7366740590603003,
      "language_loss": 0.55469286,
      "learning_rate": 2.479273078704891e-07,
      "loss": 0.57491457,
      "num_input_tokens_seen": 151793890,
      "step": 7024,
      "time_per_iteration": 2.962144136428833
    },
    {
      "auxiliary_loss_clip": 0.01014278,
      "auxiliary_loss_mlp": 0.01003095,
      "balance_loss_clip": 1.01141953,
      "balance_loss_mlp": 1.00209367,
      "epoch": 0.844706306739614,
      "flos": 62833331882880.0,
      "grad_norm": 0.7809757381921016,
      "language_loss": 0.6473707,
      "learning_rate": 2.475517855270552e-07,
      "loss": 0.66754442,
      "num_input_tokens_seen": 151853970,
      "step": 7025,
      "time_per_iteration": 3.0823192596435547
    },
    {
      "auxiliary_loss_clip": 0.0116425,
      "auxiliary_loss_mlp": 0.01024306,
      "balance_loss_clip": 1.04844451,
      "balance_loss_mlp": 1.01764858,
      "epoch": 0.8448265496302532,
      "flos": 14976114969600.0,
      "grad_norm": 1.8202925918147292,
      "language_loss": 0.72594321,
      "learning_rate": 2.4717652902862143e-07,
      "loss": 0.74782872,
      "num_input_tokens_seen": 151872945,
      "step": 7026,
      "time_per_iteration": 2.411222457885742
    },
    {
      "auxiliary_loss_clip": 0.01140705,
      "auxiliary_loss_mlp": 0.01023645,
      "balance_loss_clip": 1.04426026,
      "balance_loss_mlp": 1.01702857,
      "epoch": 0.8449467925208922,
      "flos": 23441265192960.0,
      "grad_norm": 1.617793512075884,
      "language_loss": 0.81420934,
      "learning_rate": 2.4680153843211495e-07,
      "loss": 0.8358528,
      "num_input_tokens_seen": 151892875,
      "step": 7027,
      "time_per_iteration": 2.5046610832214355
    },
    {
      "auxiliary_loss_clip": 0.01136664,
      "auxiliary_loss_mlp": 0.01022622,
      "balance_loss_clip": 1.04533029,
      "balance_loss_mlp": 1.01553178,
      "epoch": 0.8450670354115313,
      "flos": 22748045639040.0,
      "grad_norm": 1.659828788536089,
      "language_loss": 0.72587234,
      "learning_rate": 2.464268137944212e-07,
      "loss": 0.74746519,
      "num_input_tokens_seen": 151914170,
      "step": 7028,
      "time_per_iteration": 3.3836312294006348
    },
    {
      "auxiliary_loss_clip": 0.01099818,
      "auxiliary_loss_mlp": 0.01027749,
      "balance_loss_clip": 1.0400908,
      "balance_loss_mlp": 1.01974404,
      "epoch": 0.8451872783021703,
      "flos": 29825571605760.0,
      "grad_norm": 2.9001062545065763,
      "language_loss": 0.78255951,
      "learning_rate": 2.46052355172385e-07,
      "loss": 0.80383521,
      "num_input_tokens_seen": 151932210,
      "step": 7029,
      "time_per_iteration": 3.48899507522583
    },
    {
      "auxiliary_loss_clip": 0.01164604,
      "auxiliary_loss_mlp": 0.01028765,
      "balance_loss_clip": 1.04637384,
      "balance_loss_mlp": 1.02118921,
      "epoch": 0.8453075211928095,
      "flos": 21870029589120.0,
      "grad_norm": 1.7236844803737197,
      "language_loss": 0.74734879,
      "learning_rate": 2.456781626228128e-07,
      "loss": 0.76928246,
      "num_input_tokens_seen": 151951715,
      "step": 7030,
      "time_per_iteration": 3.3004095554351807
    },
    {
      "auxiliary_loss_clip": 0.01022387,
      "auxiliary_loss_mlp": 0.00753187,
      "balance_loss_clip": 1.00737798,
      "balance_loss_mlp": 0.9999218,
      "epoch": 0.8454277640834486,
      "flos": 58751869288320.0,
      "grad_norm": 0.9131877959224518,
      "language_loss": 0.66333985,
      "learning_rate": 2.453042362024675e-07,
      "loss": 0.6810956,
      "num_input_tokens_seen": 152004960,
      "step": 7031,
      "time_per_iteration": 3.135413885116577
    },
    {
      "auxiliary_loss_clip": 0.01163056,
      "auxiliary_loss_mlp": 0.0102616,
      "balance_loss_clip": 1.04623342,
      "balance_loss_mlp": 1.01947212,
      "epoch": 0.8455480069740876,
      "flos": 27090076469760.0,
      "grad_norm": 1.6454883883613196,
      "language_loss": 0.73299348,
      "learning_rate": 2.449305759680751e-07,
      "loss": 0.75488567,
      "num_input_tokens_seen": 152026285,
      "step": 7032,
      "time_per_iteration": 2.4569928646087646
    },
    {
      "auxiliary_loss_clip": 0.01121787,
      "auxiliary_loss_mlp": 0.01026833,
      "balance_loss_clip": 1.04415393,
      "balance_loss_mlp": 1.01988935,
      "epoch": 0.8456682498647268,
      "flos": 27198670262400.0,
      "grad_norm": 1.5675712988851356,
      "language_loss": 0.75331032,
      "learning_rate": 2.445571819763188e-07,
      "loss": 0.77479649,
      "num_input_tokens_seen": 152048585,
      "step": 7033,
      "time_per_iteration": 2.6643624305725098
    },
    {
      "auxiliary_loss_clip": 0.01164267,
      "auxiliary_loss_mlp": 0.01030296,
      "balance_loss_clip": 1.0476284,
      "balance_loss_mlp": 1.02287579,
      "epoch": 0.8457884927553658,
      "flos": 20631901737600.0,
      "grad_norm": 1.77588912014989,
      "language_loss": 0.58323652,
      "learning_rate": 2.4418405428384227e-07,
      "loss": 0.60518217,
      "num_input_tokens_seen": 152068795,
      "step": 7034,
      "time_per_iteration": 2.436022996902466
    },
    {
      "auxiliary_loss_clip": 0.0116601,
      "auxiliary_loss_mlp": 0.00761763,
      "balance_loss_clip": 1.04790545,
      "balance_loss_mlp": 1.00046527,
      "epoch": 0.8459087356460049,
      "flos": 15299023259520.0,
      "grad_norm": 1.7307936949013905,
      "language_loss": 0.71941221,
      "learning_rate": 2.4381119294724864e-07,
      "loss": 0.7386899,
      "num_input_tokens_seen": 152086240,
      "step": 7035,
      "time_per_iteration": 2.3920671939849854
    },
    {
      "auxiliary_loss_clip": 0.01165406,
      "auxiliary_loss_mlp": 0.01021897,
      "balance_loss_clip": 1.04743385,
      "balance_loss_mlp": 1.01508093,
      "epoch": 0.846028978536644,
      "flos": 18843155326080.0,
      "grad_norm": 3.6283295659671184,
      "language_loss": 0.53970593,
      "learning_rate": 2.434385980231004e-07,
      "loss": 0.56157893,
      "num_input_tokens_seen": 152105080,
      "step": 7036,
      "time_per_iteration": 3.1459147930145264
    },
    {
      "auxiliary_loss_clip": 0.01150569,
      "auxiliary_loss_mlp": 0.01027904,
      "balance_loss_clip": 1.04591298,
      "balance_loss_mlp": 1.02069736,
      "epoch": 0.8461492214272831,
      "flos": 52661740285440.0,
      "grad_norm": 1.492225457134181,
      "language_loss": 0.6556018,
      "learning_rate": 2.4306626956792043e-07,
      "loss": 0.67738652,
      "num_input_tokens_seen": 152130025,
      "step": 7037,
      "time_per_iteration": 2.719534158706665
    },
    {
      "auxiliary_loss_clip": 0.01148756,
      "auxiliary_loss_mlp": 0.01024081,
      "balance_loss_clip": 1.04312038,
      "balance_loss_mlp": 1.0173254,
      "epoch": 0.8462694643179222,
      "flos": 18588405093120.0,
      "grad_norm": 1.677755895706697,
      "language_loss": 0.75646615,
      "learning_rate": 2.4269420763819017e-07,
      "loss": 0.77819455,
      "num_input_tokens_seen": 152148070,
      "step": 7038,
      "time_per_iteration": 2.429368257522583
    },
    {
      "auxiliary_loss_clip": 0.01147268,
      "auxiliary_loss_mlp": 0.01022207,
      "balance_loss_clip": 1.04502559,
      "balance_loss_mlp": 1.01565695,
      "epoch": 0.8463897072085613,
      "flos": 24387080163840.0,
      "grad_norm": 2.3201240319917944,
      "language_loss": 0.83149451,
      "learning_rate": 2.4232241229035223e-07,
      "loss": 0.85318923,
      "num_input_tokens_seen": 152165825,
      "step": 7039,
      "time_per_iteration": 2.4675028324127197
    },
    {
      "auxiliary_loss_clip": 0.01052888,
      "auxiliary_loss_mlp": 0.01000745,
      "balance_loss_clip": 1.0079565,
      "balance_loss_mlp": 0.99983925,
      "epoch": 0.8465099500992004,
      "flos": 68702140258560.0,
      "grad_norm": 0.8032244963941125,
      "language_loss": 0.5678364,
      "learning_rate": 2.419508835808064e-07,
      "loss": 0.58837271,
      "num_input_tokens_seen": 152222380,
      "step": 7040,
      "time_per_iteration": 2.9750545024871826
    },
    {
      "auxiliary_loss_clip": 0.0113739,
      "auxiliary_loss_mlp": 0.01021965,
      "balance_loss_clip": 1.04579735,
      "balance_loss_mlp": 1.01506317,
      "epoch": 0.8466301929898394,
      "flos": 13735724561280.0,
      "grad_norm": 1.9433629375392678,
      "language_loss": 0.6314559,
      "learning_rate": 2.415796215659134e-07,
      "loss": 0.65304947,
      "num_input_tokens_seen": 152239085,
      "step": 7041,
      "time_per_iteration": 2.4568138122558594
    },
    {
      "auxiliary_loss_clip": 0.011259,
      "auxiliary_loss_mlp": 0.01031079,
      "balance_loss_clip": 1.04076469,
      "balance_loss_mlp": 1.02414083,
      "epoch": 0.8467504358804786,
      "flos": 19241260738560.0,
      "grad_norm": 2.011923393267298,
      "language_loss": 0.77123839,
      "learning_rate": 2.412086263019939e-07,
      "loss": 0.79280818,
      "num_input_tokens_seen": 152257110,
      "step": 7042,
      "time_per_iteration": 2.4937891960144043
    },
    {
      "auxiliary_loss_clip": 0.01160337,
      "auxiliary_loss_mlp": 0.01022558,
      "balance_loss_clip": 1.04707003,
      "balance_loss_mlp": 1.01592696,
      "epoch": 0.8468706787711177,
      "flos": 21324115710720.0,
      "grad_norm": 2.320967958914889,
      "language_loss": 0.79747272,
      "learning_rate": 2.408378978453276e-07,
      "loss": 0.81930166,
      "num_input_tokens_seen": 152277230,
      "step": 7043,
      "time_per_iteration": 2.4199469089508057
    },
    {
      "auxiliary_loss_clip": 0.01053078,
      "auxiliary_loss_mlp": 0.01002209,
      "balance_loss_clip": 1.00811934,
      "balance_loss_mlp": 1.00129664,
      "epoch": 0.8469909216617567,
      "flos": 64877439058560.0,
      "grad_norm": 0.8116887091471326,
      "language_loss": 0.63976562,
      "learning_rate": 2.404674362521533e-07,
      "loss": 0.66031849,
      "num_input_tokens_seen": 152335725,
      "step": 7044,
      "time_per_iteration": 2.9290642738342285
    },
    {
      "auxiliary_loss_clip": 0.01151659,
      "auxiliary_loss_mlp": 0.0102219,
      "balance_loss_clip": 1.04819751,
      "balance_loss_mlp": 1.01595497,
      "epoch": 0.8471111645523959,
      "flos": 19280583152640.0,
      "grad_norm": 2.2022597730265208,
      "language_loss": 0.74563384,
      "learning_rate": 2.4009724157866997e-07,
      "loss": 0.76737237,
      "num_input_tokens_seen": 152352785,
      "step": 7045,
      "time_per_iteration": 2.4344186782836914
    },
    {
      "auxiliary_loss_clip": 0.0116297,
      "auxiliary_loss_mlp": 0.01022478,
      "balance_loss_clip": 1.04652834,
      "balance_loss_mlp": 1.01587665,
      "epoch": 0.8472314074430349,
      "flos": 22015826893440.0,
      "grad_norm": 2.0010858820738973,
      "language_loss": 0.76488608,
      "learning_rate": 2.3972731388103564e-07,
      "loss": 0.78674054,
      "num_input_tokens_seen": 152371265,
      "step": 7046,
      "time_per_iteration": 2.411935806274414
    },
    {
      "auxiliary_loss_clip": 0.01005485,
      "auxiliary_loss_mlp": 0.01002247,
      "balance_loss_clip": 1.00797129,
      "balance_loss_mlp": 1.00129282,
      "epoch": 0.847351650333674,
      "flos": 57882580243200.0,
      "grad_norm": 0.8081796483934046,
      "language_loss": 0.62394696,
      "learning_rate": 2.393576532153687e-07,
      "loss": 0.64402425,
      "num_input_tokens_seen": 152435050,
      "step": 7047,
      "time_per_iteration": 3.3289384841918945
    },
    {
      "auxiliary_loss_clip": 0.01049463,
      "auxiliary_loss_mlp": 0.01000561,
      "balance_loss_clip": 1.00717735,
      "balance_loss_mlp": 0.99964905,
      "epoch": 0.8474718932243132,
      "flos": 41284238313600.0,
      "grad_norm": 1.0148745035124487,
      "language_loss": 0.57805216,
      "learning_rate": 2.389882596377453e-07,
      "loss": 0.59855247,
      "num_input_tokens_seen": 152489315,
      "step": 7048,
      "time_per_iteration": 3.1335747241973877
    },
    {
      "auxiliary_loss_clip": 0.01162949,
      "auxiliary_loss_mlp": 0.01023258,
      "balance_loss_clip": 1.04563689,
      "balance_loss_mlp": 1.01646328,
      "epoch": 0.8475921361149522,
      "flos": 38180906974080.0,
      "grad_norm": 1.783882695297717,
      "language_loss": 0.76419449,
      "learning_rate": 2.386191332042031e-07,
      "loss": 0.78605652,
      "num_input_tokens_seen": 152511210,
      "step": 7049,
      "time_per_iteration": 2.559741973876953
    },
    {
      "auxiliary_loss_clip": 0.01170272,
      "auxiliary_loss_mlp": 0.01029119,
      "balance_loss_clip": 1.04890227,
      "balance_loss_mlp": 1.02202296,
      "epoch": 0.8477123790055913,
      "flos": 25375054723200.0,
      "grad_norm": 1.6930635494095527,
      "language_loss": 0.72977793,
      "learning_rate": 2.3825027397073794e-07,
      "loss": 0.75177181,
      "num_input_tokens_seen": 152531685,
      "step": 7050,
      "time_per_iteration": 2.455000638961792
    },
    {
      "auxiliary_loss_clip": 0.01149127,
      "auxiliary_loss_mlp": 0.01024604,
      "balance_loss_clip": 1.04783297,
      "balance_loss_mlp": 1.01772869,
      "epoch": 0.8478326218962304,
      "flos": 30225185389440.0,
      "grad_norm": 1.9775603583509382,
      "language_loss": 0.66726625,
      "learning_rate": 2.3788168199330515e-07,
      "loss": 0.68900353,
      "num_input_tokens_seen": 152553245,
      "step": 7051,
      "time_per_iteration": 2.502920389175415
    },
    {
      "auxiliary_loss_clip": 0.01121482,
      "auxiliary_loss_mlp": 0.01023664,
      "balance_loss_clip": 1.03889132,
      "balance_loss_mlp": 1.01621366,
      "epoch": 0.8479528647868695,
      "flos": 38213800853760.0,
      "grad_norm": 1.5450701602903603,
      "language_loss": 0.72626805,
      "learning_rate": 2.3751335732782074e-07,
      "loss": 0.74771953,
      "num_input_tokens_seen": 152574505,
      "step": 7052,
      "time_per_iteration": 2.6080124378204346
    },
    {
      "auxiliary_loss_clip": 0.01151635,
      "auxiliary_loss_mlp": 0.01023658,
      "balance_loss_clip": 1.04819822,
      "balance_loss_mlp": 1.01690161,
      "epoch": 0.8480731076775085,
      "flos": 20957790856320.0,
      "grad_norm": 1.9118088469892158,
      "language_loss": 0.79593438,
      "learning_rate": 2.371453000301582e-07,
      "loss": 0.81768733,
      "num_input_tokens_seen": 152593190,
      "step": 7053,
      "time_per_iteration": 2.4335508346557617
    },
    {
      "auxiliary_loss_clip": 0.01119364,
      "auxiliary_loss_mlp": 0.01020221,
      "balance_loss_clip": 1.04296207,
      "balance_loss_mlp": 1.01344085,
      "epoch": 0.8481933505681477,
      "flos": 32596510487040.0,
      "grad_norm": 1.6431137248093355,
      "language_loss": 0.7410087,
      "learning_rate": 2.3677751015615222e-07,
      "loss": 0.76240456,
      "num_input_tokens_seen": 152615265,
      "step": 7054,
      "time_per_iteration": 2.5895867347717285
    },
    {
      "auxiliary_loss_clip": 0.01125532,
      "auxiliary_loss_mlp": 0.01029679,
      "balance_loss_clip": 1.04154372,
      "balance_loss_mlp": 1.02259183,
      "epoch": 0.8483135934587868,
      "flos": 20741177888640.0,
      "grad_norm": 4.149614036353006,
      "language_loss": 0.85451144,
      "learning_rate": 2.3640998776159593e-07,
      "loss": 0.87606359,
      "num_input_tokens_seen": 152632770,
      "step": 7055,
      "time_per_iteration": 4.348072290420532
    },
    {
      "auxiliary_loss_clip": 0.01137351,
      "auxiliary_loss_mlp": 0.01022682,
      "balance_loss_clip": 1.04526162,
      "balance_loss_mlp": 1.01656961,
      "epoch": 0.8484338363494258,
      "flos": 21653057485440.0,
      "grad_norm": 1.6847481393658448,
      "language_loss": 0.81471545,
      "learning_rate": 2.3604273290224253e-07,
      "loss": 0.83631575,
      "num_input_tokens_seen": 152653485,
      "step": 7056,
      "time_per_iteration": 2.500746965408325
    },
    {
      "auxiliary_loss_clip": 0.01140739,
      "auxiliary_loss_mlp": 0.01029285,
      "balance_loss_clip": 1.04679561,
      "balance_loss_mlp": 1.02154207,
      "epoch": 0.848554079240065,
      "flos": 15013964926080.0,
      "grad_norm": 3.531464086747567,
      "language_loss": 0.74845701,
      "learning_rate": 2.356757456338039e-07,
      "loss": 0.77015722,
      "num_input_tokens_seen": 152670970,
      "step": 7057,
      "time_per_iteration": 3.2922630310058594
    },
    {
      "auxiliary_loss_clip": 0.01039997,
      "auxiliary_loss_mlp": 0.01001309,
      "balance_loss_clip": 1.01098704,
      "balance_loss_mlp": 1.00036764,
      "epoch": 0.848674322130704,
      "flos": 68060453742720.0,
      "grad_norm": 0.7689343665705757,
      "language_loss": 0.59039354,
      "learning_rate": 2.3530902601195147e-07,
      "loss": 0.61080658,
      "num_input_tokens_seen": 152739460,
      "step": 7058,
      "time_per_iteration": 3.1545004844665527
    },
    {
      "auxiliary_loss_clip": 0.01147982,
      "auxiliary_loss_mlp": 0.01027185,
      "balance_loss_clip": 1.04442942,
      "balance_loss_mlp": 1.0201664,
      "epoch": 0.8487945650213431,
      "flos": 18475788977280.0,
      "grad_norm": 6.813239192788781,
      "language_loss": 0.79047465,
      "learning_rate": 2.34942574092317e-07,
      "loss": 0.8122263,
      "num_input_tokens_seen": 152754710,
      "step": 7059,
      "time_per_iteration": 2.401810646057129
    },
    {
      "auxiliary_loss_clip": 0.01154059,
      "auxiliary_loss_mlp": 0.01025475,
      "balance_loss_clip": 1.04601407,
      "balance_loss_mlp": 1.01834035,
      "epoch": 0.8489148079119821,
      "flos": 23473189405440.0,
      "grad_norm": 2.474307463828105,
      "language_loss": 0.76760042,
      "learning_rate": 2.3457638993049045e-07,
      "loss": 0.78939581,
      "num_input_tokens_seen": 152772700,
      "step": 7060,
      "time_per_iteration": 2.512019157409668
    },
    {
      "auxiliary_loss_clip": 0.01099105,
      "auxiliary_loss_mlp": 0.01023676,
      "balance_loss_clip": 1.04408264,
      "balance_loss_mlp": 1.01621664,
      "epoch": 0.8490350508026213,
      "flos": 19937604775680.0,
      "grad_norm": 1.9053257710084055,
      "language_loss": 0.64234185,
      "learning_rate": 2.3421047358202252e-07,
      "loss": 0.66356963,
      "num_input_tokens_seen": 152791550,
      "step": 7061,
      "time_per_iteration": 2.5750207901000977
    },
    {
      "auxiliary_loss_clip": 0.01152972,
      "auxiliary_loss_mlp": 0.01026238,
      "balance_loss_clip": 1.04677105,
      "balance_loss_mlp": 1.01937747,
      "epoch": 0.8491552936932604,
      "flos": 24279958828800.0,
      "grad_norm": 2.2386390924518813,
      "language_loss": 0.83409441,
      "learning_rate": 2.3384482510242144e-07,
      "loss": 0.85588646,
      "num_input_tokens_seen": 152809410,
      "step": 7062,
      "time_per_iteration": 2.45922589302063
    },
    {
      "auxiliary_loss_clip": 0.01166247,
      "auxiliary_loss_mlp": 0.01025929,
      "balance_loss_clip": 1.04649138,
      "balance_loss_mlp": 1.01872015,
      "epoch": 0.8492755365838994,
      "flos": 22522526098560.0,
      "grad_norm": 2.5269522640472566,
      "language_loss": 0.77260256,
      "learning_rate": 2.3347944454715575e-07,
      "loss": 0.79452431,
      "num_input_tokens_seen": 152825800,
      "step": 7063,
      "time_per_iteration": 3.300910472869873
    },
    {
      "auxiliary_loss_clip": 0.01168376,
      "auxiliary_loss_mlp": 0.01023725,
      "balance_loss_clip": 1.04738522,
      "balance_loss_mlp": 1.01630998,
      "epoch": 0.8493957794745386,
      "flos": 26980441182720.0,
      "grad_norm": 1.7028184501106434,
      "language_loss": 0.67474258,
      "learning_rate": 2.331143319716542e-07,
      "loss": 0.69666356,
      "num_input_tokens_seen": 152845330,
      "step": 7064,
      "time_per_iteration": 2.4657676219940186
    },
    {
      "auxiliary_loss_clip": 0.01126241,
      "auxiliary_loss_mlp": 0.01024094,
      "balance_loss_clip": 1.04288673,
      "balance_loss_mlp": 1.0171051,
      "epoch": 0.8495160223651776,
      "flos": 29861985018240.0,
      "grad_norm": 2.0395667671375084,
      "language_loss": 0.66012293,
      "learning_rate": 2.3274948743130363e-07,
      "loss": 0.68162626,
      "num_input_tokens_seen": 152865165,
      "step": 7065,
      "time_per_iteration": 2.5739710330963135
    },
    {
      "auxiliary_loss_clip": 0.01164269,
      "auxiliary_loss_mlp": 0.01022455,
      "balance_loss_clip": 1.04554188,
      "balance_loss_mlp": 1.01505852,
      "epoch": 0.8496362652558167,
      "flos": 23075443128960.0,
      "grad_norm": 1.573993371533445,
      "language_loss": 0.79392433,
      "learning_rate": 2.3238491098145085e-07,
      "loss": 0.81579161,
      "num_input_tokens_seen": 152884695,
      "step": 7066,
      "time_per_iteration": 2.4210197925567627
    },
    {
      "auxiliary_loss_clip": 0.011484,
      "auxiliary_loss_mlp": 0.01021955,
      "balance_loss_clip": 1.04456878,
      "balance_loss_mlp": 1.01509786,
      "epoch": 0.8497565081464559,
      "flos": 14609107756800.0,
      "grad_norm": 2.121251356048633,
      "language_loss": 0.73573291,
      "learning_rate": 2.3202060267740141e-07,
      "loss": 0.75743645,
      "num_input_tokens_seen": 152902220,
      "step": 7067,
      "time_per_iteration": 2.4155685901641846
    },
    {
      "auxiliary_loss_clip": 0.0110278,
      "auxiliary_loss_mlp": 0.01019291,
      "balance_loss_clip": 1.03824425,
      "balance_loss_mlp": 1.01261258,
      "epoch": 0.8498767510370949,
      "flos": 21136446126720.0,
      "grad_norm": 2.1323481787273066,
      "language_loss": 0.77249926,
      "learning_rate": 2.3165656257442044e-07,
      "loss": 0.79372001,
      "num_input_tokens_seen": 152920740,
      "step": 7068,
      "time_per_iteration": 2.5306179523468018
    },
    {
      "auxiliary_loss_clip": 0.01147328,
      "auxiliary_loss_mlp": 0.01020414,
      "balance_loss_clip": 1.04548645,
      "balance_loss_mlp": 1.01403379,
      "epoch": 0.849996993927734,
      "flos": 23654538195840.0,
      "grad_norm": 2.1418123922771297,
      "language_loss": 0.89970219,
      "learning_rate": 2.31292790727734e-07,
      "loss": 0.92137963,
      "num_input_tokens_seen": 152938305,
      "step": 7069,
      "time_per_iteration": 2.462441921234131
    },
    {
      "auxiliary_loss_clip": 0.01161451,
      "auxiliary_loss_mlp": 0.01022922,
      "balance_loss_clip": 1.04514194,
      "balance_loss_mlp": 1.01657152,
      "epoch": 0.8501172368183731,
      "flos": 20558069331840.0,
      "grad_norm": 2.718824295277621,
      "language_loss": 0.80245382,
      "learning_rate": 2.3092928719252392e-07,
      "loss": 0.82429755,
      "num_input_tokens_seen": 152956705,
      "step": 7070,
      "time_per_iteration": 2.419555187225342
    },
    {
      "auxiliary_loss_clip": 0.01148455,
      "auxiliary_loss_mlp": 0.01024365,
      "balance_loss_clip": 1.04458272,
      "balance_loss_mlp": 1.01728415,
      "epoch": 0.8502374797090122,
      "flos": 22272624201600.0,
      "grad_norm": 1.8605462975381586,
      "language_loss": 0.78503716,
      "learning_rate": 2.3056605202393475e-07,
      "loss": 0.80676532,
      "num_input_tokens_seen": 152974265,
      "step": 7071,
      "time_per_iteration": 2.4259607791900635
    },
    {
      "auxiliary_loss_clip": 0.01146862,
      "auxiliary_loss_mlp": 0.0076227,
      "balance_loss_clip": 1.04307997,
      "balance_loss_mlp": 1.00040865,
      "epoch": 0.8503577225996513,
      "flos": 23659817495040.0,
      "grad_norm": 1.741231877719285,
      "language_loss": 0.66835445,
      "learning_rate": 2.3020308527706888e-07,
      "loss": 0.68744576,
      "num_input_tokens_seen": 152993680,
      "step": 7072,
      "time_per_iteration": 2.4568064212799072
    },
    {
      "auxiliary_loss_clip": 0.01141689,
      "auxiliary_loss_mlp": 0.01024764,
      "balance_loss_clip": 1.04329515,
      "balance_loss_mlp": 1.01753986,
      "epoch": 0.8504779654902904,
      "flos": 26758513002240.0,
      "grad_norm": 1.5741104597957458,
      "language_loss": 0.88587922,
      "learning_rate": 2.2984038700698715e-07,
      "loss": 0.90754372,
      "num_input_tokens_seen": 153012990,
      "step": 7073,
      "time_per_iteration": 2.512040615081787
    },
    {
      "auxiliary_loss_clip": 0.0114737,
      "auxiliary_loss_mlp": 0.01028099,
      "balance_loss_clip": 1.04633343,
      "balance_loss_mlp": 1.02081561,
      "epoch": 0.8505982083809295,
      "flos": 26468247196800.0,
      "grad_norm": 2.330735734679678,
      "language_loss": 0.78830302,
      "learning_rate": 2.2947795726871222e-07,
      "loss": 0.8100577,
      "num_input_tokens_seen": 153034015,
      "step": 7074,
      "time_per_iteration": 2.471318483352661
    },
    {
      "auxiliary_loss_clip": 0.01151886,
      "auxiliary_loss_mlp": 0.00761997,
      "balance_loss_clip": 1.05040956,
      "balance_loss_mlp": 1.00046706,
      "epoch": 0.8507184512715685,
      "flos": 20303390926080.0,
      "grad_norm": 1.6814630545809488,
      "language_loss": 0.85835367,
      "learning_rate": 2.2911579611722253e-07,
      "loss": 0.87749255,
      "num_input_tokens_seen": 153053160,
      "step": 7075,
      "time_per_iteration": 2.4278669357299805
    },
    {
      "auxiliary_loss_clip": 0.01134444,
      "auxiliary_loss_mlp": 0.01027121,
      "balance_loss_clip": 1.04330981,
      "balance_loss_mlp": 1.01997757,
      "epoch": 0.8508386941622077,
      "flos": 19025186474880.0,
      "grad_norm": 10.859556549340144,
      "language_loss": 0.87158132,
      "learning_rate": 2.2875390360745905e-07,
      "loss": 0.89319694,
      "num_input_tokens_seen": 153072565,
      "step": 7076,
      "time_per_iteration": 2.447910785675049
    },
    {
      "auxiliary_loss_clip": 0.01127973,
      "auxiliary_loss_mlp": 0.01026943,
      "balance_loss_clip": 1.04248953,
      "balance_loss_mlp": 1.01969528,
      "epoch": 0.8509589370528468,
      "flos": 16433405654400.0,
      "grad_norm": 1.5811002385968285,
      "language_loss": 0.77661842,
      "learning_rate": 2.2839227979432008e-07,
      "loss": 0.79816759,
      "num_input_tokens_seen": 153090215,
      "step": 7077,
      "time_per_iteration": 2.4639763832092285
    },
    {
      "auxiliary_loss_clip": 0.01137051,
      "auxiliary_loss_mlp": 0.0102733,
      "balance_loss_clip": 1.04286647,
      "balance_loss_mlp": 1.02009082,
      "epoch": 0.8510791799434858,
      "flos": 18259714713600.0,
      "grad_norm": 1.8369294047695677,
      "language_loss": 0.8523283,
      "learning_rate": 2.2803092473266373e-07,
      "loss": 0.87397212,
      "num_input_tokens_seen": 153107740,
      "step": 7078,
      "time_per_iteration": 2.4514496326446533
    },
    {
      "auxiliary_loss_clip": 0.01167681,
      "auxiliary_loss_mlp": 0.01027965,
      "balance_loss_clip": 1.04919112,
      "balance_loss_mlp": 1.0211879,
      "epoch": 0.851199422834125,
      "flos": 23441372933760.0,
      "grad_norm": 2.2591622548663004,
      "language_loss": 0.87185091,
      "learning_rate": 2.2766983847730724e-07,
      "loss": 0.89380741,
      "num_input_tokens_seen": 153127410,
      "step": 7079,
      "time_per_iteration": 2.4115891456604004
    },
    {
      "auxiliary_loss_clip": 0.01130762,
      "auxiliary_loss_mlp": 0.01027365,
      "balance_loss_clip": 1.04164219,
      "balance_loss_mlp": 1.02004004,
      "epoch": 0.851319665724764,
      "flos": 16289404030080.0,
      "grad_norm": 2.1513353759889227,
      "language_loss": 0.66395998,
      "learning_rate": 2.2730902108302663e-07,
      "loss": 0.68554127,
      "num_input_tokens_seen": 153144325,
      "step": 7080,
      "time_per_iteration": 2.4805057048797607
    },
    {
      "auxiliary_loss_clip": 0.011282,
      "auxiliary_loss_mlp": 0.01024554,
      "balance_loss_clip": 1.04047132,
      "balance_loss_mlp": 1.01704419,
      "epoch": 0.8514399086154031,
      "flos": 18989347680000.0,
      "grad_norm": 1.6018961236200482,
      "language_loss": 0.68767834,
      "learning_rate": 2.269484726045583e-07,
      "loss": 0.70920593,
      "num_input_tokens_seen": 153163240,
      "step": 7081,
      "time_per_iteration": 3.2784430980682373
    },
    {
      "auxiliary_loss_clip": 0.01126317,
      "auxiliary_loss_mlp": 0.01029128,
      "balance_loss_clip": 1.04383409,
      "balance_loss_mlp": 1.02173698,
      "epoch": 0.8515601515060423,
      "flos": 24571194301440.0,
      "grad_norm": 1.706802448039936,
      "language_loss": 0.79028738,
      "learning_rate": 2.2658819309659672e-07,
      "loss": 0.81184185,
      "num_input_tokens_seen": 153183440,
      "step": 7082,
      "time_per_iteration": 3.2958412170410156
    },
    {
      "auxiliary_loss_clip": 0.01135462,
      "auxiliary_loss_mlp": 0.01020593,
      "balance_loss_clip": 1.04651475,
      "balance_loss_mlp": 1.01406932,
      "epoch": 0.8516803943966813,
      "flos": 19529443555200.0,
      "grad_norm": 2.0187326952820426,
      "language_loss": 0.84839016,
      "learning_rate": 2.2622818261379706e-07,
      "loss": 0.86995071,
      "num_input_tokens_seen": 153200460,
      "step": 7083,
      "time_per_iteration": 2.4643161296844482
    },
    {
      "auxiliary_loss_clip": 0.01134213,
      "auxiliary_loss_mlp": 0.01025148,
      "balance_loss_clip": 1.04279351,
      "balance_loss_mlp": 1.01789105,
      "epoch": 0.8518006372873204,
      "flos": 20265792364800.0,
      "grad_norm": 1.9362268411796046,
      "language_loss": 0.7506994,
      "learning_rate": 2.2586844121077142e-07,
      "loss": 0.77229297,
      "num_input_tokens_seen": 153218970,
      "step": 7084,
      "time_per_iteration": 3.321122407913208
    },
    {
      "auxiliary_loss_clip": 0.01109514,
      "auxiliary_loss_mlp": 0.01030037,
      "balance_loss_clip": 1.03928852,
      "balance_loss_mlp": 1.02259803,
      "epoch": 0.8519208801779595,
      "flos": 24133227770880.0,
      "grad_norm": 1.823985184558145,
      "language_loss": 0.72233599,
      "learning_rate": 2.2550896894209215e-07,
      "loss": 0.7437315,
      "num_input_tokens_seen": 153238485,
      "step": 7085,
      "time_per_iteration": 2.5603764057159424
    },
    {
      "auxiliary_loss_clip": 0.01012651,
      "auxiliary_loss_mlp": 0.01000803,
      "balance_loss_clip": 1.01078892,
      "balance_loss_mlp": 0.99989742,
      "epoch": 0.8520411230685986,
      "flos": 63035223252480.0,
      "grad_norm": 0.6806603840275841,
      "language_loss": 0.56611848,
      "learning_rate": 2.2514976586229184e-07,
      "loss": 0.58625305,
      "num_input_tokens_seen": 153306430,
      "step": 7086,
      "time_per_iteration": 3.2596099376678467
    },
    {
      "auxiliary_loss_clip": 0.01052337,
      "auxiliary_loss_mlp": 0.01001103,
      "balance_loss_clip": 1.00837684,
      "balance_loss_mlp": 1.00020313,
      "epoch": 0.8521613659592376,
      "flos": 65836865283840.0,
      "grad_norm": 0.7580509956980961,
      "language_loss": 0.5481354,
      "learning_rate": 2.247908320258609e-07,
      "loss": 0.5686698,
      "num_input_tokens_seen": 153366520,
      "step": 7087,
      "time_per_iteration": 3.0005438327789307
    },
    {
      "auxiliary_loss_clip": 0.01101327,
      "auxiliary_loss_mlp": 0.01024358,
      "balance_loss_clip": 1.04148126,
      "balance_loss_mlp": 1.01699662,
      "epoch": 0.8522816088498768,
      "flos": 23112323418240.0,
      "grad_norm": 2.148926769257755,
      "language_loss": 0.7967459,
      "learning_rate": 2.2443216748724914e-07,
      "loss": 0.81800264,
      "num_input_tokens_seen": 153387230,
      "step": 7088,
      "time_per_iteration": 2.5657765865325928
    },
    {
      "auxiliary_loss_clip": 0.01154512,
      "auxiliary_loss_mlp": 0.00761629,
      "balance_loss_clip": 1.04673874,
      "balance_loss_mlp": 1.00036407,
      "epoch": 0.8524018517405159,
      "flos": 31758140073600.0,
      "grad_norm": 2.414209224319873,
      "language_loss": 0.74304509,
      "learning_rate": 2.2407377230086588e-07,
      "loss": 0.76220644,
      "num_input_tokens_seen": 153409585,
      "step": 7089,
      "time_per_iteration": 3.2796361446380615
    },
    {
      "auxiliary_loss_clip": 0.01121988,
      "auxiliary_loss_mlp": 0.01020596,
      "balance_loss_clip": 1.04518378,
      "balance_loss_mlp": 1.01373291,
      "epoch": 0.8525220946311549,
      "flos": 18690318956160.0,
      "grad_norm": 1.8693936582278239,
      "language_loss": 0.83429128,
      "learning_rate": 2.23715646521079e-07,
      "loss": 0.85571712,
      "num_input_tokens_seen": 153427105,
      "step": 7090,
      "time_per_iteration": 2.5232646465301514
    },
    {
      "auxiliary_loss_clip": 0.01153943,
      "auxiliary_loss_mlp": 0.00761992,
      "balance_loss_clip": 1.04529393,
      "balance_loss_mlp": 1.00046611,
      "epoch": 0.852642337521794,
      "flos": 21793216354560.0,
      "grad_norm": 1.9545998235180504,
      "language_loss": 0.83970618,
      "learning_rate": 2.2335779020221724e-07,
      "loss": 0.85886556,
      "num_input_tokens_seen": 153443725,
      "step": 7091,
      "time_per_iteration": 2.4448933601379395
    },
    {
      "auxiliary_loss_clip": 0.0105643,
      "auxiliary_loss_mlp": 0.01002323,
      "balance_loss_clip": 1.01856983,
      "balance_loss_mlp": 1.00123203,
      "epoch": 0.8527625804124331,
      "flos": 69040132260480.0,
      "grad_norm": 0.7988800743528195,
      "language_loss": 0.56406486,
      "learning_rate": 2.2300020339856497e-07,
      "loss": 0.5846523,
      "num_input_tokens_seen": 153506410,
      "step": 7092,
      "time_per_iteration": 3.0925779342651367
    },
    {
      "auxiliary_loss_clip": 0.01132005,
      "auxiliary_loss_mlp": 0.01022019,
      "balance_loss_clip": 1.04318929,
      "balance_loss_mlp": 1.01496196,
      "epoch": 0.8528828233030722,
      "flos": 26979399688320.0,
      "grad_norm": 2.170279768260591,
      "language_loss": 0.78223598,
      "learning_rate": 2.2264288616436966e-07,
      "loss": 0.80377626,
      "num_input_tokens_seen": 153526665,
      "step": 7093,
      "time_per_iteration": 2.554919481277466
    },
    {
      "auxiliary_loss_clip": 0.01129065,
      "auxiliary_loss_mlp": 0.01025349,
      "balance_loss_clip": 1.04226148,
      "balance_loss_mlp": 1.01838422,
      "epoch": 0.8530030661937112,
      "flos": 17487598936320.0,
      "grad_norm": 2.3092569011024198,
      "language_loss": 0.72660351,
      "learning_rate": 2.222858385538351e-07,
      "loss": 0.74814761,
      "num_input_tokens_seen": 153543465,
      "step": 7094,
      "time_per_iteration": 2.4423117637634277
    },
    {
      "auxiliary_loss_clip": 0.01146373,
      "auxiliary_loss_mlp": 0.01028126,
      "balance_loss_clip": 1.04410696,
      "balance_loss_mlp": 1.02111912,
      "epoch": 0.8531233090843504,
      "flos": 22160798184960.0,
      "grad_norm": 3.44874137753195,
      "language_loss": 0.68056417,
      "learning_rate": 2.2192906062112527e-07,
      "loss": 0.70230913,
      "num_input_tokens_seen": 153563340,
      "step": 7095,
      "time_per_iteration": 2.4753835201263428
    },
    {
      "auxiliary_loss_clip": 0.01163591,
      "auxiliary_loss_mlp": 0.01021055,
      "balance_loss_clip": 1.04470897,
      "balance_loss_mlp": 1.01439142,
      "epoch": 0.8532435519749895,
      "flos": 37635388145280.0,
      "grad_norm": 1.584784203027402,
      "language_loss": 0.70633185,
      "learning_rate": 2.2157255242036377e-07,
      "loss": 0.72817832,
      "num_input_tokens_seen": 153587005,
      "step": 7096,
      "time_per_iteration": 2.5972514152526855
    },
    {
      "auxiliary_loss_clip": 0.01118378,
      "auxiliary_loss_mlp": 0.01025146,
      "balance_loss_clip": 1.04140472,
      "balance_loss_mlp": 1.01800263,
      "epoch": 0.8533637948656285,
      "flos": 21398163598080.0,
      "grad_norm": 2.844418129219226,
      "language_loss": 0.74186468,
      "learning_rate": 2.2121631400563135e-07,
      "loss": 0.76329982,
      "num_input_tokens_seen": 153606835,
      "step": 7097,
      "time_per_iteration": 2.513479709625244
    },
    {
      "auxiliary_loss_clip": 0.01049414,
      "auxiliary_loss_mlp": 0.01001437,
      "balance_loss_clip": 1.01043344,
      "balance_loss_mlp": 1.00050151,
      "epoch": 0.8534840377562677,
      "flos": 53345122490880.0,
      "grad_norm": 0.7640189036336291,
      "language_loss": 0.53004527,
      "learning_rate": 2.208603454309701e-07,
      "loss": 0.55055374,
      "num_input_tokens_seen": 153664925,
      "step": 7098,
      "time_per_iteration": 3.006751775741577
    },
    {
      "auxiliary_loss_clip": 0.01109014,
      "auxiliary_loss_mlp": 0.01025397,
      "balance_loss_clip": 1.04181433,
      "balance_loss_mlp": 1.01755607,
      "epoch": 0.8536042806469067,
      "flos": 20814148368000.0,
      "grad_norm": 2.2556983047475723,
      "language_loss": 0.70660621,
      "learning_rate": 2.2050464675037994e-07,
      "loss": 0.72795033,
      "num_input_tokens_seen": 153683550,
      "step": 7099,
      "time_per_iteration": 2.573735237121582
    },
    {
      "auxiliary_loss_clip": 0.01136092,
      "auxiliary_loss_mlp": 0.01026461,
      "balance_loss_clip": 1.0445708,
      "balance_loss_mlp": 1.01921868,
      "epoch": 0.8537245235375458,
      "flos": 24681368292480.0,
      "grad_norm": 2.2744954366880434,
      "language_loss": 0.72964811,
      "learning_rate": 2.2014921801782016e-07,
      "loss": 0.75127357,
      "num_input_tokens_seen": 153703040,
      "step": 7100,
      "time_per_iteration": 2.505391836166382
    },
    {
      "auxiliary_loss_clip": 0.01136182,
      "auxiliary_loss_mlp": 0.0102226,
      "balance_loss_clip": 1.03957009,
      "balance_loss_mlp": 1.01538169,
      "epoch": 0.853844766428185,
      "flos": 24384817607040.0,
      "grad_norm": 1.8797281719053744,
      "language_loss": 0.74105167,
      "learning_rate": 2.1979405928720872e-07,
      "loss": 0.76263613,
      "num_input_tokens_seen": 153722695,
      "step": 7101,
      "time_per_iteration": 2.5264322757720947
    },
    {
      "auxiliary_loss_clip": 0.01139696,
      "auxiliary_loss_mlp": 0.01021695,
      "balance_loss_clip": 1.04485679,
      "balance_loss_mlp": 1.01508152,
      "epoch": 0.853965009318824,
      "flos": 20955707867520.0,
      "grad_norm": 1.5139654686054798,
      "language_loss": 0.79297006,
      "learning_rate": 2.1943917061242257e-07,
      "loss": 0.81458396,
      "num_input_tokens_seen": 153742550,
      "step": 7102,
      "time_per_iteration": 2.4963033199310303
    },
    {
      "auxiliary_loss_clip": 0.01158273,
      "auxiliary_loss_mlp": 0.00762013,
      "balance_loss_clip": 1.04658568,
      "balance_loss_mlp": 1.00049496,
      "epoch": 0.8540852522094631,
      "flos": 24201816791040.0,
      "grad_norm": 1.7341689569974803,
      "language_loss": 0.66479659,
      "learning_rate": 2.1908455204729903e-07,
      "loss": 0.68399942,
      "num_input_tokens_seen": 153761700,
      "step": 7103,
      "time_per_iteration": 2.471831798553467
    },
    {
      "auxiliary_loss_clip": 0.01136763,
      "auxiliary_loss_mlp": 0.0102547,
      "balance_loss_clip": 1.04289103,
      "balance_loss_mlp": 1.01815653,
      "epoch": 0.8542054951001022,
      "flos": 25082921410560.0,
      "grad_norm": 2.693554586123264,
      "language_loss": 0.78179169,
      "learning_rate": 2.1873020364563265e-07,
      "loss": 0.80341399,
      "num_input_tokens_seen": 153780765,
      "step": 7104,
      "time_per_iteration": 2.512988567352295
    },
    {
      "auxiliary_loss_clip": 0.01146736,
      "auxiliary_loss_mlp": 0.0102334,
      "balance_loss_clip": 1.04565895,
      "balance_loss_mlp": 1.01646447,
      "epoch": 0.8543257379907413,
      "flos": 24316551809280.0,
      "grad_norm": 24.71493465391521,
      "language_loss": 0.75604618,
      "learning_rate": 2.183761254611789e-07,
      "loss": 0.77774698,
      "num_input_tokens_seen": 153801090,
      "step": 7105,
      "time_per_iteration": 2.463827610015869
    },
    {
      "auxiliary_loss_clip": 0.01149472,
      "auxiliary_loss_mlp": 0.01027142,
      "balance_loss_clip": 1.04569709,
      "balance_loss_mlp": 1.02007318,
      "epoch": 0.8544459808813804,
      "flos": 55286630467200.0,
      "grad_norm": 2.015847224516383,
      "language_loss": 0.70016515,
      "learning_rate": 2.1802231754764987e-07,
      "loss": 0.72193128,
      "num_input_tokens_seen": 153826530,
      "step": 7106,
      "time_per_iteration": 2.7713136672973633
    },
    {
      "auxiliary_loss_clip": 0.01136835,
      "auxiliary_loss_mlp": 0.01027235,
      "balance_loss_clip": 1.04165757,
      "balance_loss_mlp": 1.01954257,
      "epoch": 0.8545662237720195,
      "flos": 25776248705280.0,
      "grad_norm": 1.8350454229343163,
      "language_loss": 0.76465547,
      "learning_rate": 2.17668779958718e-07,
      "loss": 0.78629619,
      "num_input_tokens_seen": 153849110,
      "step": 7107,
      "time_per_iteration": 2.527698516845703
    },
    {
      "auxiliary_loss_clip": 0.01164663,
      "auxiliary_loss_mlp": 0.01023817,
      "balance_loss_clip": 1.04628968,
      "balance_loss_mlp": 1.01670957,
      "epoch": 0.8546864666626586,
      "flos": 11108320427520.0,
      "grad_norm": 2.8153978998934415,
      "language_loss": 0.80400735,
      "learning_rate": 2.1731551274801553e-07,
      "loss": 0.82589221,
      "num_input_tokens_seen": 153865550,
      "step": 7108,
      "time_per_iteration": 3.35726261138916
    },
    {
      "auxiliary_loss_clip": 0.01140238,
      "auxiliary_loss_mlp": 0.01024911,
      "balance_loss_clip": 1.04658413,
      "balance_loss_mlp": 1.01740098,
      "epoch": 0.8548067095532976,
      "flos": 25520169669120.0,
      "grad_norm": 2.9064952533683606,
      "language_loss": 0.61815995,
      "learning_rate": 2.169625159691324e-07,
      "loss": 0.63981146,
      "num_input_tokens_seen": 153885425,
      "step": 7109,
      "time_per_iteration": 3.27755069732666
    },
    {
      "auxiliary_loss_clip": 0.01118058,
      "auxiliary_loss_mlp": 0.01024274,
      "balance_loss_clip": 1.04166985,
      "balance_loss_mlp": 1.01730919,
      "epoch": 0.8549269524439368,
      "flos": 24717853532160.0,
      "grad_norm": 2.3224754343060345,
      "language_loss": 0.7424742,
      "learning_rate": 2.1660978967561784e-07,
      "loss": 0.76389754,
      "num_input_tokens_seen": 153904760,
      "step": 7110,
      "time_per_iteration": 3.447495698928833
    },
    {
      "auxiliary_loss_clip": 0.01163556,
      "auxiliary_loss_mlp": 0.0102038,
      "balance_loss_clip": 1.04498148,
      "balance_loss_mlp": 1.01380217,
      "epoch": 0.8550471953345758,
      "flos": 19825599191040.0,
      "grad_norm": 2.7017082195453903,
      "language_loss": 0.78816509,
      "learning_rate": 2.1625733392098035e-07,
      "loss": 0.81000435,
      "num_input_tokens_seen": 153920370,
      "step": 7111,
      "time_per_iteration": 2.401700735092163
    },
    {
      "auxiliary_loss_clip": 0.01164089,
      "auxiliary_loss_mlp": 0.01023094,
      "balance_loss_clip": 1.04617691,
      "balance_loss_mlp": 1.01623917,
      "epoch": 0.8551674382252149,
      "flos": 22820441500800.0,
      "grad_norm": 1.583238807419085,
      "language_loss": 0.79560065,
      "learning_rate": 2.159051487586867e-07,
      "loss": 0.81747246,
      "num_input_tokens_seen": 153940500,
      "step": 7112,
      "time_per_iteration": 2.4194259643554688
    },
    {
      "auxiliary_loss_clip": 0.01141066,
      "auxiliary_loss_mlp": 0.01029279,
      "balance_loss_clip": 1.04665303,
      "balance_loss_mlp": 1.02133965,
      "epoch": 0.8552876811158541,
      "flos": 20631255292800.0,
      "grad_norm": 2.010515579972332,
      "language_loss": 0.72758937,
      "learning_rate": 2.155532342421642e-07,
      "loss": 0.74929285,
      "num_input_tokens_seen": 153958500,
      "step": 7113,
      "time_per_iteration": 2.4786014556884766
    },
    {
      "auxiliary_loss_clip": 0.01154349,
      "auxiliary_loss_mlp": 0.01028762,
      "balance_loss_clip": 1.04639828,
      "balance_loss_mlp": 1.02130246,
      "epoch": 0.8554079240064931,
      "flos": 23112359331840.0,
      "grad_norm": 1.7578985421022326,
      "language_loss": 0.78201163,
      "learning_rate": 2.1520159042479636e-07,
      "loss": 0.80384278,
      "num_input_tokens_seen": 153976790,
      "step": 7114,
      "time_per_iteration": 2.466628074645996
    },
    {
      "auxiliary_loss_clip": 0.01150476,
      "auxiliary_loss_mlp": 0.01024776,
      "balance_loss_clip": 1.04641366,
      "balance_loss_mlp": 1.01762962,
      "epoch": 0.8555281668971322,
      "flos": 22128047959680.0,
      "grad_norm": 2.1415884181085025,
      "language_loss": 0.71078467,
      "learning_rate": 2.148502173599287e-07,
      "loss": 0.73253727,
      "num_input_tokens_seen": 153994930,
      "step": 7115,
      "time_per_iteration": 2.4356510639190674
    },
    {
      "auxiliary_loss_clip": 0.01130847,
      "auxiliary_loss_mlp": 0.01019869,
      "balance_loss_clip": 1.04307628,
      "balance_loss_mlp": 1.01238227,
      "epoch": 0.8556484097877713,
      "flos": 31139040234240.0,
      "grad_norm": 1.7688118373270425,
      "language_loss": 0.66214544,
      "learning_rate": 2.1449911510086372e-07,
      "loss": 0.68365264,
      "num_input_tokens_seen": 154014400,
      "step": 7116,
      "time_per_iteration": 3.298696517944336
    },
    {
      "auxiliary_loss_clip": 0.01147216,
      "auxiliary_loss_mlp": 0.0102453,
      "balance_loss_clip": 1.04423118,
      "balance_loss_mlp": 1.01752079,
      "epoch": 0.8557686526784104,
      "flos": 24316551809280.0,
      "grad_norm": 1.752646639779243,
      "language_loss": 0.76703835,
      "learning_rate": 2.141482837008628e-07,
      "loss": 0.78875583,
      "num_input_tokens_seen": 154034940,
      "step": 7117,
      "time_per_iteration": 2.4745707511901855
    },
    {
      "auxiliary_loss_clip": 0.01141461,
      "auxiliary_loss_mlp": 0.01028354,
      "balance_loss_clip": 1.04265523,
      "balance_loss_mlp": 1.02091861,
      "epoch": 0.8558888955690495,
      "flos": 17712723427200.0,
      "grad_norm": 1.915362085385774,
      "language_loss": 0.71950197,
      "learning_rate": 2.1379772321314826e-07,
      "loss": 0.74120009,
      "num_input_tokens_seen": 154052985,
      "step": 7118,
      "time_per_iteration": 2.4002320766448975
    },
    {
      "auxiliary_loss_clip": 0.01088184,
      "auxiliary_loss_mlp": 0.01026403,
      "balance_loss_clip": 1.04164338,
      "balance_loss_mlp": 1.01889861,
      "epoch": 0.8560091384596886,
      "flos": 19171702051200.0,
      "grad_norm": 1.9415124180322658,
      "language_loss": 0.81642765,
      "learning_rate": 2.1344743369089802e-07,
      "loss": 0.83757353,
      "num_input_tokens_seen": 154068765,
      "step": 7119,
      "time_per_iteration": 2.5673701763153076
    },
    {
      "auxiliary_loss_clip": 0.01138132,
      "auxiliary_loss_mlp": 0.01024128,
      "balance_loss_clip": 1.0464741,
      "balance_loss_mlp": 1.01710999,
      "epoch": 0.8561293813503277,
      "flos": 23914855036800.0,
      "grad_norm": 1.6660649511231194,
      "language_loss": 0.81984925,
      "learning_rate": 2.130974151872522e-07,
      "loss": 0.84147185,
      "num_input_tokens_seen": 154089100,
      "step": 7120,
      "time_per_iteration": 2.5017502307891846
    },
    {
      "auxiliary_loss_clip": 0.01127324,
      "auxiliary_loss_mlp": 0.01023549,
      "balance_loss_clip": 1.04591632,
      "balance_loss_mlp": 1.01660848,
      "epoch": 0.8562496242409667,
      "flos": 22529206028160.0,
      "grad_norm": 1.6972027605980051,
      "language_loss": 0.78405559,
      "learning_rate": 2.1274766775530773e-07,
      "loss": 0.80556428,
      "num_input_tokens_seen": 154108965,
      "step": 7121,
      "time_per_iteration": 2.5228111743927
    },
    {
      "auxiliary_loss_clip": 0.01166229,
      "auxiliary_loss_mlp": 0.01021753,
      "balance_loss_clip": 1.04575682,
      "balance_loss_mlp": 1.01455534,
      "epoch": 0.8563698671316058,
      "flos": 14712745472640.0,
      "grad_norm": 2.262499353113867,
      "language_loss": 0.79953611,
      "learning_rate": 2.1239819144812077e-07,
      "loss": 0.8214159,
      "num_input_tokens_seen": 154123425,
      "step": 7122,
      "time_per_iteration": 2.382805109024048
    },
    {
      "auxiliary_loss_clip": 0.01115698,
      "auxiliary_loss_mlp": 0.01024639,
      "balance_loss_clip": 1.03922987,
      "balance_loss_mlp": 1.01757264,
      "epoch": 0.856490110022245,
      "flos": 39167768211840.0,
      "grad_norm": 1.7684929227127204,
      "language_loss": 0.70028448,
      "learning_rate": 2.1204898631870716e-07,
      "loss": 0.72168791,
      "num_input_tokens_seen": 154148315,
      "step": 7123,
      "time_per_iteration": 2.654613733291626
    },
    {
      "auxiliary_loss_clip": 0.01137594,
      "auxiliary_loss_mlp": 0.01022538,
      "balance_loss_clip": 1.04618323,
      "balance_loss_mlp": 1.01592505,
      "epoch": 0.856610352912884,
      "flos": 29059345658880.0,
      "grad_norm": 1.713443754030315,
      "language_loss": 0.75897026,
      "learning_rate": 2.1170005242004006e-07,
      "loss": 0.78057152,
      "num_input_tokens_seen": 154169665,
      "step": 7124,
      "time_per_iteration": 2.5396406650543213
    },
    {
      "auxiliary_loss_clip": 0.0114077,
      "auxiliary_loss_mlp": 0.01020307,
      "balance_loss_clip": 1.04329169,
      "balance_loss_mlp": 1.01335979,
      "epoch": 0.8567305958035231,
      "flos": 23878333883520.0,
      "grad_norm": 1.6401032456238034,
      "language_loss": 0.77739376,
      "learning_rate": 2.1135138980505384e-07,
      "loss": 0.7990045,
      "num_input_tokens_seen": 154190335,
      "step": 7125,
      "time_per_iteration": 2.5006911754608154
    },
    {
      "auxiliary_loss_clip": 0.01133988,
      "auxiliary_loss_mlp": 0.01021823,
      "balance_loss_clip": 1.04508853,
      "balance_loss_mlp": 1.01464081,
      "epoch": 0.8568508386941622,
      "flos": 22200120599040.0,
      "grad_norm": 1.6923256701763048,
      "language_loss": 0.7226361,
      "learning_rate": 2.110029985266395e-07,
      "loss": 0.74419427,
      "num_input_tokens_seen": 154210040,
      "step": 7126,
      "time_per_iteration": 2.481630563735962
    },
    {
      "auxiliary_loss_clip": 0.01139962,
      "auxiliary_loss_mlp": 0.01023347,
      "balance_loss_clip": 1.04299104,
      "balance_loss_mlp": 1.01628399,
      "epoch": 0.8569710815848013,
      "flos": 17307507121920.0,
      "grad_norm": 1.717440676243043,
      "language_loss": 0.73825502,
      "learning_rate": 2.1065487863764787e-07,
      "loss": 0.75988805,
      "num_input_tokens_seen": 154228385,
      "step": 7127,
      "time_per_iteration": 2.4848124980926514
    },
    {
      "auxiliary_loss_clip": 0.01099867,
      "auxiliary_loss_mlp": 0.01022333,
      "balance_loss_clip": 1.03663266,
      "balance_loss_mlp": 1.01476645,
      "epoch": 0.8570913244754403,
      "flos": 23732285184000.0,
      "grad_norm": 1.5067486900678617,
      "language_loss": 0.85707498,
      "learning_rate": 2.1030703019088846e-07,
      "loss": 0.87829697,
      "num_input_tokens_seen": 154249015,
      "step": 7128,
      "time_per_iteration": 2.5588924884796143
    },
    {
      "auxiliary_loss_clip": 0.01144073,
      "auxiliary_loss_mlp": 0.01022251,
      "balance_loss_clip": 1.04383111,
      "balance_loss_mlp": 1.01531649,
      "epoch": 0.8572115673660795,
      "flos": 20048748433920.0,
      "grad_norm": 1.8404305668588965,
      "language_loss": 0.71009517,
      "learning_rate": 2.099594532391291e-07,
      "loss": 0.73175842,
      "num_input_tokens_seen": 154267700,
      "step": 7129,
      "time_per_iteration": 2.447481870651245
    },
    {
      "auxiliary_loss_clip": 0.01141461,
      "auxiliary_loss_mlp": 0.01022812,
      "balance_loss_clip": 1.04364192,
      "balance_loss_mlp": 1.01587701,
      "epoch": 0.8573318102567186,
      "flos": 27160389342720.0,
      "grad_norm": 2.135642708660653,
      "language_loss": 0.79491234,
      "learning_rate": 2.0961214783509806e-07,
      "loss": 0.81655508,
      "num_input_tokens_seen": 154290580,
      "step": 7130,
      "time_per_iteration": 2.4881370067596436
    },
    {
      "auxiliary_loss_clip": 0.01141851,
      "auxiliary_loss_mlp": 0.01022728,
      "balance_loss_clip": 1.04389179,
      "balance_loss_mlp": 1.01544392,
      "epoch": 0.8574520531473576,
      "flos": 24936585402240.0,
      "grad_norm": 1.672699755340515,
      "language_loss": 0.74431634,
      "learning_rate": 2.0926511403148051e-07,
      "loss": 0.76596206,
      "num_input_tokens_seen": 154309545,
      "step": 7131,
      "time_per_iteration": 2.5279552936553955
    },
    {
      "auxiliary_loss_clip": 0.01131397,
      "auxiliary_loss_mlp": 0.01028293,
      "balance_loss_clip": 1.04535735,
      "balance_loss_mlp": 1.02132261,
      "epoch": 0.8575722960379968,
      "flos": 18771154513920.0,
      "grad_norm": 2.010697116662194,
      "language_loss": 0.75646871,
      "learning_rate": 2.0891835188092143e-07,
      "loss": 0.77806562,
      "num_input_tokens_seen": 154326545,
      "step": 7132,
      "time_per_iteration": 2.485581398010254
    },
    {
      "auxiliary_loss_clip": 0.01130087,
      "auxiliary_loss_mlp": 0.0102527,
      "balance_loss_clip": 1.04269528,
      "balance_loss_mlp": 1.01760507,
      "epoch": 0.8576925389286358,
      "flos": 22200300167040.0,
      "grad_norm": 1.785602718310032,
      "language_loss": 0.81554157,
      "learning_rate": 2.0857186143602434e-07,
      "loss": 0.83709514,
      "num_input_tokens_seen": 154345190,
      "step": 7133,
      "time_per_iteration": 2.5050699710845947
    },
    {
      "auxiliary_loss_clip": 0.01112251,
      "auxiliary_loss_mlp": 0.01029203,
      "balance_loss_clip": 1.03923154,
      "balance_loss_mlp": 1.02162695,
      "epoch": 0.8578127818192749,
      "flos": 22894345733760.0,
      "grad_norm": 2.153421500470876,
      "language_loss": 0.67821896,
      "learning_rate": 2.0822564274935094e-07,
      "loss": 0.69963348,
      "num_input_tokens_seen": 154364615,
      "step": 7134,
      "time_per_iteration": 3.3863205909729004
    },
    {
      "auxiliary_loss_clip": 0.01138747,
      "auxiliary_loss_mlp": 0.01024078,
      "balance_loss_clip": 1.04717696,
      "balance_loss_mlp": 1.01645803,
      "epoch": 0.8579330247099141,
      "flos": 34824839541120.0,
      "grad_norm": 1.6992535868343805,
      "language_loss": 0.67037719,
      "learning_rate": 2.078796958734239e-07,
      "loss": 0.69200546,
      "num_input_tokens_seen": 154387335,
      "step": 7135,
      "time_per_iteration": 3.3488194942474365
    },
    {
      "auxiliary_loss_clip": 0.01150362,
      "auxiliary_loss_mlp": 0.01025291,
      "balance_loss_clip": 1.04596567,
      "balance_loss_mlp": 1.01869583,
      "epoch": 0.8580532676005531,
      "flos": 19755681367680.0,
      "grad_norm": 1.810673972587797,
      "language_loss": 0.75111914,
      "learning_rate": 2.0753402086072124e-07,
      "loss": 0.77287567,
      "num_input_tokens_seen": 154405965,
      "step": 7136,
      "time_per_iteration": 2.4711809158325195
    },
    {
      "auxiliary_loss_clip": 0.01092402,
      "auxiliary_loss_mlp": 0.01028217,
      "balance_loss_clip": 1.04108095,
      "balance_loss_mlp": 1.02101052,
      "epoch": 0.8581735104911922,
      "flos": 22739318634240.0,
      "grad_norm": 3.4370197576638586,
      "language_loss": 0.75543499,
      "learning_rate": 2.071886177636828e-07,
      "loss": 0.77664119,
      "num_input_tokens_seen": 154422750,
      "step": 7137,
      "time_per_iteration": 3.4663712978363037
    },
    {
      "auxiliary_loss_clip": 0.01148035,
      "auxiliary_loss_mlp": 0.01022027,
      "balance_loss_clip": 1.04576111,
      "balance_loss_mlp": 1.01519358,
      "epoch": 0.8582937533818313,
      "flos": 23149131880320.0,
      "grad_norm": 1.8716892172047015,
      "language_loss": 0.83404833,
      "learning_rate": 2.0684348663470575e-07,
      "loss": 0.85574889,
      "num_input_tokens_seen": 154442930,
      "step": 7138,
      "time_per_iteration": 2.4559977054595947
    },
    {
      "auxiliary_loss_clip": 0.01133601,
      "auxiliary_loss_mlp": 0.01025404,
      "balance_loss_clip": 1.03944218,
      "balance_loss_mlp": 1.01821876,
      "epoch": 0.8584139962724704,
      "flos": 19498668577920.0,
      "grad_norm": 1.9761921902968498,
      "language_loss": 0.61640847,
      "learning_rate": 2.0649862752614555e-07,
      "loss": 0.63799858,
      "num_input_tokens_seen": 154461640,
      "step": 7139,
      "time_per_iteration": 2.475545644760132
    },
    {
      "auxiliary_loss_clip": 0.01042598,
      "auxiliary_loss_mlp": 0.01000632,
      "balance_loss_clip": 1.00727153,
      "balance_loss_mlp": 0.9997142,
      "epoch": 0.8585342391631094,
      "flos": 71276577788160.0,
      "grad_norm": 0.7948786302469087,
      "language_loss": 0.57112753,
      "learning_rate": 2.0615404049031838e-07,
      "loss": 0.59155977,
      "num_input_tokens_seen": 154518610,
      "step": 7140,
      "time_per_iteration": 3.0986568927764893
    },
    {
      "auxiliary_loss_clip": 0.01152329,
      "auxiliary_loss_mlp": 0.01026267,
      "balance_loss_clip": 1.04647374,
      "balance_loss_mlp": 1.01862538,
      "epoch": 0.8586544820537486,
      "flos": 10815432929280.0,
      "grad_norm": 2.54544364075567,
      "language_loss": 0.77928972,
      "learning_rate": 2.0580972557949616e-07,
      "loss": 0.8010757,
      "num_input_tokens_seen": 154533700,
      "step": 7141,
      "time_per_iteration": 2.4081368446350098
    },
    {
      "auxiliary_loss_clip": 0.01053219,
      "auxiliary_loss_mlp": 0.01000548,
      "balance_loss_clip": 1.0076679,
      "balance_loss_mlp": 0.99956435,
      "epoch": 0.8587747249443877,
      "flos": 64811184422400.0,
      "grad_norm": 0.7943313617794318,
      "language_loss": 0.54253036,
      "learning_rate": 2.054656828459125e-07,
      "loss": 0.56306803,
      "num_input_tokens_seen": 154597810,
      "step": 7142,
      "time_per_iteration": 3.787107229232788
    },
    {
      "auxiliary_loss_clip": 0.01105922,
      "auxiliary_loss_mlp": 0.01025544,
      "balance_loss_clip": 1.04046488,
      "balance_loss_mlp": 1.01786041,
      "epoch": 0.8588949678350267,
      "flos": 26834607964800.0,
      "grad_norm": 1.792004687903429,
      "language_loss": 0.77316689,
      "learning_rate": 2.051219123417578e-07,
      "loss": 0.79448158,
      "num_input_tokens_seen": 154617870,
      "step": 7143,
      "time_per_iteration": 2.601811170578003
    },
    {
      "auxiliary_loss_clip": 0.01166115,
      "auxiliary_loss_mlp": 0.01024218,
      "balance_loss_clip": 1.04571009,
      "balance_loss_mlp": 1.01630831,
      "epoch": 0.8590152107256659,
      "flos": 26104256726400.0,
      "grad_norm": 2.120210973560474,
      "language_loss": 0.59869969,
      "learning_rate": 2.0477841411918196e-07,
      "loss": 0.62060302,
      "num_input_tokens_seen": 154637395,
      "step": 7144,
      "time_per_iteration": 2.4668514728546143
    },
    {
      "auxiliary_loss_clip": 0.01144708,
      "auxiliary_loss_mlp": 0.01022133,
      "balance_loss_clip": 1.04402995,
      "balance_loss_mlp": 1.01509368,
      "epoch": 0.859135453616305,
      "flos": 26140885620480.0,
      "grad_norm": 1.8277466197563197,
      "language_loss": 0.75026977,
      "learning_rate": 2.0443518823029326e-07,
      "loss": 0.7719382,
      "num_input_tokens_seen": 154657935,
      "step": 7145,
      "time_per_iteration": 2.522007465362549
    },
    {
      "auxiliary_loss_clip": 0.01117454,
      "auxiliary_loss_mlp": 0.01027138,
      "balance_loss_clip": 1.04153407,
      "balance_loss_mlp": 1.0197289,
      "epoch": 0.859255696506944,
      "flos": 12969319046400.0,
      "grad_norm": 2.0723992810253353,
      "language_loss": 0.76774371,
      "learning_rate": 2.0409223472715854e-07,
      "loss": 0.78918958,
      "num_input_tokens_seen": 154675080,
      "step": 7146,
      "time_per_iteration": 2.488037586212158
    },
    {
      "auxiliary_loss_clip": 0.01124392,
      "auxiliary_loss_mlp": 0.00761389,
      "balance_loss_clip": 1.04345751,
      "balance_loss_mlp": 1.00045025,
      "epoch": 0.8593759393975832,
      "flos": 18475753063680.0,
      "grad_norm": 1.8579247381037083,
      "language_loss": 0.75002623,
      "learning_rate": 2.0374955366180434e-07,
      "loss": 0.76888394,
      "num_input_tokens_seen": 154692720,
      "step": 7147,
      "time_per_iteration": 2.5174143314361572
    },
    {
      "auxiliary_loss_clip": 0.01126578,
      "auxiliary_loss_mlp": 0.01021481,
      "balance_loss_clip": 1.04173207,
      "balance_loss_mlp": 1.01446247,
      "epoch": 0.8594961822882222,
      "flos": 22200156512640.0,
      "grad_norm": 1.6797549329276067,
      "language_loss": 0.72642541,
      "learning_rate": 2.034071450862147e-07,
      "loss": 0.74790597,
      "num_input_tokens_seen": 154710190,
      "step": 7148,
      "time_per_iteration": 2.5408291816711426
    },
    {
      "auxiliary_loss_clip": 0.01139129,
      "auxiliary_loss_mlp": 0.01023014,
      "balance_loss_clip": 1.04301882,
      "balance_loss_mlp": 1.015432,
      "epoch": 0.8596164251788613,
      "flos": 23294749616640.0,
      "grad_norm": 1.8180564129992038,
      "language_loss": 0.76850498,
      "learning_rate": 2.030650090523327e-07,
      "loss": 0.79012644,
      "num_input_tokens_seen": 154729380,
      "step": 7149,
      "time_per_iteration": 2.5200419425964355
    },
    {
      "auxiliary_loss_clip": 0.01120077,
      "auxiliary_loss_mlp": 0.01023761,
      "balance_loss_clip": 1.04114318,
      "balance_loss_mlp": 1.01627517,
      "epoch": 0.8597366680695004,
      "flos": 31649905416960.0,
      "grad_norm": 1.7057502844911314,
      "language_loss": 0.59345913,
      "learning_rate": 2.0272314561205995e-07,
      "loss": 0.61489749,
      "num_input_tokens_seen": 154749775,
      "step": 7150,
      "time_per_iteration": 2.5961899757385254
    },
    {
      "auxiliary_loss_clip": 0.01115746,
      "auxiliary_loss_mlp": 0.01019331,
      "balance_loss_clip": 1.03978515,
      "balance_loss_mlp": 1.01245558,
      "epoch": 0.8598569109601395,
      "flos": 21287738211840.0,
      "grad_norm": 2.2297157319955887,
      "language_loss": 0.72871649,
      "learning_rate": 2.023815548172567e-07,
      "loss": 0.75006723,
      "num_input_tokens_seen": 154769845,
      "step": 7151,
      "time_per_iteration": 2.5622732639312744
    },
    {
      "auxiliary_loss_clip": 0.01150406,
      "auxiliary_loss_mlp": 0.01023706,
      "balance_loss_clip": 1.04482269,
      "balance_loss_mlp": 1.01644015,
      "epoch": 0.8599771538507786,
      "flos": 25447809720960.0,
      "grad_norm": 1.6101062331573892,
      "language_loss": 0.6585561,
      "learning_rate": 2.0204023671974267e-07,
      "loss": 0.68029726,
      "num_input_tokens_seen": 154789230,
      "step": 7152,
      "time_per_iteration": 2.4884986877441406
    },
    {
      "auxiliary_loss_clip": 0.01143532,
      "auxiliary_loss_mlp": 0.01023821,
      "balance_loss_clip": 1.04296422,
      "balance_loss_mlp": 1.01630533,
      "epoch": 0.8600973967414177,
      "flos": 16723958768640.0,
      "grad_norm": 2.0342596018549015,
      "language_loss": 0.81103659,
      "learning_rate": 2.0169919137129532e-07,
      "loss": 0.83271015,
      "num_input_tokens_seen": 154807670,
      "step": 7153,
      "time_per_iteration": 2.436661720275879
    },
    {
      "auxiliary_loss_clip": 0.01151922,
      "auxiliary_loss_mlp": 0.01026136,
      "balance_loss_clip": 1.04653025,
      "balance_loss_mlp": 1.01864386,
      "epoch": 0.8602176396320568,
      "flos": 25227928615680.0,
      "grad_norm": 3.207644420848581,
      "language_loss": 0.70785671,
      "learning_rate": 2.013584188236508e-07,
      "loss": 0.72963732,
      "num_input_tokens_seen": 154825575,
      "step": 7154,
      "time_per_iteration": 2.4701919555664062
    },
    {
      "auxiliary_loss_clip": 0.01166557,
      "auxiliary_loss_mlp": 0.01024054,
      "balance_loss_clip": 1.04713452,
      "balance_loss_mlp": 1.0166328,
      "epoch": 0.8603378825226958,
      "flos": 20412236113920.0,
      "grad_norm": 1.8171104162711655,
      "language_loss": 0.7927283,
      "learning_rate": 2.0101791912850396e-07,
      "loss": 0.81463444,
      "num_input_tokens_seen": 154845115,
      "step": 7155,
      "time_per_iteration": 2.4896609783172607
    },
    {
      "auxiliary_loss_clip": 0.01136658,
      "auxiliary_loss_mlp": 0.01018456,
      "balance_loss_clip": 1.04521883,
      "balance_loss_mlp": 1.01131868,
      "epoch": 0.8604581254133349,
      "flos": 34930201109760.0,
      "grad_norm": 2.06850491880919,
      "language_loss": 0.64094603,
      "learning_rate": 2.006776923375082e-07,
      "loss": 0.66249716,
      "num_input_tokens_seen": 154866770,
      "step": 7156,
      "time_per_iteration": 2.5888118743896484
    },
    {
      "auxiliary_loss_clip": 0.01164964,
      "auxiliary_loss_mlp": 0.01020246,
      "balance_loss_clip": 1.04633975,
      "balance_loss_mlp": 1.01316762,
      "epoch": 0.860578368303974,
      "flos": 22596538072320.0,
      "grad_norm": 1.615905795665432,
      "language_loss": 0.71076477,
      "learning_rate": 2.003377385022764e-07,
      "loss": 0.7326169,
      "num_input_tokens_seen": 154885595,
      "step": 7157,
      "time_per_iteration": 2.442112445831299
    },
    {
      "auxiliary_loss_clip": 0.01136359,
      "auxiliary_loss_mlp": 0.0102615,
      "balance_loss_clip": 1.04194939,
      "balance_loss_mlp": 1.01952505,
      "epoch": 0.8606986111946131,
      "flos": 21324331192320.0,
      "grad_norm": 2.7986868252528163,
      "language_loss": 0.77131236,
      "learning_rate": 1.9999805767437826e-07,
      "loss": 0.7929374,
      "num_input_tokens_seen": 154904485,
      "step": 7158,
      "time_per_iteration": 2.4537353515625
    },
    {
      "auxiliary_loss_clip": 0.01130876,
      "auxiliary_loss_mlp": 0.01022342,
      "balance_loss_clip": 1.04168081,
      "balance_loss_mlp": 1.01549327,
      "epoch": 0.8608188540852522,
      "flos": 28877206769280.0,
      "grad_norm": 1.6227244744477838,
      "language_loss": 0.71838284,
      "learning_rate": 1.9965864990534386e-07,
      "loss": 0.73991501,
      "num_input_tokens_seen": 154925010,
      "step": 7159,
      "time_per_iteration": 2.5383620262145996
    },
    {
      "auxiliary_loss_clip": 0.01114805,
      "auxiliary_loss_mlp": 0.01022494,
      "balance_loss_clip": 1.03851306,
      "balance_loss_mlp": 1.01589882,
      "epoch": 0.8609390969758913,
      "flos": 29716187713920.0,
      "grad_norm": 1.7403032928859041,
      "language_loss": 0.77594042,
      "learning_rate": 1.9931951524666092e-07,
      "loss": 0.79731339,
      "num_input_tokens_seen": 154946100,
      "step": 7160,
      "time_per_iteration": 2.657500743865967
    },
    {
      "auxiliary_loss_clip": 0.01153075,
      "auxiliary_loss_mlp": 0.00761612,
      "balance_loss_clip": 1.04609132,
      "balance_loss_mlp": 1.00046122,
      "epoch": 0.8610593398665304,
      "flos": 21249349551360.0,
      "grad_norm": 1.6085437579817292,
      "language_loss": 0.81116873,
      "learning_rate": 1.9898065374977534e-07,
      "loss": 0.83031559,
      "num_input_tokens_seen": 154966305,
      "step": 7161,
      "time_per_iteration": 3.298006772994995
    },
    {
      "auxiliary_loss_clip": 0.01118675,
      "auxiliary_loss_mlp": 0.01018596,
      "balance_loss_clip": 1.04061389,
      "balance_loss_mlp": 1.01259685,
      "epoch": 0.8611795827571694,
      "flos": 14830102183680.0,
      "grad_norm": 2.334776530974076,
      "language_loss": 0.72395498,
      "learning_rate": 1.9864206546609342e-07,
      "loss": 0.74532765,
      "num_input_tokens_seen": 154985145,
      "step": 7162,
      "time_per_iteration": 2.4654252529144287
    },
    {
      "auxiliary_loss_clip": 0.01162847,
      "auxiliary_loss_mlp": 0.01021165,
      "balance_loss_clip": 1.04496765,
      "balance_loss_mlp": 1.01436126,
      "epoch": 0.8612998256478086,
      "flos": 24243258107520.0,
      "grad_norm": 1.7584485460681074,
      "language_loss": 0.84370685,
      "learning_rate": 1.983037504469771e-07,
      "loss": 0.86554694,
      "num_input_tokens_seen": 155003855,
      "step": 7163,
      "time_per_iteration": 2.439310312271118
    },
    {
      "auxiliary_loss_clip": 0.01154268,
      "auxiliary_loss_mlp": 0.01023178,
      "balance_loss_clip": 1.04794371,
      "balance_loss_mlp": 1.01624954,
      "epoch": 0.8614200685384477,
      "flos": 21252653602560.0,
      "grad_norm": 1.7550456793430336,
      "language_loss": 0.6641385,
      "learning_rate": 1.9796570874374984e-07,
      "loss": 0.68591297,
      "num_input_tokens_seen": 155023960,
      "step": 7164,
      "time_per_iteration": 3.2776477336883545
    },
    {
      "auxiliary_loss_clip": 0.01138227,
      "auxiliary_loss_mlp": 0.01019308,
      "balance_loss_clip": 1.04431295,
      "balance_loss_mlp": 1.01208401,
      "epoch": 0.8615403114290867,
      "flos": 20007738080640.0,
      "grad_norm": 1.9202443777637868,
      "language_loss": 0.77560365,
      "learning_rate": 1.976279404076917e-07,
      "loss": 0.79717898,
      "num_input_tokens_seen": 155043360,
      "step": 7165,
      "time_per_iteration": 2.489232301712036
    },
    {
      "auxiliary_loss_clip": 0.01121536,
      "auxiliary_loss_mlp": 0.01024967,
      "balance_loss_clip": 1.04301095,
      "balance_loss_mlp": 1.01811802,
      "epoch": 0.8616605543197259,
      "flos": 29789373674880.0,
      "grad_norm": 2.3160453414399793,
      "language_loss": 0.76121736,
      "learning_rate": 1.9729044549004193e-07,
      "loss": 0.78268236,
      "num_input_tokens_seen": 155064745,
      "step": 7166,
      "time_per_iteration": 2.5777416229248047
    },
    {
      "auxiliary_loss_clip": 0.01149603,
      "auxiliary_loss_mlp": 0.01021178,
      "balance_loss_clip": 1.04632354,
      "balance_loss_mlp": 1.01432967,
      "epoch": 0.8617807972103649,
      "flos": 28911609020160.0,
      "grad_norm": 2.392063385774023,
      "language_loss": 0.70326626,
      "learning_rate": 1.9695322404199822e-07,
      "loss": 0.72497404,
      "num_input_tokens_seen": 155086790,
      "step": 7167,
      "time_per_iteration": 2.4902288913726807
    },
    {
      "auxiliary_loss_clip": 0.01139383,
      "auxiliary_loss_mlp": 0.01025531,
      "balance_loss_clip": 1.04743254,
      "balance_loss_mlp": 1.01831341,
      "epoch": 0.861901040101004,
      "flos": 27673804391040.0,
      "grad_norm": 2.8428791164739757,
      "language_loss": 0.82057333,
      "learning_rate": 1.9661627611471654e-07,
      "loss": 0.84222245,
      "num_input_tokens_seen": 155106585,
      "step": 7168,
      "time_per_iteration": 2.53017520904541
    },
    {
      "auxiliary_loss_clip": 0.01143684,
      "auxiliary_loss_mlp": 0.01021376,
      "balance_loss_clip": 1.04487944,
      "balance_loss_mlp": 1.01368058,
      "epoch": 0.8620212829916432,
      "flos": 49748056755840.0,
      "grad_norm": 1.9479253952305635,
      "language_loss": 0.7032724,
      "learning_rate": 1.9627960175931246e-07,
      "loss": 0.72492301,
      "num_input_tokens_seen": 155131285,
      "step": 7169,
      "time_per_iteration": 3.5001347064971924
    },
    {
      "auxiliary_loss_clip": 0.01150959,
      "auxiliary_loss_mlp": 0.01027892,
      "balance_loss_clip": 1.04737639,
      "balance_loss_mlp": 1.02125502,
      "epoch": 0.8621415258822822,
      "flos": 21138672769920.0,
      "grad_norm": 1.9796452057515024,
      "language_loss": 0.74033833,
      "learning_rate": 1.9594320102685847e-07,
      "loss": 0.7621268,
      "num_input_tokens_seen": 155150555,
      "step": 7170,
      "time_per_iteration": 2.4481654167175293
    },
    {
      "auxiliary_loss_clip": 0.01125506,
      "auxiliary_loss_mlp": 0.00761345,
      "balance_loss_clip": 1.04096925,
      "balance_loss_mlp": 1.00041819,
      "epoch": 0.8622617687729213,
      "flos": 21689039934720.0,
      "grad_norm": 1.8623388644017276,
      "language_loss": 0.63945937,
      "learning_rate": 1.956070739683864e-07,
      "loss": 0.65832794,
      "num_input_tokens_seen": 155169890,
      "step": 7171,
      "time_per_iteration": 2.482297897338867
    },
    {
      "auxiliary_loss_clip": 0.01107715,
      "auxiliary_loss_mlp": 0.01020904,
      "balance_loss_clip": 1.03871894,
      "balance_loss_mlp": 1.01387656,
      "epoch": 0.8623820116635604,
      "flos": 26250592734720.0,
      "grad_norm": 1.4883773715103235,
      "language_loss": 0.74448526,
      "learning_rate": 1.9527122063488678e-07,
      "loss": 0.76577145,
      "num_input_tokens_seen": 155191005,
      "step": 7172,
      "time_per_iteration": 2.5603301525115967
    },
    {
      "auxiliary_loss_clip": 0.01132689,
      "auxiliary_loss_mlp": 0.01019137,
      "balance_loss_clip": 1.03947949,
      "balance_loss_mlp": 1.01248479,
      "epoch": 0.8625022545541995,
      "flos": 19647554451840.0,
      "grad_norm": 1.7552696607859688,
      "language_loss": 0.8015672,
      "learning_rate": 1.9493564107730755e-07,
      "loss": 0.82308543,
      "num_input_tokens_seen": 155211005,
      "step": 7173,
      "time_per_iteration": 2.484396457672119
    },
    {
      "auxiliary_loss_clip": 0.01130122,
      "auxiliary_loss_mlp": 0.01024789,
      "balance_loss_clip": 1.04005027,
      "balance_loss_mlp": 1.01823866,
      "epoch": 0.8626224974448385,
      "flos": 21908382336000.0,
      "grad_norm": 1.8993551317961008,
      "language_loss": 0.61120141,
      "learning_rate": 1.9460033534655684e-07,
      "loss": 0.63275051,
      "num_input_tokens_seen": 155230365,
      "step": 7174,
      "time_per_iteration": 2.513669967651367
    },
    {
      "auxiliary_loss_clip": 0.011275,
      "auxiliary_loss_mlp": 0.01024181,
      "balance_loss_clip": 1.03730571,
      "balance_loss_mlp": 1.01709414,
      "epoch": 0.8627427403354777,
      "flos": 23331198942720.0,
      "grad_norm": 2.01607008177014,
      "language_loss": 0.84329581,
      "learning_rate": 1.9426530349349978e-07,
      "loss": 0.86481261,
      "num_input_tokens_seen": 155250815,
      "step": 7175,
      "time_per_iteration": 2.4828202724456787
    },
    {
      "auxiliary_loss_clip": 0.01148406,
      "auxiliary_loss_mlp": 0.00761486,
      "balance_loss_clip": 1.04410505,
      "balance_loss_mlp": 1.00044036,
      "epoch": 0.8628629832261168,
      "flos": 16362877299840.0,
      "grad_norm": 1.8052870818472062,
      "language_loss": 0.64746886,
      "learning_rate": 1.9393054556896038e-07,
      "loss": 0.6665678,
      "num_input_tokens_seen": 155268515,
      "step": 7176,
      "time_per_iteration": 2.498678684234619
    },
    {
      "auxiliary_loss_clip": 0.01118835,
      "auxiliary_loss_mlp": 0.01023025,
      "balance_loss_clip": 1.03963065,
      "balance_loss_mlp": 1.01558614,
      "epoch": 0.8629832261167558,
      "flos": 28103941756800.0,
      "grad_norm": 2.2220566490366784,
      "language_loss": 0.69286215,
      "learning_rate": 1.9359606162372133e-07,
      "loss": 0.71428072,
      "num_input_tokens_seen": 155290120,
      "step": 7177,
      "time_per_iteration": 2.6455132961273193
    },
    {
      "auxiliary_loss_clip": 0.01163659,
      "auxiliary_loss_mlp": 0.01021851,
      "balance_loss_clip": 1.0466615,
      "balance_loss_mlp": 1.01476979,
      "epoch": 0.863103469007395,
      "flos": 20230061310720.0,
      "grad_norm": 1.6731357986143505,
      "language_loss": 0.70456815,
      "learning_rate": 1.9326185170852293e-07,
      "loss": 0.72642326,
      "num_input_tokens_seen": 155309085,
      "step": 7178,
      "time_per_iteration": 2.419935941696167
    },
    {
      "auxiliary_loss_clip": 0.01148719,
      "auxiliary_loss_mlp": 0.01022166,
      "balance_loss_clip": 1.0447228,
      "balance_loss_mlp": 1.01529956,
      "epoch": 0.863223711898034,
      "flos": 24498547044480.0,
      "grad_norm": 2.2492280640774993,
      "language_loss": 0.72275078,
      "learning_rate": 1.9292791587406598e-07,
      "loss": 0.74445969,
      "num_input_tokens_seen": 155327945,
      "step": 7179,
      "time_per_iteration": 2.484071969985962
    },
    {
      "auxiliary_loss_clip": 0.01148161,
      "auxiliary_loss_mlp": 0.00761599,
      "balance_loss_clip": 1.04366326,
      "balance_loss_mlp": 1.00045872,
      "epoch": 0.8633439547886731,
      "flos": 17675376261120.0,
      "grad_norm": 1.9709774704506708,
      "language_loss": 0.86926478,
      "learning_rate": 1.9259425417100661e-07,
      "loss": 0.88836235,
      "num_input_tokens_seen": 155344060,
      "step": 7180,
      "time_per_iteration": 2.4398000240325928
    },
    {
      "auxiliary_loss_clip": 0.01089203,
      "auxiliary_loss_mlp": 0.01028645,
      "balance_loss_clip": 1.03364253,
      "balance_loss_mlp": 1.02115881,
      "epoch": 0.8634641976793123,
      "flos": 12895055677440.0,
      "grad_norm": 2.273902149310325,
      "language_loss": 0.75118923,
      "learning_rate": 1.9226086664996234e-07,
      "loss": 0.7723676,
      "num_input_tokens_seen": 155362305,
      "step": 7181,
      "time_per_iteration": 2.5962767601013184
    },
    {
      "auxiliary_loss_clip": 0.01140575,
      "auxiliary_loss_mlp": 0.01025767,
      "balance_loss_clip": 1.04629219,
      "balance_loss_mlp": 1.01902294,
      "epoch": 0.8635844405699513,
      "flos": 23878980328320.0,
      "grad_norm": 2.1145054036243787,
      "language_loss": 0.74558616,
      "learning_rate": 1.9192775336150712e-07,
      "loss": 0.76724958,
      "num_input_tokens_seen": 155382605,
      "step": 7182,
      "time_per_iteration": 2.522078037261963
    },
    {
      "auxiliary_loss_clip": 0.01047765,
      "auxiliary_loss_mlp": 0.01001411,
      "balance_loss_clip": 1.00723147,
      "balance_loss_mlp": 1.00045192,
      "epoch": 0.8637046834605904,
      "flos": 60453387521280.0,
      "grad_norm": 0.7791210614873482,
      "language_loss": 0.56297916,
      "learning_rate": 1.915949143561739e-07,
      "loss": 0.58347094,
      "num_input_tokens_seen": 155437280,
      "step": 7183,
      "time_per_iteration": 3.0353150367736816
    },
    {
      "auxiliary_loss_clip": 0.01150454,
      "auxiliary_loss_mlp": 0.01025263,
      "balance_loss_clip": 1.04654932,
      "balance_loss_mlp": 1.01849771,
      "epoch": 0.8638249263512295,
      "flos": 20558751690240.0,
      "grad_norm": 1.7818436930383539,
      "language_loss": 0.78101188,
      "learning_rate": 1.9126234968445498e-07,
      "loss": 0.80276906,
      "num_input_tokens_seen": 155456970,
      "step": 7184,
      "time_per_iteration": 2.4756267070770264
    },
    {
      "auxiliary_loss_clip": 0.01166765,
      "auxiliary_loss_mlp": 0.01029052,
      "balance_loss_clip": 1.04843652,
      "balance_loss_mlp": 1.02172065,
      "epoch": 0.8639451692418686,
      "flos": 26615768353920.0,
      "grad_norm": 1.4424492467860675,
      "language_loss": 0.67526162,
      "learning_rate": 1.9093005939679884e-07,
      "loss": 0.69721973,
      "num_input_tokens_seen": 155478925,
      "step": 7185,
      "time_per_iteration": 2.46785044670105
    },
    {
      "auxiliary_loss_clip": 0.01151267,
      "auxiliary_loss_mlp": 0.01027334,
      "balance_loss_clip": 1.04599619,
      "balance_loss_mlp": 1.01983833,
      "epoch": 0.8640654121325076,
      "flos": 15122450977920.0,
      "grad_norm": 1.7826463810661388,
      "language_loss": 0.76235586,
      "learning_rate": 1.9059804354361452e-07,
      "loss": 0.78414184,
      "num_input_tokens_seen": 155496700,
      "step": 7186,
      "time_per_iteration": 2.3923046588897705
    },
    {
      "auxiliary_loss_clip": 0.0112752,
      "auxiliary_loss_mlp": 0.01022135,
      "balance_loss_clip": 1.03850758,
      "balance_loss_mlp": 1.01474404,
      "epoch": 0.8641856550231467,
      "flos": 31869068250240.0,
      "grad_norm": 2.7666134160528597,
      "language_loss": 0.70189577,
      "learning_rate": 1.902663021752684e-07,
      "loss": 0.72339237,
      "num_input_tokens_seen": 155518130,
      "step": 7187,
      "time_per_iteration": 2.5652294158935547
    },
    {
      "auxiliary_loss_clip": 0.01167919,
      "auxiliary_loss_mlp": 0.0102202,
      "balance_loss_clip": 1.04863822,
      "balance_loss_mlp": 1.01492715,
      "epoch": 0.8643058979137859,
      "flos": 14976545932800.0,
      "grad_norm": 2.0883023640443947,
      "language_loss": 0.82624209,
      "learning_rate": 1.8993483534208556e-07,
      "loss": 0.84814155,
      "num_input_tokens_seen": 155537040,
      "step": 7188,
      "time_per_iteration": 3.165694236755371
    },
    {
      "auxiliary_loss_clip": 0.01132931,
      "auxiliary_loss_mlp": 0.01025959,
      "balance_loss_clip": 1.04535341,
      "balance_loss_mlp": 1.0183475,
      "epoch": 0.8644261408044249,
      "flos": 13115726881920.0,
      "grad_norm": 2.3793782609293195,
      "language_loss": 0.74976712,
      "learning_rate": 1.8960364309434884e-07,
      "loss": 0.77135599,
      "num_input_tokens_seen": 155554535,
      "step": 7189,
      "time_per_iteration": 3.3004629611968994
    },
    {
      "auxiliary_loss_clip": 0.01089953,
      "auxiliary_loss_mlp": 0.00761578,
      "balance_loss_clip": 1.03765702,
      "balance_loss_mlp": 1.00045216,
      "epoch": 0.864546383695064,
      "flos": 20850920916480.0,
      "grad_norm": 1.6633452739898058,
      "language_loss": 0.78486216,
      "learning_rate": 1.8927272548229967e-07,
      "loss": 0.80337751,
      "num_input_tokens_seen": 155574225,
      "step": 7190,
      "time_per_iteration": 3.425333261489868
    },
    {
      "auxiliary_loss_clip": 0.0110905,
      "auxiliary_loss_mlp": 0.01024414,
      "balance_loss_clip": 1.04110312,
      "balance_loss_mlp": 1.01752329,
      "epoch": 0.8646666265857031,
      "flos": 21324582587520.0,
      "grad_norm": 1.5625766720170955,
      "language_loss": 0.83032167,
      "learning_rate": 1.8894208255613876e-07,
      "loss": 0.85165632,
      "num_input_tokens_seen": 155593540,
      "step": 7191,
      "time_per_iteration": 2.5528724193573
    },
    {
      "auxiliary_loss_clip": 0.01164949,
      "auxiliary_loss_mlp": 0.01021209,
      "balance_loss_clip": 1.04750764,
      "balance_loss_mlp": 1.01422906,
      "epoch": 0.8647868694763422,
      "flos": 19750833031680.0,
      "grad_norm": 2.0039131677404303,
      "language_loss": 0.77587777,
      "learning_rate": 1.8861171436602397e-07,
      "loss": 0.79773939,
      "num_input_tokens_seen": 155610655,
      "step": 7192,
      "time_per_iteration": 2.401355028152466
    },
    {
      "auxiliary_loss_clip": 0.01153267,
      "auxiliary_loss_mlp": 0.01026935,
      "balance_loss_clip": 1.0469681,
      "balance_loss_mlp": 1.01959729,
      "epoch": 0.8649071123669813,
      "flos": 26176760328960.0,
      "grad_norm": 2.2980797467456395,
      "language_loss": 0.80351007,
      "learning_rate": 1.882816209620719e-07,
      "loss": 0.82531208,
      "num_input_tokens_seen": 155627365,
      "step": 7193,
      "time_per_iteration": 2.4719905853271484
    },
    {
      "auxiliary_loss_clip": 0.01143419,
      "auxiliary_loss_mlp": 0.01026071,
      "balance_loss_clip": 1.05009413,
      "balance_loss_mlp": 1.0185132,
      "epoch": 0.8650273552576204,
      "flos": 20302888135680.0,
      "grad_norm": 4.460382097870268,
      "language_loss": 0.76593405,
      "learning_rate": 1.8795180239435738e-07,
      "loss": 0.78762901,
      "num_input_tokens_seen": 155646220,
      "step": 7194,
      "time_per_iteration": 2.4722025394439697
    },
    {
      "auxiliary_loss_clip": 0.01143013,
      "auxiliary_loss_mlp": 0.01025932,
      "balance_loss_clip": 1.045964,
      "balance_loss_mlp": 1.01846695,
      "epoch": 0.8651475981482595,
      "flos": 23951088881280.0,
      "grad_norm": 3.0105802304119966,
      "language_loss": 0.7641834,
      "learning_rate": 1.8762225871291348e-07,
      "loss": 0.78587282,
      "num_input_tokens_seen": 155662095,
      "step": 7195,
      "time_per_iteration": 2.491901397705078
    },
    {
      "auxiliary_loss_clip": 0.01165321,
      "auxiliary_loss_mlp": 0.0076158,
      "balance_loss_clip": 1.0470469,
      "balance_loss_mlp": 1.00047779,
      "epoch": 0.8652678410388985,
      "flos": 21684622561920.0,
      "grad_norm": 1.664628418272885,
      "language_loss": 0.80965567,
      "learning_rate": 1.8729298996773201e-07,
      "loss": 0.82892466,
      "num_input_tokens_seen": 155680845,
      "step": 7196,
      "time_per_iteration": 3.188472270965576
    },
    {
      "auxiliary_loss_clip": 0.01047082,
      "auxiliary_loss_mlp": 0.01002488,
      "balance_loss_clip": 1.00778794,
      "balance_loss_mlp": 1.00148618,
      "epoch": 0.8653880839295377,
      "flos": 65224660855680.0,
      "grad_norm": 0.8355544559823973,
      "language_loss": 0.61004162,
      "learning_rate": 1.8696399620876301e-07,
      "loss": 0.63053727,
      "num_input_tokens_seen": 155737875,
      "step": 7197,
      "time_per_iteration": 2.9900219440460205
    },
    {
      "auxiliary_loss_clip": 0.01120454,
      "auxiliary_loss_mlp": 0.01024241,
      "balance_loss_clip": 1.03924894,
      "balance_loss_mlp": 1.01658797,
      "epoch": 0.8655083268201768,
      "flos": 17749172753280.0,
      "grad_norm": 2.3259976094731494,
      "language_loss": 0.78793573,
      "learning_rate": 1.866352774859141e-07,
      "loss": 0.80938268,
      "num_input_tokens_seen": 155753100,
      "step": 7198,
      "time_per_iteration": 2.471815824508667
    },
    {
      "auxiliary_loss_clip": 0.0112523,
      "auxiliary_loss_mlp": 0.01022292,
      "balance_loss_clip": 1.0403744,
      "balance_loss_mlp": 1.01574779,
      "epoch": 0.8656285697108158,
      "flos": 20703974376960.0,
      "grad_norm": 2.6370103477383484,
      "language_loss": 0.69603854,
      "learning_rate": 1.8630683384905188e-07,
      "loss": 0.71751374,
      "num_input_tokens_seen": 155772430,
      "step": 7199,
      "time_per_iteration": 2.489495277404785
    },
    {
      "auxiliary_loss_clip": 0.01167,
      "auxiliary_loss_mlp": 0.00761697,
      "balance_loss_clip": 1.04912877,
      "balance_loss_mlp": 1.00046921,
      "epoch": 0.865748812601455,
      "flos": 18653833716480.0,
      "grad_norm": 2.1157043708059677,
      "language_loss": 0.88626057,
      "learning_rate": 1.8597866534800045e-07,
      "loss": 0.90554756,
      "num_input_tokens_seen": 155787545,
      "step": 7200,
      "time_per_iteration": 2.3899731636047363
    },
    {
      "auxiliary_loss_clip": 0.01153518,
      "auxiliary_loss_mlp": 0.00762037,
      "balance_loss_clip": 1.04650259,
      "balance_loss_mlp": 1.00039446,
      "epoch": 0.865869055492094,
      "flos": 70652554807680.0,
      "grad_norm": 2.4030041371908037,
      "language_loss": 0.74556565,
      "learning_rate": 1.8565077203254398e-07,
      "loss": 0.76472116,
      "num_input_tokens_seen": 155813005,
      "step": 7201,
      "time_per_iteration": 2.8548152446746826
    },
    {
      "auxiliary_loss_clip": 0.01125996,
      "auxiliary_loss_mlp": 0.01027375,
      "balance_loss_clip": 1.04730594,
      "balance_loss_mlp": 1.0201304,
      "epoch": 0.8659892983827331,
      "flos": 17383961220480.0,
      "grad_norm": 2.8187071375143,
      "language_loss": 0.72664392,
      "learning_rate": 1.8532315395242203e-07,
      "loss": 0.74817753,
      "num_input_tokens_seen": 155829455,
      "step": 7202,
      "time_per_iteration": 2.478963851928711
    },
    {
      "auxiliary_loss_clip": 0.01123371,
      "auxiliary_loss_mlp": 0.01021681,
      "balance_loss_clip": 1.04116428,
      "balance_loss_mlp": 1.01486802,
      "epoch": 0.8661095412733723,
      "flos": 17895221452800.0,
      "grad_norm": 2.187291378597507,
      "language_loss": 0.7256676,
      "learning_rate": 1.849958111573353e-07,
      "loss": 0.74711818,
      "num_input_tokens_seen": 155848060,
      "step": 7203,
      "time_per_iteration": 2.4842655658721924
    },
    {
      "auxiliary_loss_clip": 0.01162012,
      "auxiliary_loss_mlp": 0.01021479,
      "balance_loss_clip": 1.0460043,
      "balance_loss_mlp": 1.01482391,
      "epoch": 0.8662297841640113,
      "flos": 18224163227520.0,
      "grad_norm": 1.69114508267875,
      "language_loss": 0.64192033,
      "learning_rate": 1.8466874369694074e-07,
      "loss": 0.66375524,
      "num_input_tokens_seen": 155865755,
      "step": 7204,
      "time_per_iteration": 2.408764362335205
    },
    {
      "auxiliary_loss_clip": 0.01123255,
      "auxiliary_loss_mlp": 0.01023464,
      "balance_loss_clip": 1.03958607,
      "balance_loss_mlp": 1.01678586,
      "epoch": 0.8663500270546504,
      "flos": 16362159027840.0,
      "grad_norm": 2.5639042246858006,
      "language_loss": 0.7051791,
      "learning_rate": 1.843419516208542e-07,
      "loss": 0.7266463,
      "num_input_tokens_seen": 155882680,
      "step": 7205,
      "time_per_iteration": 2.4768755435943604
    },
    {
      "auxiliary_loss_clip": 0.01153133,
      "auxiliary_loss_mlp": 0.01023827,
      "balance_loss_clip": 1.04743052,
      "balance_loss_mlp": 1.01622105,
      "epoch": 0.8664702699452895,
      "flos": 17894431353600.0,
      "grad_norm": 2.30049604041738,
      "language_loss": 0.79530966,
      "learning_rate": 1.8401543497865047e-07,
      "loss": 0.81707931,
      "num_input_tokens_seen": 155900680,
      "step": 7206,
      "time_per_iteration": 2.4018285274505615
    },
    {
      "auxiliary_loss_clip": 0.0115393,
      "auxiliary_loss_mlp": 0.00761717,
      "balance_loss_clip": 1.0466845,
      "balance_loss_mlp": 1.00041735,
      "epoch": 0.8665905128359286,
      "flos": 30736373794560.0,
      "grad_norm": 1.8444418785065984,
      "language_loss": 0.64331782,
      "learning_rate": 1.836891938198608e-07,
      "loss": 0.66247427,
      "num_input_tokens_seen": 155921105,
      "step": 7207,
      "time_per_iteration": 2.519784450531006
    },
    {
      "auxiliary_loss_clip": 0.01135419,
      "auxiliary_loss_mlp": 0.01026907,
      "balance_loss_clip": 1.04418874,
      "balance_loss_mlp": 1.01985538,
      "epoch": 0.8667107557265676,
      "flos": 18656419495680.0,
      "grad_norm": 2.112179523624969,
      "language_loss": 0.71469545,
      "learning_rate": 1.8336322819397677e-07,
      "loss": 0.73631871,
      "num_input_tokens_seen": 155938640,
      "step": 7208,
      "time_per_iteration": 2.4382028579711914
    },
    {
      "auxiliary_loss_clip": 0.01127471,
      "auxiliary_loss_mlp": 0.01024617,
      "balance_loss_clip": 1.04024673,
      "balance_loss_mlp": 1.01731586,
      "epoch": 0.8668309986172068,
      "flos": 20083725302400.0,
      "grad_norm": 3.634784215260924,
      "language_loss": 0.6269623,
      "learning_rate": 1.8303753815044654e-07,
      "loss": 0.64848322,
      "num_input_tokens_seen": 155957945,
      "step": 7209,
      "time_per_iteration": 2.5275588035583496
    },
    {
      "auxiliary_loss_clip": 0.01143566,
      "auxiliary_loss_mlp": 0.01023894,
      "balance_loss_clip": 1.04285336,
      "balance_loss_mlp": 1.01640725,
      "epoch": 0.8669512415078459,
      "flos": 21615099788160.0,
      "grad_norm": 21.689983554009228,
      "language_loss": 0.70910311,
      "learning_rate": 1.827121237386773e-07,
      "loss": 0.73077774,
      "num_input_tokens_seen": 155975390,
      "step": 7210,
      "time_per_iteration": 2.5133891105651855
    },
    {
      "auxiliary_loss_clip": 0.01138843,
      "auxiliary_loss_mlp": 0.01026775,
      "balance_loss_clip": 1.04371071,
      "balance_loss_mlp": 1.01966405,
      "epoch": 0.8670714843984849,
      "flos": 17703601372800.0,
      "grad_norm": 2.5703796765308438,
      "language_loss": 0.74971163,
      "learning_rate": 1.8238698500803374e-07,
      "loss": 0.77136779,
      "num_input_tokens_seen": 155988155,
      "step": 7211,
      "time_per_iteration": 2.404639482498169
    },
    {
      "auxiliary_loss_clip": 0.01052786,
      "auxiliary_loss_mlp": 0.01001665,
      "balance_loss_clip": 1.00803161,
      "balance_loss_mlp": 1.00067604,
      "epoch": 0.8671917272891241,
      "flos": 60705483125760.0,
      "grad_norm": 0.7177639008424767,
      "language_loss": 0.56275952,
      "learning_rate": 1.820621220078391e-07,
      "loss": 0.58330399,
      "num_input_tokens_seen": 156052065,
      "step": 7212,
      "time_per_iteration": 3.0944032669067383
    },
    {
      "auxiliary_loss_clip": 0.01163049,
      "auxiliary_loss_mlp": 0.01018756,
      "balance_loss_clip": 1.04521549,
      "balance_loss_mlp": 1.01137066,
      "epoch": 0.8673119701797631,
      "flos": 20451881750400.0,
      "grad_norm": 1.6343438347074417,
      "language_loss": 0.67810798,
      "learning_rate": 1.8173753478737553e-07,
      "loss": 0.69992602,
      "num_input_tokens_seen": 156072500,
      "step": 7213,
      "time_per_iteration": 2.422515392303467
    },
    {
      "auxiliary_loss_clip": 0.01168366,
      "auxiliary_loss_mlp": 0.01026591,
      "balance_loss_clip": 1.04770541,
      "balance_loss_mlp": 1.01931643,
      "epoch": 0.8674322130704022,
      "flos": 19647410797440.0,
      "grad_norm": 2.129531267401255,
      "language_loss": 0.79784894,
      "learning_rate": 1.8141322339588205e-07,
      "loss": 0.81979847,
      "num_input_tokens_seen": 156089840,
      "step": 7214,
      "time_per_iteration": 2.3884010314941406
    },
    {
      "auxiliary_loss_clip": 0.01165535,
      "auxiliary_loss_mlp": 0.01025992,
      "balance_loss_clip": 1.04775238,
      "balance_loss_mlp": 1.01889634,
      "epoch": 0.8675524559610414,
      "flos": 26025001367040.0,
      "grad_norm": 1.9739554735825269,
      "language_loss": 0.70256305,
      "learning_rate": 1.810891878825569e-07,
      "loss": 0.72447836,
      "num_input_tokens_seen": 156109815,
      "step": 7215,
      "time_per_iteration": 3.151298999786377
    },
    {
      "auxiliary_loss_clip": 0.01132954,
      "auxiliary_loss_mlp": 0.0102297,
      "balance_loss_clip": 1.04019213,
      "balance_loss_mlp": 1.01597226,
      "epoch": 0.8676726988516804,
      "flos": 15049444584960.0,
      "grad_norm": 2.3792088920886827,
      "language_loss": 0.7182318,
      "learning_rate": 1.8076542829655561e-07,
      "loss": 0.73979104,
      "num_input_tokens_seen": 156128620,
      "step": 7216,
      "time_per_iteration": 3.5483880043029785
    },
    {
      "auxiliary_loss_clip": 0.01140362,
      "auxiliary_loss_mlp": 0.01027005,
      "balance_loss_clip": 1.0467881,
      "balance_loss_mlp": 1.01904202,
      "epoch": 0.8677929417423195,
      "flos": 16288111140480.0,
      "grad_norm": 1.8380978494496327,
      "language_loss": 0.79279149,
      "learning_rate": 1.8044194468699203e-07,
      "loss": 0.81446517,
      "num_input_tokens_seen": 156145930,
      "step": 7217,
      "time_per_iteration": 3.3012497425079346
    },
    {
      "auxiliary_loss_clip": 0.01136375,
      "auxiliary_loss_mlp": 0.0102343,
      "balance_loss_clip": 1.04667044,
      "balance_loss_mlp": 1.01642644,
      "epoch": 0.8679131846329585,
      "flos": 18844160906880.0,
      "grad_norm": 2.6265625464221776,
      "language_loss": 0.76021826,
      "learning_rate": 1.8011873710293912e-07,
      "loss": 0.7818163,
      "num_input_tokens_seen": 156164435,
      "step": 7218,
      "time_per_iteration": 2.518362522125244
    },
    {
      "auxiliary_loss_clip": 0.01147029,
      "auxiliary_loss_mlp": 0.01023731,
      "balance_loss_clip": 1.0453763,
      "balance_loss_mlp": 1.01647723,
      "epoch": 0.8680334275235977,
      "flos": 33620718890880.0,
      "grad_norm": 1.7668753068412386,
      "language_loss": 0.69502074,
      "learning_rate": 1.7979580559342677e-07,
      "loss": 0.71672833,
      "num_input_tokens_seen": 156185165,
      "step": 7219,
      "time_per_iteration": 2.539588451385498
    },
    {
      "auxiliary_loss_clip": 0.0113732,
      "auxiliary_loss_mlp": 0.01024839,
      "balance_loss_clip": 1.04524732,
      "balance_loss_mlp": 1.01791036,
      "epoch": 0.8681536704142367,
      "flos": 24681152810880.0,
      "grad_norm": 1.661279573481478,
      "language_loss": 0.66628891,
      "learning_rate": 1.7947315020744358e-07,
      "loss": 0.6879105,
      "num_input_tokens_seen": 156206260,
      "step": 7220,
      "time_per_iteration": 2.5168042182922363
    },
    {
      "auxiliary_loss_clip": 0.01133822,
      "auxiliary_loss_mlp": 0.01020396,
      "balance_loss_clip": 1.04179478,
      "balance_loss_mlp": 1.01354706,
      "epoch": 0.8682739133048758,
      "flos": 20011042131840.0,
      "grad_norm": 1.7472550506304882,
      "language_loss": 0.8028605,
      "learning_rate": 1.7915077099393594e-07,
      "loss": 0.82440269,
      "num_input_tokens_seen": 156222860,
      "step": 7221,
      "time_per_iteration": 2.4560141563415527
    },
    {
      "auxiliary_loss_clip": 0.01153419,
      "auxiliary_loss_mlp": 0.01025319,
      "balance_loss_clip": 1.04447746,
      "balance_loss_mlp": 1.01802301,
      "epoch": 0.868394156195515,
      "flos": 16654759217280.0,
      "grad_norm": 1.8490439489410155,
      "language_loss": 0.73337972,
      "learning_rate": 1.788286680018083e-07,
      "loss": 0.75516713,
      "num_input_tokens_seen": 156241570,
      "step": 7222,
      "time_per_iteration": 2.440154552459717
    },
    {
      "auxiliary_loss_clip": 0.01141224,
      "auxiliary_loss_mlp": 0.01025568,
      "balance_loss_clip": 1.04470885,
      "balance_loss_mlp": 1.01887417,
      "epoch": 0.868514399086154,
      "flos": 28001381448960.0,
      "grad_norm": 1.8230106415532108,
      "language_loss": 0.72670174,
      "learning_rate": 1.7850684127992443e-07,
      "loss": 0.74836957,
      "num_input_tokens_seen": 156261315,
      "step": 7223,
      "time_per_iteration": 3.2229413986206055
    },
    {
      "auxiliary_loss_clip": 0.01123506,
      "auxiliary_loss_mlp": 0.01024615,
      "balance_loss_clip": 1.04491425,
      "balance_loss_mlp": 1.0175817,
      "epoch": 0.8686346419767931,
      "flos": 20084587228800.0,
      "grad_norm": 1.735764597549995,
      "language_loss": 0.70108593,
      "learning_rate": 1.7818529087710378e-07,
      "loss": 0.72256708,
      "num_input_tokens_seen": 156281670,
      "step": 7224,
      "time_per_iteration": 2.5162742137908936
    },
    {
      "auxiliary_loss_clip": 0.01147513,
      "auxiliary_loss_mlp": 0.00761875,
      "balance_loss_clip": 1.04344988,
      "balance_loss_mlp": 1.00043964,
      "epoch": 0.8687548848674322,
      "flos": 18223516782720.0,
      "grad_norm": 1.6537983379543508,
      "language_loss": 0.84135675,
      "learning_rate": 1.7786401684212637e-07,
      "loss": 0.86045063,
      "num_input_tokens_seen": 156300500,
      "step": 7225,
      "time_per_iteration": 2.427837610244751
    },
    {
      "auxiliary_loss_clip": 0.01030253,
      "auxiliary_loss_mlp": 0.01003754,
      "balance_loss_clip": 1.01050448,
      "balance_loss_mlp": 1.00300622,
      "epoch": 0.8688751277580713,
      "flos": 70457885049600.0,
      "grad_norm": 0.7448156500959605,
      "language_loss": 0.55952418,
      "learning_rate": 1.7754301922372883e-07,
      "loss": 0.57986426,
      "num_input_tokens_seen": 156350145,
      "step": 7226,
      "time_per_iteration": 2.960059881210327
    },
    {
      "auxiliary_loss_clip": 0.01101348,
      "auxiliary_loss_mlp": 0.01023442,
      "balance_loss_clip": 1.04107022,
      "balance_loss_mlp": 1.01597977,
      "epoch": 0.8689953706487104,
      "flos": 26906788344960.0,
      "grad_norm": 1.9821852930127382,
      "language_loss": 0.80675328,
      "learning_rate": 1.7722229807060617e-07,
      "loss": 0.82800114,
      "num_input_tokens_seen": 156368725,
      "step": 7227,
      "time_per_iteration": 2.626340389251709
    },
    {
      "auxiliary_loss_clip": 0.01113914,
      "auxiliary_loss_mlp": 0.01023528,
      "balance_loss_clip": 1.03940463,
      "balance_loss_mlp": 1.01679015,
      "epoch": 0.8691156135393495,
      "flos": 34637385438720.0,
      "grad_norm": 2.6854853968846255,
      "language_loss": 0.82057846,
      "learning_rate": 1.7690185343141172e-07,
      "loss": 0.84195292,
      "num_input_tokens_seen": 156388640,
      "step": 7228,
      "time_per_iteration": 2.6248438358306885
    },
    {
      "auxiliary_loss_clip": 0.01136533,
      "auxiliary_loss_mlp": 0.01020992,
      "balance_loss_clip": 1.04294658,
      "balance_loss_mlp": 1.01432788,
      "epoch": 0.8692358564299886,
      "flos": 18989814556800.0,
      "grad_norm": 2.2716592416477717,
      "language_loss": 0.69993812,
      "learning_rate": 1.7658168535475615e-07,
      "loss": 0.72151327,
      "num_input_tokens_seen": 156406425,
      "step": 7229,
      "time_per_iteration": 2.4600648880004883
    },
    {
      "auxiliary_loss_clip": 0.0114353,
      "auxiliary_loss_mlp": 0.01026314,
      "balance_loss_clip": 1.04729962,
      "balance_loss_mlp": 1.0191381,
      "epoch": 0.8693560993206276,
      "flos": 30370839039360.0,
      "grad_norm": 1.6102992224663388,
      "language_loss": 0.64323294,
      "learning_rate": 1.7626179388920948e-07,
      "loss": 0.66493136,
      "num_input_tokens_seen": 156427705,
      "step": 7230,
      "time_per_iteration": 2.569403886795044
    },
    {
      "auxiliary_loss_clip": 0.01137886,
      "auxiliary_loss_mlp": 0.00761504,
      "balance_loss_clip": 1.04577625,
      "balance_loss_mlp": 1.00051188,
      "epoch": 0.8694763422112668,
      "flos": 27200430028800.0,
      "grad_norm": 1.6627130175782878,
      "language_loss": 0.80660558,
      "learning_rate": 1.7594217908329866e-07,
      "loss": 0.82559949,
      "num_input_tokens_seen": 156449890,
      "step": 7231,
      "time_per_iteration": 2.549978494644165
    },
    {
      "auxiliary_loss_clip": 0.01129133,
      "auxiliary_loss_mlp": 0.01020288,
      "balance_loss_clip": 1.04360998,
      "balance_loss_mlp": 1.01343369,
      "epoch": 0.8695965851019059,
      "flos": 26139161767680.0,
      "grad_norm": 2.1040868894728995,
      "language_loss": 0.74063492,
      "learning_rate": 1.7562284098550895e-07,
      "loss": 0.76212919,
      "num_input_tokens_seen": 156469600,
      "step": 7232,
      "time_per_iteration": 2.504659652709961
    },
    {
      "auxiliary_loss_clip": 0.01036728,
      "auxiliary_loss_mlp": 0.01001842,
      "balance_loss_clip": 1.00906968,
      "balance_loss_mlp": 1.00084066,
      "epoch": 0.8697168279925449,
      "flos": 67332616456320.0,
      "grad_norm": 0.8363226528400701,
      "language_loss": 0.62288058,
      "learning_rate": 1.753037796442838e-07,
      "loss": 0.64326626,
      "num_input_tokens_seen": 156529040,
      "step": 7233,
      "time_per_iteration": 3.0336718559265137
    },
    {
      "auxiliary_loss_clip": 0.0116483,
      "auxiliary_loss_mlp": 0.0102388,
      "balance_loss_clip": 1.04627991,
      "balance_loss_mlp": 1.0163815,
      "epoch": 0.8698370708831841,
      "flos": 19718693337600.0,
      "grad_norm": 2.1924795882386596,
      "language_loss": 0.75254858,
      "learning_rate": 1.74984995108024e-07,
      "loss": 0.77443564,
      "num_input_tokens_seen": 156546970,
      "step": 7234,
      "time_per_iteration": 2.4014999866485596
    },
    {
      "auxiliary_loss_clip": 0.01152445,
      "auxiliary_loss_mlp": 0.01020582,
      "balance_loss_clip": 1.04538691,
      "balance_loss_mlp": 1.01382589,
      "epoch": 0.8699573137738231,
      "flos": 12859971068160.0,
      "grad_norm": 2.0490979757867835,
      "language_loss": 0.83430582,
      "learning_rate": 1.7466648742508981e-07,
      "loss": 0.85603607,
      "num_input_tokens_seen": 156563155,
      "step": 7235,
      "time_per_iteration": 2.3954050540924072
    },
    {
      "auxiliary_loss_clip": 0.01134735,
      "auxiliary_loss_mlp": 0.01028024,
      "balance_loss_clip": 1.04438365,
      "balance_loss_mlp": 1.02059102,
      "epoch": 0.8700775566644622,
      "flos": 17420733768960.0,
      "grad_norm": 1.8509984896470082,
      "language_loss": 0.84744728,
      "learning_rate": 1.7434825664379837e-07,
      "loss": 0.86907488,
      "num_input_tokens_seen": 156581660,
      "step": 7236,
      "time_per_iteration": 2.4437103271484375
    },
    {
      "auxiliary_loss_clip": 0.01152251,
      "auxiliary_loss_mlp": 0.01020869,
      "balance_loss_clip": 1.04583621,
      "balance_loss_mlp": 1.01381803,
      "epoch": 0.8701977995551013,
      "flos": 13735221770880.0,
      "grad_norm": 2.472246871219972,
      "language_loss": 0.86005932,
      "learning_rate": 1.740303028124246e-07,
      "loss": 0.88179052,
      "num_input_tokens_seen": 156597720,
      "step": 7237,
      "time_per_iteration": 2.4325177669525146
    },
    {
      "auxiliary_loss_clip": 0.01087302,
      "auxiliary_loss_mlp": 0.01022647,
      "balance_loss_clip": 1.03820395,
      "balance_loss_mlp": 1.01557159,
      "epoch": 0.8703180424457404,
      "flos": 30555707362560.0,
      "grad_norm": 2.1474210848133737,
      "language_loss": 0.75847697,
      "learning_rate": 1.7371262597920212e-07,
      "loss": 0.77957654,
      "num_input_tokens_seen": 156619780,
      "step": 7238,
      "time_per_iteration": 2.6330466270446777
    },
    {
      "auxiliary_loss_clip": 0.01110035,
      "auxiliary_loss_mlp": 0.01031193,
      "balance_loss_clip": 1.04637206,
      "balance_loss_mlp": 1.02401686,
      "epoch": 0.8704382853363795,
      "flos": 19608986223360.0,
      "grad_norm": 1.4507670531626606,
      "language_loss": 0.76313674,
      "learning_rate": 1.7339522619232195e-07,
      "loss": 0.784549,
      "num_input_tokens_seen": 156638160,
      "step": 7239,
      "time_per_iteration": 2.5429558753967285
    },
    {
      "auxiliary_loss_clip": 0.01144011,
      "auxiliary_loss_mlp": 0.01025196,
      "balance_loss_clip": 1.04311478,
      "balance_loss_mlp": 1.01756704,
      "epoch": 0.8705585282270186,
      "flos": 26613900846720.0,
      "grad_norm": 1.8145665293701223,
      "language_loss": 0.7527864,
      "learning_rate": 1.730781034999338e-07,
      "loss": 0.77447844,
      "num_input_tokens_seen": 156659740,
      "step": 7240,
      "time_per_iteration": 2.5225019454956055
    },
    {
      "auxiliary_loss_clip": 0.01162477,
      "auxiliary_loss_mlp": 0.01025346,
      "balance_loss_clip": 1.04853749,
      "balance_loss_mlp": 1.01852775,
      "epoch": 0.8706787711176577,
      "flos": 34090465979520.0,
      "grad_norm": 2.189082264317166,
      "language_loss": 0.7326014,
      "learning_rate": 1.7276125795014497e-07,
      "loss": 0.75447965,
      "num_input_tokens_seen": 156678190,
      "step": 7241,
      "time_per_iteration": 3.258859872817993
    },
    {
      "auxiliary_loss_clip": 0.0113959,
      "auxiliary_loss_mlp": 0.01024554,
      "balance_loss_clip": 1.04195809,
      "balance_loss_mlp": 1.01709104,
      "epoch": 0.8707990140082967,
      "flos": 14611513968000.0,
      "grad_norm": 1.8861881373261027,
      "language_loss": 0.67529809,
      "learning_rate": 1.7244468959102054e-07,
      "loss": 0.69693953,
      "num_input_tokens_seen": 156695245,
      "step": 7242,
      "time_per_iteration": 3.2692439556121826
    },
    {
      "auxiliary_loss_clip": 0.01152093,
      "auxiliary_loss_mlp": 0.0102106,
      "balance_loss_clip": 1.04796839,
      "balance_loss_mlp": 1.01383018,
      "epoch": 0.8709192568989359,
      "flos": 20084156265600.0,
      "grad_norm": 2.162124046544784,
      "language_loss": 0.85106719,
      "learning_rate": 1.7212839847058348e-07,
      "loss": 0.87279868,
      "num_input_tokens_seen": 156710375,
      "step": 7243,
      "time_per_iteration": 3.285090923309326
    },
    {
      "auxiliary_loss_clip": 0.01102517,
      "auxiliary_loss_mlp": 0.01021966,
      "balance_loss_clip": 1.03975677,
      "balance_loss_mlp": 1.01538289,
      "epoch": 0.871039499789575,
      "flos": 16727083251840.0,
      "grad_norm": 1.9652953171757017,
      "language_loss": 0.73743653,
      "learning_rate": 1.718123846368147e-07,
      "loss": 0.7586813,
      "num_input_tokens_seen": 156729420,
      "step": 7244,
      "time_per_iteration": 2.563595771789551
    },
    {
      "auxiliary_loss_clip": 0.01137063,
      "auxiliary_loss_mlp": 0.0076149,
      "balance_loss_clip": 1.04633033,
      "balance_loss_mlp": 1.00042486,
      "epoch": 0.871159742680214,
      "flos": 21068790860160.0,
      "grad_norm": 1.6732765310763076,
      "language_loss": 0.71722603,
      "learning_rate": 1.714966481376543e-07,
      "loss": 0.7362116,
      "num_input_tokens_seen": 156746100,
      "step": 7245,
      "time_per_iteration": 2.4658870697021484
    },
    {
      "auxiliary_loss_clip": 0.01149281,
      "auxiliary_loss_mlp": 0.01023633,
      "balance_loss_clip": 1.04388595,
      "balance_loss_mlp": 1.0166893,
      "epoch": 0.8712799855708532,
      "flos": 28256526731520.0,
      "grad_norm": 1.9632740234594193,
      "language_loss": 0.83133149,
      "learning_rate": 1.7118118902099797e-07,
      "loss": 0.8530606,
      "num_input_tokens_seen": 156764185,
      "step": 7246,
      "time_per_iteration": 2.474360466003418
    },
    {
      "auxiliary_loss_clip": 0.01150039,
      "auxiliary_loss_mlp": 0.01029283,
      "balance_loss_clip": 1.04477,
      "balance_loss_mlp": 1.02224934,
      "epoch": 0.8714002284614922,
      "flos": 22236677665920.0,
      "grad_norm": 1.5796608125500102,
      "language_loss": 0.8056609,
      "learning_rate": 1.7086600733470146e-07,
      "loss": 0.82745409,
      "num_input_tokens_seen": 156784855,
      "step": 7247,
      "time_per_iteration": 2.445566177368164
    },
    {
      "auxiliary_loss_clip": 0.01147125,
      "auxiliary_loss_mlp": 0.01026997,
      "balance_loss_clip": 1.04464507,
      "balance_loss_mlp": 1.02038074,
      "epoch": 0.8715204713521313,
      "flos": 21431919404160.0,
      "grad_norm": 1.935767532876687,
      "language_loss": 0.77335012,
      "learning_rate": 1.7055110312657738e-07,
      "loss": 0.79509139,
      "num_input_tokens_seen": 156804350,
      "step": 7248,
      "time_per_iteration": 2.4370718002319336
    },
    {
      "auxiliary_loss_clip": 0.01132677,
      "auxiliary_loss_mlp": 0.01027483,
      "balance_loss_clip": 1.04434276,
      "balance_loss_mlp": 1.01993704,
      "epoch": 0.8716407142427703,
      "flos": 23440439180160.0,
      "grad_norm": 2.5153054770464633,
      "language_loss": 0.74408627,
      "learning_rate": 1.702364764443962e-07,
      "loss": 0.76568788,
      "num_input_tokens_seen": 156823425,
      "step": 7249,
      "time_per_iteration": 3.2174181938171387
    },
    {
      "auxiliary_loss_clip": 0.01089067,
      "auxiliary_loss_mlp": 0.01021527,
      "balance_loss_clip": 1.03563094,
      "balance_loss_mlp": 1.0139842,
      "epoch": 0.8717609571334095,
      "flos": 27958683156480.0,
      "grad_norm": 1.8951449523988524,
      "language_loss": 0.72447848,
      "learning_rate": 1.6992212733588685e-07,
      "loss": 0.74558449,
      "num_input_tokens_seen": 156843090,
      "step": 7250,
      "time_per_iteration": 2.640659809112549
    },
    {
      "auxiliary_loss_clip": 0.01133421,
      "auxiliary_loss_mlp": 0.01024201,
      "balance_loss_clip": 1.04252803,
      "balance_loss_mlp": 1.01718819,
      "epoch": 0.8718812000240486,
      "flos": 25479482538240.0,
      "grad_norm": 1.7649682355255958,
      "language_loss": 0.75195658,
      "learning_rate": 1.6960805584873538e-07,
      "loss": 0.77353281,
      "num_input_tokens_seen": 156861090,
      "step": 7251,
      "time_per_iteration": 2.500795841217041
    },
    {
      "auxiliary_loss_clip": 0.01111381,
      "auxiliary_loss_mlp": 0.01025281,
      "balance_loss_clip": 1.04045248,
      "balance_loss_mlp": 1.01842391,
      "epoch": 0.8720014429146876,
      "flos": 23403056100480.0,
      "grad_norm": 2.812243738197062,
      "language_loss": 0.78263247,
      "learning_rate": 1.6929426203058684e-07,
      "loss": 0.80399907,
      "num_input_tokens_seen": 156881515,
      "step": 7252,
      "time_per_iteration": 2.5852882862091064
    },
    {
      "auxiliary_loss_clip": 0.01167568,
      "auxiliary_loss_mlp": 0.00762262,
      "balance_loss_clip": 1.04532886,
      "balance_loss_mlp": 1.00051188,
      "epoch": 0.8721216858053268,
      "flos": 24352821567360.0,
      "grad_norm": 2.770436258071441,
      "language_loss": 0.80376977,
      "learning_rate": 1.689807459290431e-07,
      "loss": 0.82306808,
      "num_input_tokens_seen": 156900170,
      "step": 7253,
      "time_per_iteration": 2.434401512145996
    },
    {
      "auxiliary_loss_clip": 0.01137395,
      "auxiliary_loss_mlp": 0.01023688,
      "balance_loss_clip": 1.04412496,
      "balance_loss_mlp": 1.01732874,
      "epoch": 0.8722419286959658,
      "flos": 33869687034240.0,
      "grad_norm": 1.9828134672737794,
      "language_loss": 0.70552158,
      "learning_rate": 1.6866750759166437e-07,
      "loss": 0.72713244,
      "num_input_tokens_seen": 156920150,
      "step": 7254,
      "time_per_iteration": 2.590894937515259
    },
    {
      "auxiliary_loss_clip": 0.01117179,
      "auxiliary_loss_mlp": 0.01021353,
      "balance_loss_clip": 1.03890204,
      "balance_loss_mlp": 1.0138483,
      "epoch": 0.8723621715866049,
      "flos": 18369385914240.0,
      "grad_norm": 2.325752208452468,
      "language_loss": 0.77237034,
      "learning_rate": 1.6835454706596865e-07,
      "loss": 0.79375571,
      "num_input_tokens_seen": 156937980,
      "step": 7255,
      "time_per_iteration": 2.482780933380127
    },
    {
      "auxiliary_loss_clip": 0.01165945,
      "auxiliary_loss_mlp": 0.01026586,
      "balance_loss_clip": 1.04769075,
      "balance_loss_mlp": 1.01935589,
      "epoch": 0.8724824144772441,
      "flos": 22013348855040.0,
      "grad_norm": 2.354655660311847,
      "language_loss": 0.737064,
      "learning_rate": 1.680418643994317e-07,
      "loss": 0.75898921,
      "num_input_tokens_seen": 156956550,
      "step": 7256,
      "time_per_iteration": 2.4169082641601562
    },
    {
      "auxiliary_loss_clip": 0.01062112,
      "auxiliary_loss_mlp": 0.01001034,
      "balance_loss_clip": 1.00829339,
      "balance_loss_mlp": 1.00016332,
      "epoch": 0.8726026573678831,
      "flos": 66698720213760.0,
      "grad_norm": 0.8921880064192208,
      "language_loss": 0.64547169,
      "learning_rate": 1.6772945963948738e-07,
      "loss": 0.66610312,
      "num_input_tokens_seen": 157014715,
      "step": 7257,
      "time_per_iteration": 3.026306629180908
    },
    {
      "auxiliary_loss_clip": 0.01131867,
      "auxiliary_loss_mlp": 0.01023775,
      "balance_loss_clip": 1.04413319,
      "balance_loss_mlp": 1.01699543,
      "epoch": 0.8727229002585222,
      "flos": 13370908078080.0,
      "grad_norm": 2.9417976147240394,
      "language_loss": 0.77364326,
      "learning_rate": 1.6741733283352733e-07,
      "loss": 0.79519969,
      "num_input_tokens_seen": 157032320,
      "step": 7258,
      "time_per_iteration": 2.4624507427215576
    },
    {
      "auxiliary_loss_clip": 0.01116296,
      "auxiliary_loss_mlp": 0.01026885,
      "balance_loss_clip": 1.0433048,
      "balance_loss_mlp": 1.01999807,
      "epoch": 0.8728431431491613,
      "flos": 21796987282560.0,
      "grad_norm": 1.5145087344391084,
      "language_loss": 0.8384738,
      "learning_rate": 1.6710548402890102e-07,
      "loss": 0.8599056,
      "num_input_tokens_seen": 157052845,
      "step": 7259,
      "time_per_iteration": 2.553818941116333
    },
    {
      "auxiliary_loss_clip": 0.01168785,
      "auxiliary_loss_mlp": 0.01024988,
      "balance_loss_clip": 1.04685569,
      "balance_loss_mlp": 1.01740909,
      "epoch": 0.8729633860398004,
      "flos": 36173823742080.0,
      "grad_norm": 1.8175940698179587,
      "language_loss": 0.66873902,
      "learning_rate": 1.6679391327291527e-07,
      "loss": 0.69067675,
      "num_input_tokens_seen": 157074050,
      "step": 7260,
      "time_per_iteration": 2.5198378562927246
    },
    {
      "auxiliary_loss_clip": 0.01135614,
      "auxiliary_loss_mlp": 0.01023822,
      "balance_loss_clip": 1.04127288,
      "balance_loss_mlp": 1.01706612,
      "epoch": 0.8730836289304394,
      "flos": 16359680989440.0,
      "grad_norm": 4.247101649933086,
      "language_loss": 0.67918015,
      "learning_rate": 1.6648262061283492e-07,
      "loss": 0.70077455,
      "num_input_tokens_seen": 157089350,
      "step": 7261,
      "time_per_iteration": 2.42807936668396
    },
    {
      "auxiliary_loss_clip": 0.01121677,
      "auxiliary_loss_mlp": 0.01024106,
      "balance_loss_clip": 1.03930736,
      "balance_loss_mlp": 1.01723695,
      "epoch": 0.8732038718210786,
      "flos": 21215126868480.0,
      "grad_norm": 1.877826568790639,
      "language_loss": 0.73299396,
      "learning_rate": 1.6617160609588353e-07,
      "loss": 0.75445175,
      "num_input_tokens_seen": 157108525,
      "step": 7262,
      "time_per_iteration": 2.4914398193359375
    },
    {
      "auxiliary_loss_clip": 0.01142879,
      "auxiliary_loss_mlp": 0.01027586,
      "balance_loss_clip": 1.04569161,
      "balance_loss_mlp": 1.01995111,
      "epoch": 0.8733241147117177,
      "flos": 16610696208000.0,
      "grad_norm": 4.5376051386047065,
      "language_loss": 0.71651846,
      "learning_rate": 1.6586086976924163e-07,
      "loss": 0.73822308,
      "num_input_tokens_seen": 157124025,
      "step": 7263,
      "time_per_iteration": 2.4247589111328125
    },
    {
      "auxiliary_loss_clip": 0.01150642,
      "auxiliary_loss_mlp": 0.01021705,
      "balance_loss_clip": 1.04396796,
      "balance_loss_mlp": 1.01519895,
      "epoch": 0.8734443576023567,
      "flos": 20193935207040.0,
      "grad_norm": 1.936928895674092,
      "language_loss": 0.7832166,
      "learning_rate": 1.6555041168004747e-07,
      "loss": 0.80494004,
      "num_input_tokens_seen": 157143345,
      "step": 7264,
      "time_per_iteration": 2.4338228702545166
    },
    {
      "auxiliary_loss_clip": 0.01130214,
      "auxiliary_loss_mlp": 0.01022533,
      "balance_loss_clip": 1.04194391,
      "balance_loss_mlp": 1.01580715,
      "epoch": 0.8735646004929959,
      "flos": 18041162411520.0,
      "grad_norm": 1.7579067613630046,
      "language_loss": 0.69390976,
      "learning_rate": 1.6524023187539715e-07,
      "loss": 0.71543723,
      "num_input_tokens_seen": 157161630,
      "step": 7265,
      "time_per_iteration": 2.4513816833496094
    },
    {
      "auxiliary_loss_clip": 0.01137891,
      "auxiliary_loss_mlp": 0.01023288,
      "balance_loss_clip": 1.04442573,
      "balance_loss_mlp": 1.01650167,
      "epoch": 0.873684843383635,
      "flos": 20262344659200.0,
      "grad_norm": 2.0068267346133606,
      "language_loss": 0.74740982,
      "learning_rate": 1.649303304023446e-07,
      "loss": 0.76902163,
      "num_input_tokens_seen": 157181385,
      "step": 7266,
      "time_per_iteration": 2.471242666244507
    },
    {
      "auxiliary_loss_clip": 0.011177,
      "auxiliary_loss_mlp": 0.01022204,
      "balance_loss_clip": 1.04360771,
      "balance_loss_mlp": 1.01524258,
      "epoch": 0.873805086274274,
      "flos": 16947287579520.0,
      "grad_norm": 1.8667044926312626,
      "language_loss": 0.78711295,
      "learning_rate": 1.6462070730790246e-07,
      "loss": 0.80851197,
      "num_input_tokens_seen": 157200545,
      "step": 7267,
      "time_per_iteration": 2.478363275527954
    },
    {
      "auxiliary_loss_clip": 0.01132664,
      "auxiliary_loss_mlp": 0.01024071,
      "balance_loss_clip": 1.0407896,
      "balance_loss_mlp": 1.01688588,
      "epoch": 0.8739253291649132,
      "flos": 18041270152320.0,
      "grad_norm": 2.6493317700626657,
      "language_loss": 0.7863695,
      "learning_rate": 1.6431136263903912e-07,
      "loss": 0.80793685,
      "num_input_tokens_seen": 157219545,
      "step": 7268,
      "time_per_iteration": 4.352779865264893
    },
    {
      "auxiliary_loss_clip": 0.01153717,
      "auxiliary_loss_mlp": 0.00761389,
      "balance_loss_clip": 1.04387498,
      "balance_loss_mlp": 1.00046408,
      "epoch": 0.8740455720555522,
      "flos": 21325085377920.0,
      "grad_norm": 2.452105333380181,
      "language_loss": 0.73743081,
      "learning_rate": 1.6400229644268282e-07,
      "loss": 0.7565819,
      "num_input_tokens_seen": 157237900,
      "step": 7269,
      "time_per_iteration": 2.4299886226654053
    },
    {
      "auxiliary_loss_clip": 0.01118299,
      "auxiliary_loss_mlp": 0.01028769,
      "balance_loss_clip": 1.04583395,
      "balance_loss_mlp": 1.0210557,
      "epoch": 0.8741658149461913,
      "flos": 15158684822400.0,
      "grad_norm": 1.8136251130857342,
      "language_loss": 0.8119356,
      "learning_rate": 1.6369350876571852e-07,
      "loss": 0.83340633,
      "num_input_tokens_seen": 157256055,
      "step": 7270,
      "time_per_iteration": 3.328937292098999
    },
    {
      "auxiliary_loss_clip": 0.01105456,
      "auxiliary_loss_mlp": 0.01023493,
      "balance_loss_clip": 1.0391233,
      "balance_loss_mlp": 1.01657021,
      "epoch": 0.8742860578368304,
      "flos": 23039855729280.0,
      "grad_norm": 1.8416639093662597,
      "language_loss": 0.81421393,
      "learning_rate": 1.6338499965498874e-07,
      "loss": 0.83550346,
      "num_input_tokens_seen": 157274785,
      "step": 7271,
      "time_per_iteration": 2.5355899333953857
    },
    {
      "auxiliary_loss_clip": 0.01120467,
      "auxiliary_loss_mlp": 0.01030033,
      "balance_loss_clip": 1.04358613,
      "balance_loss_mlp": 1.02230835,
      "epoch": 0.8744063007274695,
      "flos": 28145347159680.0,
      "grad_norm": 1.439658537946952,
      "language_loss": 0.77447951,
      "learning_rate": 1.630767691572943e-07,
      "loss": 0.79598451,
      "num_input_tokens_seen": 157294805,
      "step": 7272,
      "time_per_iteration": 2.5608465671539307
    },
    {
      "auxiliary_loss_clip": 0.01042944,
      "auxiliary_loss_mlp": 0.0100128,
      "balance_loss_clip": 1.00839281,
      "balance_loss_mlp": 1.00034416,
      "epoch": 0.8745265436181086,
      "flos": 64034076654720.0,
      "grad_norm": 0.7358329192011749,
      "language_loss": 0.53526986,
      "learning_rate": 1.6276881731939306e-07,
      "loss": 0.5557121,
      "num_input_tokens_seen": 157356695,
      "step": 7273,
      "time_per_iteration": 3.1215240955352783
    },
    {
      "auxiliary_loss_clip": 0.01146961,
      "auxiliary_loss_mlp": 0.0102357,
      "balance_loss_clip": 1.04521716,
      "balance_loss_mlp": 1.01659966,
      "epoch": 0.8746467865087477,
      "flos": 28658618553600.0,
      "grad_norm": 1.852533439932542,
      "language_loss": 0.75572109,
      "learning_rate": 1.6246114418800193e-07,
      "loss": 0.77742648,
      "num_input_tokens_seen": 157376975,
      "step": 7274,
      "time_per_iteration": 2.499825954437256
    },
    {
      "auxiliary_loss_clip": 0.01145811,
      "auxiliary_loss_mlp": 0.01026986,
      "balance_loss_clip": 1.04466081,
      "balance_loss_mlp": 1.01937413,
      "epoch": 0.8747670293993868,
      "flos": 23985850268160.0,
      "grad_norm": 1.7093377120631499,
      "language_loss": 0.76604038,
      "learning_rate": 1.6215374980979423e-07,
      "loss": 0.78776836,
      "num_input_tokens_seen": 157397385,
      "step": 7275,
      "time_per_iteration": 2.4932079315185547
    },
    {
      "auxiliary_loss_clip": 0.01145528,
      "auxiliary_loss_mlp": 0.01027972,
      "balance_loss_clip": 1.04585576,
      "balance_loss_mlp": 1.02143073,
      "epoch": 0.8748872722900258,
      "flos": 45221624478720.0,
      "grad_norm": 1.9740980187618742,
      "language_loss": 0.68492776,
      "learning_rate": 1.6184663423140133e-07,
      "loss": 0.70666277,
      "num_input_tokens_seen": 157417685,
      "step": 7276,
      "time_per_iteration": 3.3338396549224854
    },
    {
      "auxiliary_loss_clip": 0.01112649,
      "auxiliary_loss_mlp": 0.0102933,
      "balance_loss_clip": 1.04239416,
      "balance_loss_mlp": 1.02246654,
      "epoch": 0.875007515180665,
      "flos": 19754280737280.0,
      "grad_norm": 1.7860318129655117,
      "language_loss": 0.63926178,
      "learning_rate": 1.615397974994126e-07,
      "loss": 0.66068155,
      "num_input_tokens_seen": 157435490,
      "step": 7277,
      "time_per_iteration": 2.5295870304107666
    },
    {
      "auxiliary_loss_clip": 0.01162282,
      "auxiliary_loss_mlp": 0.01022699,
      "balance_loss_clip": 1.04633284,
      "balance_loss_mlp": 1.01594031,
      "epoch": 0.875127758071304,
      "flos": 22710734386560.0,
      "grad_norm": 1.4724612628082139,
      "language_loss": 0.80776554,
      "learning_rate": 1.6123323966037438e-07,
      "loss": 0.82961535,
      "num_input_tokens_seen": 157454010,
      "step": 7278,
      "time_per_iteration": 2.4248545169830322
    },
    {
      "auxiliary_loss_clip": 0.01165152,
      "auxiliary_loss_mlp": 0.01031969,
      "balance_loss_clip": 1.04795313,
      "balance_loss_mlp": 1.02484655,
      "epoch": 0.8752480009619431,
      "flos": 23403846199680.0,
      "grad_norm": 2.220548201688487,
      "language_loss": 0.78453571,
      "learning_rate": 1.6092696076079216e-07,
      "loss": 0.80650693,
      "num_input_tokens_seen": 157472385,
      "step": 7279,
      "time_per_iteration": 2.405405044555664
    },
    {
      "auxiliary_loss_clip": 0.01111073,
      "auxiliary_loss_mlp": 0.01022329,
      "balance_loss_clip": 1.04066515,
      "balance_loss_mlp": 1.01545322,
      "epoch": 0.8753682438525822,
      "flos": 26213101914240.0,
      "grad_norm": 1.5701699245583525,
      "language_loss": 0.74108648,
      "learning_rate": 1.6062096084712785e-07,
      "loss": 0.76242048,
      "num_input_tokens_seen": 157493735,
      "step": 7280,
      "time_per_iteration": 2.5235345363616943
    },
    {
      "auxiliary_loss_clip": 0.01127196,
      "auxiliary_loss_mlp": 0.00761459,
      "balance_loss_clip": 1.03945374,
      "balance_loss_mlp": 1.00047183,
      "epoch": 0.8754884867432213,
      "flos": 23326745656320.0,
      "grad_norm": 2.105885957597929,
      "language_loss": 0.70599437,
      "learning_rate": 1.6031523996580098e-07,
      "loss": 0.72488093,
      "num_input_tokens_seen": 157511295,
      "step": 7281,
      "time_per_iteration": 2.472104549407959
    },
    {
      "auxiliary_loss_clip": 0.01131526,
      "auxiliary_loss_mlp": 0.01026154,
      "balance_loss_clip": 1.04429686,
      "balance_loss_mlp": 1.01880455,
      "epoch": 0.8756087296338604,
      "flos": 12495226412160.0,
      "grad_norm": 2.5097686155391026,
      "language_loss": 0.66421473,
      "learning_rate": 1.6000979816318981e-07,
      "loss": 0.68579155,
      "num_input_tokens_seen": 157529760,
      "step": 7282,
      "time_per_iteration": 2.4813733100891113
    },
    {
      "auxiliary_loss_clip": 0.01144003,
      "auxiliary_loss_mlp": 0.01020992,
      "balance_loss_clip": 1.04466677,
      "balance_loss_mlp": 1.01388156,
      "epoch": 0.8757289725244994,
      "flos": 18952898353920.0,
      "grad_norm": 2.1019711633317932,
      "language_loss": 0.75089264,
      "learning_rate": 1.5970463548562886e-07,
      "loss": 0.7725426,
      "num_input_tokens_seen": 157548915,
      "step": 7283,
      "time_per_iteration": 2.4103548526763916
    },
    {
      "auxiliary_loss_clip": 0.01135534,
      "auxiliary_loss_mlp": 0.01023007,
      "balance_loss_clip": 1.0453856,
      "balance_loss_mlp": 1.01614642,
      "epoch": 0.8758492154151386,
      "flos": 25265958140160.0,
      "grad_norm": 1.794976923770229,
      "language_loss": 0.70912701,
      "learning_rate": 1.5939975197941192e-07,
      "loss": 0.73071241,
      "num_input_tokens_seen": 157570570,
      "step": 7284,
      "time_per_iteration": 2.501824378967285
    },
    {
      "auxiliary_loss_clip": 0.01044147,
      "auxiliary_loss_mlp": 0.01000643,
      "balance_loss_clip": 1.00908566,
      "balance_loss_mlp": 0.99974883,
      "epoch": 0.8759694583057777,
      "flos": 65571664193280.0,
      "grad_norm": 0.8480058578704652,
      "language_loss": 0.53354669,
      "learning_rate": 1.5909514769078892e-07,
      "loss": 0.55399466,
      "num_input_tokens_seen": 157635675,
      "step": 7285,
      "time_per_iteration": 3.11342453956604
    },
    {
      "auxiliary_loss_clip": 0.01115919,
      "auxiliary_loss_mlp": 0.01025474,
      "balance_loss_clip": 1.04525328,
      "balance_loss_mlp": 1.01875901,
      "epoch": 0.8760897011964167,
      "flos": 25446193608960.0,
      "grad_norm": 1.5350151270816048,
      "language_loss": 0.77772117,
      "learning_rate": 1.5879082266596867e-07,
      "loss": 0.79913509,
      "num_input_tokens_seen": 157657015,
      "step": 7286,
      "time_per_iteration": 2.5437259674072266
    },
    {
      "auxiliary_loss_clip": 0.01128602,
      "auxiliary_loss_mlp": 0.01022283,
      "balance_loss_clip": 1.03911042,
      "balance_loss_mlp": 1.01504064,
      "epoch": 0.8762099440870559,
      "flos": 28984830894720.0,
      "grad_norm": 1.832594416387367,
      "language_loss": 0.71956927,
      "learning_rate": 1.5848677695111645e-07,
      "loss": 0.74107808,
      "num_input_tokens_seen": 157678615,
      "step": 7287,
      "time_per_iteration": 2.5251834392547607
    },
    {
      "auxiliary_loss_clip": 0.01127496,
      "auxiliary_loss_mlp": 0.01024736,
      "balance_loss_clip": 1.04283404,
      "balance_loss_mlp": 1.01744914,
      "epoch": 0.8763301869776949,
      "flos": 21609461352960.0,
      "grad_norm": 2.254454354289077,
      "language_loss": 0.69463563,
      "learning_rate": 1.5818301059235562e-07,
      "loss": 0.71615803,
      "num_input_tokens_seen": 157693790,
      "step": 7288,
      "time_per_iteration": 2.4859535694122314
    },
    {
      "auxiliary_loss_clip": 0.01137619,
      "auxiliary_loss_mlp": 0.01020425,
      "balance_loss_clip": 1.04543161,
      "balance_loss_mlp": 1.01324582,
      "epoch": 0.876450429868334,
      "flos": 24644416176000.0,
      "grad_norm": 1.8075034109766241,
      "language_loss": 0.813995,
      "learning_rate": 1.578795236357684e-07,
      "loss": 0.83557546,
      "num_input_tokens_seen": 157715255,
      "step": 7289,
      "time_per_iteration": 2.508932590484619
    },
    {
      "auxiliary_loss_clip": 0.0113726,
      "auxiliary_loss_mlp": 0.01022899,
      "balance_loss_clip": 1.04508615,
      "balance_loss_mlp": 1.01613379,
      "epoch": 0.8765706727589732,
      "flos": 20260046188800.0,
      "grad_norm": 3.5040755570939806,
      "language_loss": 0.85885584,
      "learning_rate": 1.5757631612739218e-07,
      "loss": 0.8804574,
      "num_input_tokens_seen": 157728800,
      "step": 7290,
      "time_per_iteration": 2.4742484092712402
    },
    {
      "auxiliary_loss_clip": 0.01061895,
      "auxiliary_loss_mlp": 0.01001848,
      "balance_loss_clip": 1.0079484,
      "balance_loss_mlp": 1.00092435,
      "epoch": 0.8766909156496122,
      "flos": 71371165276800.0,
      "grad_norm": 0.7815669485717385,
      "language_loss": 0.61447883,
      "learning_rate": 1.572733881132242e-07,
      "loss": 0.63511622,
      "num_input_tokens_seen": 157789445,
      "step": 7291,
      "time_per_iteration": 3.063943386077881
    },
    {
      "auxiliary_loss_clip": 0.01028644,
      "auxiliary_loss_mlp": 0.01002972,
      "balance_loss_clip": 1.01076031,
      "balance_loss_mlp": 1.00190496,
      "epoch": 0.8768111585402513,
      "flos": 69523490603520.0,
      "grad_norm": 0.7819245990748084,
      "language_loss": 0.58509946,
      "learning_rate": 1.5697073963921814e-07,
      "loss": 0.60541558,
      "num_input_tokens_seen": 157848685,
      "step": 7292,
      "time_per_iteration": 3.016230344772339
    },
    {
      "auxiliary_loss_clip": 0.01151952,
      "auxiliary_loss_mlp": 0.01018753,
      "balance_loss_clip": 1.04669309,
      "balance_loss_mlp": 1.01141024,
      "epoch": 0.8769314014308904,
      "flos": 18838558385280.0,
      "grad_norm": 2.264413964133642,
      "language_loss": 0.84785247,
      "learning_rate": 1.566683707512857e-07,
      "loss": 0.86955953,
      "num_input_tokens_seen": 157866360,
      "step": 7293,
      "time_per_iteration": 2.421010971069336
    },
    {
      "auxiliary_loss_clip": 0.0113355,
      "auxiliary_loss_mlp": 0.01028772,
      "balance_loss_clip": 1.04351854,
      "balance_loss_mlp": 1.02105916,
      "epoch": 0.8770516443215295,
      "flos": 14976402278400.0,
      "grad_norm": 1.9243735443994014,
      "language_loss": 0.79577112,
      "learning_rate": 1.5636628149529553e-07,
      "loss": 0.81739438,
      "num_input_tokens_seen": 157884150,
      "step": 7294,
      "time_per_iteration": 3.179569959640503
    },
    {
      "auxiliary_loss_clip": 0.01133557,
      "auxiliary_loss_mlp": 0.01022716,
      "balance_loss_clip": 1.04266167,
      "balance_loss_mlp": 1.01615024,
      "epoch": 0.8771718872121685,
      "flos": 31649654021760.0,
      "grad_norm": 2.573076373108681,
      "language_loss": 0.79259431,
      "learning_rate": 1.560644719170743e-07,
      "loss": 0.81415707,
      "num_input_tokens_seen": 157905020,
      "step": 7295,
      "time_per_iteration": 3.405374765396118
    },
    {
      "auxiliary_loss_clip": 0.01120247,
      "auxiliary_loss_mlp": 0.01025591,
      "balance_loss_clip": 1.04078662,
      "balance_loss_mlp": 1.01785469,
      "epoch": 0.8772921301028077,
      "flos": 36095466222720.0,
      "grad_norm": 1.7139196897832787,
      "language_loss": 0.72097123,
      "learning_rate": 1.5576294206240692e-07,
      "loss": 0.74242961,
      "num_input_tokens_seen": 157924545,
      "step": 7296,
      "time_per_iteration": 2.63200306892395
    },
    {
      "auxiliary_loss_clip": 0.01134486,
      "auxiliary_loss_mlp": 0.01024436,
      "balance_loss_clip": 1.04464006,
      "balance_loss_mlp": 1.01770353,
      "epoch": 0.8774123729934468,
      "flos": 57116961849600.0,
      "grad_norm": 1.599752716293309,
      "language_loss": 0.67424309,
      "learning_rate": 1.5546169197703507e-07,
      "loss": 0.69583237,
      "num_input_tokens_seen": 157950820,
      "step": 7297,
      "time_per_iteration": 3.6132898330688477
    },
    {
      "auxiliary_loss_clip": 0.01141044,
      "auxiliary_loss_mlp": 0.01026423,
      "balance_loss_clip": 1.04180646,
      "balance_loss_mlp": 1.0195744,
      "epoch": 0.8775326158840858,
      "flos": 23914495900800.0,
      "grad_norm": 2.819588830314426,
      "language_loss": 0.77361566,
      "learning_rate": 1.5516072170665774e-07,
      "loss": 0.79529035,
      "num_input_tokens_seen": 157968790,
      "step": 7298,
      "time_per_iteration": 2.4898064136505127
    },
    {
      "auxiliary_loss_clip": 0.01151527,
      "auxiliary_loss_mlp": 0.01020035,
      "balance_loss_clip": 1.04496181,
      "balance_loss_mlp": 1.01342487,
      "epoch": 0.877652858774725,
      "flos": 17123285243520.0,
      "grad_norm": 1.9497699785209313,
      "language_loss": 0.86900139,
      "learning_rate": 1.5486003129693214e-07,
      "loss": 0.89071703,
      "num_input_tokens_seen": 157986155,
      "step": 7299,
      "time_per_iteration": 2.4063479900360107
    },
    {
      "auxiliary_loss_clip": 0.01151676,
      "auxiliary_loss_mlp": 0.01021452,
      "balance_loss_clip": 1.04569602,
      "balance_loss_mlp": 1.01438928,
      "epoch": 0.877773101665364,
      "flos": 16508961912960.0,
      "grad_norm": 1.9450436910258881,
      "language_loss": 0.78161234,
      "learning_rate": 1.545596207934725e-07,
      "loss": 0.80334359,
      "num_input_tokens_seen": 158004640,
      "step": 7300,
      "time_per_iteration": 2.410071611404419
    },
    {
      "auxiliary_loss_clip": 0.01128689,
      "auxiliary_loss_mlp": 0.0102387,
      "balance_loss_clip": 1.04172814,
      "balance_loss_mlp": 1.01659501,
      "epoch": 0.8778933445560031,
      "flos": 22053209973120.0,
      "grad_norm": 1.7879151151134898,
      "language_loss": 0.77681673,
      "learning_rate": 1.5425949024185147e-07,
      "loss": 0.79834229,
      "num_input_tokens_seen": 158024665,
      "step": 7301,
      "time_per_iteration": 2.486607313156128
    },
    {
      "auxiliary_loss_clip": 0.01138102,
      "auxiliary_loss_mlp": 0.01025856,
      "balance_loss_clip": 1.04291797,
      "balance_loss_mlp": 1.01881361,
      "epoch": 0.8780135874466423,
      "flos": 22564757514240.0,
      "grad_norm": 2.075972349259541,
      "language_loss": 0.67669159,
      "learning_rate": 1.5395963968759818e-07,
      "loss": 0.69833124,
      "num_input_tokens_seen": 158044940,
      "step": 7302,
      "time_per_iteration": 3.2607688903808594
    },
    {
      "auxiliary_loss_clip": 0.01137313,
      "auxiliary_loss_mlp": 0.01020815,
      "balance_loss_clip": 1.04224372,
      "balance_loss_mlp": 1.01370716,
      "epoch": 0.8781338303372813,
      "flos": 61531999073280.0,
      "grad_norm": 1.5629112086878276,
      "language_loss": 0.64401269,
      "learning_rate": 1.536600691761998e-07,
      "loss": 0.66559398,
      "num_input_tokens_seen": 158070770,
      "step": 7303,
      "time_per_iteration": 2.8303260803222656
    },
    {
      "auxiliary_loss_clip": 0.01127785,
      "auxiliary_loss_mlp": 0.0102383,
      "balance_loss_clip": 1.04722524,
      "balance_loss_mlp": 1.01747036,
      "epoch": 0.8782540732279204,
      "flos": 22674751937280.0,
      "grad_norm": 2.908706747026936,
      "language_loss": 0.71378052,
      "learning_rate": 1.5336077875310084e-07,
      "loss": 0.73529673,
      "num_input_tokens_seen": 158089995,
      "step": 7304,
      "time_per_iteration": 2.513988971710205
    },
    {
      "auxiliary_loss_clip": 0.01112788,
      "auxiliary_loss_mlp": 0.01021601,
      "balance_loss_clip": 1.04157197,
      "balance_loss_mlp": 1.01475227,
      "epoch": 0.8783743161185595,
      "flos": 16070348937600.0,
      "grad_norm": 1.8643091032824237,
      "language_loss": 0.74091262,
      "learning_rate": 1.5306176846370321e-07,
      "loss": 0.7622565,
      "num_input_tokens_seen": 158108140,
      "step": 7305,
      "time_per_iteration": 2.500054121017456
    },
    {
      "auxiliary_loss_clip": 0.01144422,
      "auxiliary_loss_mlp": 0.01032515,
      "balance_loss_clip": 1.04369247,
      "balance_loss_mlp": 1.0251832,
      "epoch": 0.8784945590091986,
      "flos": 26067879227520.0,
      "grad_norm": 2.006700364624822,
      "language_loss": 0.74064058,
      "learning_rate": 1.5276303835336712e-07,
      "loss": 0.76240993,
      "num_input_tokens_seen": 158128680,
      "step": 7306,
      "time_per_iteration": 2.5232291221618652
    },
    {
      "auxiliary_loss_clip": 0.01053913,
      "auxiliary_loss_mlp": 0.01000644,
      "balance_loss_clip": 1.00853157,
      "balance_loss_mlp": 0.99977398,
      "epoch": 0.8786148018998376,
      "flos": 62720643939840.0,
      "grad_norm": 0.7645305825577966,
      "language_loss": 0.53551519,
      "learning_rate": 1.524645884674094e-07,
      "loss": 0.55606073,
      "num_input_tokens_seen": 158185610,
      "step": 7307,
      "time_per_iteration": 3.014658212661743
    },
    {
      "auxiliary_loss_clip": 0.01164924,
      "auxiliary_loss_mlp": 0.00762419,
      "balance_loss_clip": 1.04495406,
      "balance_loss_mlp": 1.00043738,
      "epoch": 0.8787350447904768,
      "flos": 21652734263040.0,
      "grad_norm": 3.863340952209873,
      "language_loss": 0.79001743,
      "learning_rate": 1.521664188511047e-07,
      "loss": 0.80929089,
      "num_input_tokens_seen": 158205635,
      "step": 7308,
      "time_per_iteration": 2.4349019527435303
    },
    {
      "auxiliary_loss_clip": 0.01139687,
      "auxiliary_loss_mlp": 0.00761752,
      "balance_loss_clip": 1.04875147,
      "balance_loss_mlp": 1.00054598,
      "epoch": 0.8788552876811159,
      "flos": 25478476957440.0,
      "grad_norm": 1.8669475001719467,
      "language_loss": 0.80641675,
      "learning_rate": 1.518685295496851e-07,
      "loss": 0.82543111,
      "num_input_tokens_seen": 158223495,
      "step": 7309,
      "time_per_iteration": 2.5166094303131104
    },
    {
      "auxiliary_loss_clip": 0.0115075,
      "auxiliary_loss_mlp": 0.01022878,
      "balance_loss_clip": 1.04389,
      "balance_loss_mlp": 1.01619339,
      "epoch": 0.8789755305717549,
      "flos": 22310222762880.0,
      "grad_norm": 1.5892033441673044,
      "language_loss": 0.85107481,
      "learning_rate": 1.5157092060833975e-07,
      "loss": 0.87281114,
      "num_input_tokens_seen": 158243145,
      "step": 7310,
      "time_per_iteration": 2.471359968185425
    },
    {
      "auxiliary_loss_clip": 0.01134144,
      "auxiliary_loss_mlp": 0.01018158,
      "balance_loss_clip": 1.04197431,
      "balance_loss_mlp": 1.01142299,
      "epoch": 0.879095773462394,
      "flos": 29310971408640.0,
      "grad_norm": 1.524385952489216,
      "language_loss": 0.65837979,
      "learning_rate": 1.5127359207221658e-07,
      "loss": 0.67990279,
      "num_input_tokens_seen": 158262625,
      "step": 7311,
      "time_per_iteration": 2.5396535396575928
    },
    {
      "auxiliary_loss_clip": 0.01083491,
      "auxiliary_loss_mlp": 0.0102234,
      "balance_loss_clip": 1.03510821,
      "balance_loss_mlp": 1.01461518,
      "epoch": 0.8792160163530331,
      "flos": 16690023394560.0,
      "grad_norm": 1.8527610637499978,
      "language_loss": 0.73353505,
      "learning_rate": 1.5097654398641923e-07,
      "loss": 0.75459343,
      "num_input_tokens_seen": 158280530,
      "step": 7312,
      "time_per_iteration": 2.576570987701416
    },
    {
      "auxiliary_loss_clip": 0.01155488,
      "auxiliary_loss_mlp": 0.01025056,
      "balance_loss_clip": 1.04616594,
      "balance_loss_mlp": 1.01764083,
      "epoch": 0.8793362592436722,
      "flos": 24499301230080.0,
      "grad_norm": 1.5507347272198129,
      "language_loss": 0.73067212,
      "learning_rate": 1.5067977639601014e-07,
      "loss": 0.75247753,
      "num_input_tokens_seen": 158303290,
      "step": 7313,
      "time_per_iteration": 2.491948366165161
    },
    {
      "auxiliary_loss_clip": 0.0113365,
      "auxiliary_loss_mlp": 0.01024339,
      "balance_loss_clip": 1.04443944,
      "balance_loss_mlp": 1.01734138,
      "epoch": 0.8794565021343113,
      "flos": 14538399834240.0,
      "grad_norm": 3.035761442661135,
      "language_loss": 0.71191573,
      "learning_rate": 1.5038328934600864e-07,
      "loss": 0.73349553,
      "num_input_tokens_seen": 158319925,
      "step": 7314,
      "time_per_iteration": 2.4762117862701416
    },
    {
      "auxiliary_loss_clip": 0.01134273,
      "auxiliary_loss_mlp": 0.01025895,
      "balance_loss_clip": 1.04428303,
      "balance_loss_mlp": 1.01915979,
      "epoch": 0.8795767450249504,
      "flos": 39530286224640.0,
      "grad_norm": 2.275108579227439,
      "language_loss": 0.70001376,
      "learning_rate": 1.5008708288139161e-07,
      "loss": 0.72161543,
      "num_input_tokens_seen": 158342285,
      "step": 7315,
      "time_per_iteration": 2.642104387283325
    },
    {
      "auxiliary_loss_clip": 0.01151542,
      "auxiliary_loss_mlp": 0.01025986,
      "balance_loss_clip": 1.04667854,
      "balance_loss_mlp": 1.01880705,
      "epoch": 0.8796969879155895,
      "flos": 22960672197120.0,
      "grad_norm": 1.907857171125672,
      "language_loss": 0.73190546,
      "learning_rate": 1.497911570470931e-07,
      "loss": 0.75368077,
      "num_input_tokens_seen": 158362290,
      "step": 7316,
      "time_per_iteration": 2.471937894821167
    },
    {
      "auxiliary_loss_clip": 0.01112257,
      "auxiliary_loss_mlp": 0.01028597,
      "balance_loss_clip": 1.04162276,
      "balance_loss_mlp": 1.02133727,
      "epoch": 0.8798172308062285,
      "flos": 28362427004160.0,
      "grad_norm": 1.7013615199981986,
      "language_loss": 0.85545564,
      "learning_rate": 1.494955118880048e-07,
      "loss": 0.87686414,
      "num_input_tokens_seen": 158383275,
      "step": 7317,
      "time_per_iteration": 2.5632169246673584
    },
    {
      "auxiliary_loss_clip": 0.01150999,
      "auxiliary_loss_mlp": 0.01025349,
      "balance_loss_clip": 1.04445624,
      "balance_loss_mlp": 1.01834583,
      "epoch": 0.8799374736968677,
      "flos": 23988974751360.0,
      "grad_norm": 1.6072975066153066,
      "language_loss": 0.72797132,
      "learning_rate": 1.4920014744897634e-07,
      "loss": 0.74973476,
      "num_input_tokens_seen": 158402690,
      "step": 7318,
      "time_per_iteration": 2.5129966735839844
    },
    {
      "auxiliary_loss_clip": 0.01127804,
      "auxiliary_loss_mlp": 0.01018451,
      "balance_loss_clip": 1.04282677,
      "balance_loss_mlp": 1.01139379,
      "epoch": 0.8800577165875068,
      "flos": 25630271832960.0,
      "grad_norm": 1.9919640693505132,
      "language_loss": 0.86248398,
      "learning_rate": 1.4890506377481392e-07,
      "loss": 0.88394654,
      "num_input_tokens_seen": 158421780,
      "step": 7319,
      "time_per_iteration": 2.5378336906433105
    },
    {
      "auxiliary_loss_clip": 0.01094993,
      "auxiliary_loss_mlp": 0.01030529,
      "balance_loss_clip": 1.04215813,
      "balance_loss_mlp": 1.02373707,
      "epoch": 0.8801779594781458,
      "flos": 23440331439360.0,
      "grad_norm": 1.598586831054208,
      "language_loss": 0.6399948,
      "learning_rate": 1.486102609102815e-07,
      "loss": 0.66125,
      "num_input_tokens_seen": 158442330,
      "step": 7320,
      "time_per_iteration": 2.613802433013916
    },
    {
      "auxiliary_loss_clip": 0.01129238,
      "auxiliary_loss_mlp": 0.01023758,
      "balance_loss_clip": 1.04291725,
      "balance_loss_mlp": 1.01688814,
      "epoch": 0.880298202368785,
      "flos": 11508580656000.0,
      "grad_norm": 2.4365163959647145,
      "language_loss": 0.8557207,
      "learning_rate": 1.483157389001004e-07,
      "loss": 0.87725067,
      "num_input_tokens_seen": 158459890,
      "step": 7321,
      "time_per_iteration": 3.2902450561523438
    },
    {
      "auxiliary_loss_clip": 0.01136308,
      "auxiliary_loss_mlp": 0.01022919,
      "balance_loss_clip": 1.04140711,
      "balance_loss_mlp": 1.01521778,
      "epoch": 0.880418445259424,
      "flos": 22671447886080.0,
      "grad_norm": 2.066468582129408,
      "language_loss": 0.78696632,
      "learning_rate": 1.4802149778894933e-07,
      "loss": 0.80855858,
      "num_input_tokens_seen": 158478680,
      "step": 7322,
      "time_per_iteration": 2.4976484775543213
    },
    {
      "auxiliary_loss_clip": 0.01140318,
      "auxiliary_loss_mlp": 0.0102217,
      "balance_loss_clip": 1.04095674,
      "balance_loss_mlp": 1.01589322,
      "epoch": 0.8805386881500631,
      "flos": 20522158709760.0,
      "grad_norm": 1.7618588769863543,
      "language_loss": 0.87580609,
      "learning_rate": 1.4772753762146484e-07,
      "loss": 0.89743096,
      "num_input_tokens_seen": 158497935,
      "step": 7323,
      "time_per_iteration": 2.436676502227783
    },
    {
      "auxiliary_loss_clip": 0.01145624,
      "auxiliary_loss_mlp": 0.01019832,
      "balance_loss_clip": 1.0443821,
      "balance_loss_mlp": 1.01225281,
      "epoch": 0.8806589310407023,
      "flos": 36538891620480.0,
      "grad_norm": 1.591509823588181,
      "language_loss": 0.70549273,
      "learning_rate": 1.474338584422401e-07,
      "loss": 0.72714734,
      "num_input_tokens_seen": 158523145,
      "step": 7324,
      "time_per_iteration": 3.442779779434204
    },
    {
      "auxiliary_loss_clip": 0.01145237,
      "auxiliary_loss_mlp": 0.01020684,
      "balance_loss_clip": 1.04453063,
      "balance_loss_mlp": 1.01381409,
      "epoch": 0.8807791739313413,
      "flos": 23440187784960.0,
      "grad_norm": 1.6276190756949633,
      "language_loss": 0.75873756,
      "learning_rate": 1.4714046029582595e-07,
      "loss": 0.78039676,
      "num_input_tokens_seen": 158542210,
      "step": 7325,
      "time_per_iteration": 2.455404758453369
    },
    {
      "auxiliary_loss_clip": 0.01124952,
      "auxiliary_loss_mlp": 0.01022688,
      "balance_loss_clip": 1.0425694,
      "balance_loss_mlp": 1.01545227,
      "epoch": 0.8808994168219804,
      "flos": 25956843310080.0,
      "grad_norm": 1.818360019062367,
      "language_loss": 0.75533503,
      "learning_rate": 1.46847343226731e-07,
      "loss": 0.77681148,
      "num_input_tokens_seen": 158563250,
      "step": 7326,
      "time_per_iteration": 2.552887439727783
    },
    {
      "auxiliary_loss_clip": 0.01151537,
      "auxiliary_loss_mlp": 0.01023485,
      "balance_loss_clip": 1.04391599,
      "balance_loss_mlp": 1.01612389,
      "epoch": 0.8810196597126195,
      "flos": 17092079303040.0,
      "grad_norm": 2.0285178791236946,
      "language_loss": 0.69577301,
      "learning_rate": 1.465545072794203e-07,
      "loss": 0.71752322,
      "num_input_tokens_seen": 158581125,
      "step": 7327,
      "time_per_iteration": 2.424048900604248
    },
    {
      "auxiliary_loss_clip": 0.01107508,
      "auxiliary_loss_mlp": 0.01023891,
      "balance_loss_clip": 1.04375744,
      "balance_loss_mlp": 1.01673532,
      "epoch": 0.8811399026032586,
      "flos": 23002831785600.0,
      "grad_norm": 2.5769298564056418,
      "language_loss": 0.75683004,
      "learning_rate": 1.4626195249831774e-07,
      "loss": 0.778144,
      "num_input_tokens_seen": 158602025,
      "step": 7328,
      "time_per_iteration": 2.5792641639709473
    },
    {
      "auxiliary_loss_clip": 0.01148546,
      "auxiliary_loss_mlp": 0.01021431,
      "balance_loss_clip": 1.04422379,
      "balance_loss_mlp": 1.0146066,
      "epoch": 0.8812601454938976,
      "flos": 14463813242880.0,
      "grad_norm": 1.9528449440150601,
      "language_loss": 0.71760786,
      "learning_rate": 1.4596967892780244e-07,
      "loss": 0.73930764,
      "num_input_tokens_seen": 158618355,
      "step": 7329,
      "time_per_iteration": 3.194915771484375
    },
    {
      "auxiliary_loss_clip": 0.01162686,
      "auxiliary_loss_mlp": 0.01023766,
      "balance_loss_clip": 1.04621303,
      "balance_loss_mlp": 1.01692367,
      "epoch": 0.8813803883845368,
      "flos": 22493223578880.0,
      "grad_norm": 1.6494222605649067,
      "language_loss": 0.74884796,
      "learning_rate": 1.4567768661221314e-07,
      "loss": 0.77071249,
      "num_input_tokens_seen": 158638925,
      "step": 7330,
      "time_per_iteration": 2.4850776195526123
    },
    {
      "auxiliary_loss_clip": 0.01155302,
      "auxiliary_loss_mlp": 0.00761926,
      "balance_loss_clip": 1.04727781,
      "balance_loss_mlp": 1.00046337,
      "epoch": 0.8815006312751759,
      "flos": 21506901045120.0,
      "grad_norm": 1.9400371862733354,
      "language_loss": 0.74573803,
      "learning_rate": 1.4538597559584442e-07,
      "loss": 0.76491034,
      "num_input_tokens_seen": 158656715,
      "step": 7331,
      "time_per_iteration": 2.478764295578003
    },
    {
      "auxiliary_loss_clip": 0.01133124,
      "auxiliary_loss_mlp": 0.01025317,
      "balance_loss_clip": 1.04287696,
      "balance_loss_mlp": 1.01793778,
      "epoch": 0.8816208741658149,
      "flos": 22784566792320.0,
      "grad_norm": 2.3234348253468395,
      "language_loss": 0.79236937,
      "learning_rate": 1.4509454592294823e-07,
      "loss": 0.81395376,
      "num_input_tokens_seen": 158677200,
      "step": 7332,
      "time_per_iteration": 2.5042765140533447
    },
    {
      "auxiliary_loss_clip": 0.01126054,
      "auxiliary_loss_mlp": 0.00761949,
      "balance_loss_clip": 1.04570723,
      "balance_loss_mlp": 1.00043428,
      "epoch": 0.8817411170564541,
      "flos": 17779409026560.0,
      "grad_norm": 1.9214944616600578,
      "language_loss": 0.79064822,
      "learning_rate": 1.448033976377354e-07,
      "loss": 0.80952829,
      "num_input_tokens_seen": 158692185,
      "step": 7333,
      "time_per_iteration": 2.497204303741455
    },
    {
      "auxiliary_loss_clip": 0.01151412,
      "auxiliary_loss_mlp": 0.01020055,
      "balance_loss_clip": 1.0434835,
      "balance_loss_mlp": 1.01328373,
      "epoch": 0.8818613599470931,
      "flos": 18551812112640.0,
      "grad_norm": 1.8108215662815614,
      "language_loss": 0.7412011,
      "learning_rate": 1.445125307843713e-07,
      "loss": 0.76291573,
      "num_input_tokens_seen": 158710410,
      "step": 7334,
      "time_per_iteration": 2.422093629837036
    },
    {
      "auxiliary_loss_clip": 0.01149064,
      "auxiliary_loss_mlp": 0.01021286,
      "balance_loss_clip": 1.04654717,
      "balance_loss_mlp": 1.01507831,
      "epoch": 0.8819816028377322,
      "flos": 27599792417280.0,
      "grad_norm": 1.5808760505769883,
      "language_loss": 0.75867236,
      "learning_rate": 1.442219454069813e-07,
      "loss": 0.78037584,
      "num_input_tokens_seen": 158731435,
      "step": 7335,
      "time_per_iteration": 2.5133044719696045
    },
    {
      "auxiliary_loss_clip": 0.01110849,
      "auxiliary_loss_mlp": 0.01026464,
      "balance_loss_clip": 1.0421983,
      "balance_loss_mlp": 1.0197438,
      "epoch": 0.8821018457283714,
      "flos": 23404600385280.0,
      "grad_norm": 2.1464745733324277,
      "language_loss": 0.66530502,
      "learning_rate": 1.4393164154964676e-07,
      "loss": 0.68667817,
      "num_input_tokens_seen": 158750965,
      "step": 7336,
      "time_per_iteration": 2.555781364440918
    },
    {
      "auxiliary_loss_clip": 0.01150348,
      "auxiliary_loss_mlp": 0.01025165,
      "balance_loss_clip": 1.04768014,
      "balance_loss_mlp": 1.01837564,
      "epoch": 0.8822220886190104,
      "flos": 29132459792640.0,
      "grad_norm": 2.126371595911774,
      "language_loss": 0.94086659,
      "learning_rate": 1.4364161925640649e-07,
      "loss": 0.96262175,
      "num_input_tokens_seen": 158772365,
      "step": 7337,
      "time_per_iteration": 2.5190610885620117
    },
    {
      "auxiliary_loss_clip": 0.01163945,
      "auxiliary_loss_mlp": 0.01022252,
      "balance_loss_clip": 1.04698277,
      "balance_loss_mlp": 1.01548684,
      "epoch": 0.8823423315096495,
      "flos": 20485422074880.0,
      "grad_norm": 1.8771424008116642,
      "language_loss": 0.85013425,
      "learning_rate": 1.4335187857125663e-07,
      "loss": 0.87199628,
      "num_input_tokens_seen": 158791065,
      "step": 7338,
      "time_per_iteration": 2.4469211101531982
    },
    {
      "auxiliary_loss_clip": 0.01151542,
      "auxiliary_loss_mlp": 0.01020202,
      "balance_loss_clip": 1.04482007,
      "balance_loss_mlp": 1.01351452,
      "epoch": 0.8824625744002886,
      "flos": 24206377818240.0,
      "grad_norm": 1.707140800204605,
      "language_loss": 0.75492656,
      "learning_rate": 1.4306241953815023e-07,
      "loss": 0.77664405,
      "num_input_tokens_seen": 158812125,
      "step": 7339,
      "time_per_iteration": 2.481071949005127
    },
    {
      "auxiliary_loss_clip": 0.01152356,
      "auxiliary_loss_mlp": 0.01021534,
      "balance_loss_clip": 1.04634786,
      "balance_loss_mlp": 1.01454782,
      "epoch": 0.8825828172909277,
      "flos": 24679500785280.0,
      "grad_norm": 1.8240121142624837,
      "language_loss": 0.70930111,
      "learning_rate": 1.4277324220099862e-07,
      "loss": 0.73104006,
      "num_input_tokens_seen": 158834035,
      "step": 7340,
      "time_per_iteration": 2.4850611686706543
    },
    {
      "auxiliary_loss_clip": 0.01116506,
      "auxiliary_loss_mlp": 0.010231,
      "balance_loss_clip": 1.04049945,
      "balance_loss_mlp": 1.01630521,
      "epoch": 0.8827030601815667,
      "flos": 22456163721600.0,
      "grad_norm": 1.9610634645826541,
      "language_loss": 0.7435261,
      "learning_rate": 1.4248434660366938e-07,
      "loss": 0.76492214,
      "num_input_tokens_seen": 158853510,
      "step": 7341,
      "time_per_iteration": 2.502406358718872
    },
    {
      "auxiliary_loss_clip": 0.01134861,
      "auxiliary_loss_mlp": 0.01025517,
      "balance_loss_clip": 1.04488158,
      "balance_loss_mlp": 1.01854336,
      "epoch": 0.8828233030722058,
      "flos": 19865639877120.0,
      "grad_norm": 2.2529110446295153,
      "language_loss": 0.70449144,
      "learning_rate": 1.4219573278998808e-07,
      "loss": 0.7260952,
      "num_input_tokens_seen": 158871970,
      "step": 7342,
      "time_per_iteration": 2.465952157974243
    },
    {
      "auxiliary_loss_clip": 0.01134606,
      "auxiliary_loss_mlp": 0.01023129,
      "balance_loss_clip": 1.04135799,
      "balance_loss_mlp": 1.01561308,
      "epoch": 0.882943545962845,
      "flos": 39347213581440.0,
      "grad_norm": 2.4144026349215286,
      "language_loss": 0.65100288,
      "learning_rate": 1.4190740080373685e-07,
      "loss": 0.67258024,
      "num_input_tokens_seen": 158892250,
      "step": 7343,
      "time_per_iteration": 2.651212453842163
    },
    {
      "auxiliary_loss_clip": 0.01108483,
      "auxiliary_loss_mlp": 0.01024688,
      "balance_loss_clip": 1.04243755,
      "balance_loss_mlp": 1.01709437,
      "epoch": 0.883063788853484,
      "flos": 19054524908160.0,
      "grad_norm": 1.9286691086263699,
      "language_loss": 0.84236878,
      "learning_rate": 1.4161935068865538e-07,
      "loss": 0.86370045,
      "num_input_tokens_seen": 158907395,
      "step": 7344,
      "time_per_iteration": 2.551642656326294
    },
    {
      "auxiliary_loss_clip": 0.01164032,
      "auxiliary_loss_mlp": 0.01021445,
      "balance_loss_clip": 1.0461843,
      "balance_loss_mlp": 1.01421499,
      "epoch": 0.8831840317441231,
      "flos": 18733196816640.0,
      "grad_norm": 2.1069957650828313,
      "language_loss": 0.75574142,
      "learning_rate": 1.4133158248844113e-07,
      "loss": 0.77759618,
      "num_input_tokens_seen": 158926300,
      "step": 7345,
      "time_per_iteration": 2.3864638805389404
    },
    {
      "auxiliary_loss_clip": 0.01126088,
      "auxiliary_loss_mlp": 0.01023549,
      "balance_loss_clip": 1.04333997,
      "balance_loss_mlp": 1.01577067,
      "epoch": 0.8833042746347622,
      "flos": 26827712553600.0,
      "grad_norm": 2.282310599232799,
      "language_loss": 0.73526591,
      "learning_rate": 1.4104409624674785e-07,
      "loss": 0.75676233,
      "num_input_tokens_seen": 158946085,
      "step": 7346,
      "time_per_iteration": 2.5429930686950684
    },
    {
      "auxiliary_loss_clip": 0.01152918,
      "auxiliary_loss_mlp": 0.01019155,
      "balance_loss_clip": 1.04880488,
      "balance_loss_mlp": 1.01244402,
      "epoch": 0.8834245175254013,
      "flos": 26104077158400.0,
      "grad_norm": 1.8701023507103074,
      "language_loss": 0.78414667,
      "learning_rate": 1.407568920071873e-07,
      "loss": 0.80586743,
      "num_input_tokens_seen": 158964950,
      "step": 7347,
      "time_per_iteration": 2.4870550632476807
    },
    {
      "auxiliary_loss_clip": 0.01169903,
      "auxiliary_loss_mlp": 0.0102551,
      "balance_loss_clip": 1.04780555,
      "balance_loss_mlp": 1.01800549,
      "epoch": 0.8835447604160404,
      "flos": 30629036977920.0,
      "grad_norm": 1.8127838562903202,
      "language_loss": 0.68040216,
      "learning_rate": 1.4046996981332782e-07,
      "loss": 0.70235622,
      "num_input_tokens_seen": 158984835,
      "step": 7348,
      "time_per_iteration": 3.5452449321746826
    },
    {
      "auxiliary_loss_clip": 0.01125377,
      "auxiliary_loss_mlp": 0.01022474,
      "balance_loss_clip": 1.04293323,
      "balance_loss_mlp": 1.01482034,
      "epoch": 0.8836650033066795,
      "flos": 24718356322560.0,
      "grad_norm": 1.8635928875697585,
      "language_loss": 0.77988631,
      "learning_rate": 1.4018332970869516e-07,
      "loss": 0.80136484,
      "num_input_tokens_seen": 159002775,
      "step": 7349,
      "time_per_iteration": 2.5430126190185547
    },
    {
      "auxiliary_loss_clip": 0.01132065,
      "auxiliary_loss_mlp": 0.0102571,
      "balance_loss_clip": 1.04517722,
      "balance_loss_mlp": 1.01832163,
      "epoch": 0.8837852461973186,
      "flos": 25413371556480.0,
      "grad_norm": 1.8206539258283447,
      "language_loss": 0.85138249,
      "learning_rate": 1.398969717367733e-07,
      "loss": 0.87296033,
      "num_input_tokens_seen": 159024100,
      "step": 7350,
      "time_per_iteration": 3.366276264190674
    },
    {
      "auxiliary_loss_clip": 0.01108142,
      "auxiliary_loss_mlp": 0.01025792,
      "balance_loss_clip": 1.04441047,
      "balance_loss_mlp": 1.01903009,
      "epoch": 0.8839054890879576,
      "flos": 17822574195840.0,
      "grad_norm": 2.448918069540013,
      "language_loss": 0.75964421,
      "learning_rate": 1.396108959410014e-07,
      "loss": 0.78098363,
      "num_input_tokens_seen": 159043315,
      "step": 7351,
      "time_per_iteration": 2.516375780105591
    },
    {
      "auxiliary_loss_clip": 0.0115221,
      "auxiliary_loss_mlp": 0.00761962,
      "balance_loss_clip": 1.04712796,
      "balance_loss_mlp": 1.00036573,
      "epoch": 0.8840257319785968,
      "flos": 23769021818880.0,
      "grad_norm": 1.7945940777075668,
      "language_loss": 0.81374007,
      "learning_rate": 1.3932510236477745e-07,
      "loss": 0.83288181,
      "num_input_tokens_seen": 159063985,
      "step": 7352,
      "time_per_iteration": 2.462893486022949
    },
    {
      "auxiliary_loss_clip": 0.0114889,
      "auxiliary_loss_mlp": 0.01025624,
      "balance_loss_clip": 1.04266739,
      "balance_loss_mlp": 1.01796532,
      "epoch": 0.8841459748692359,
      "flos": 29059776622080.0,
      "grad_norm": 1.7131172184681667,
      "language_loss": 0.5630846,
      "learning_rate": 1.3903959105145636e-07,
      "loss": 0.58482969,
      "num_input_tokens_seen": 159084475,
      "step": 7353,
      "time_per_iteration": 2.5107421875
    },
    {
      "auxiliary_loss_clip": 0.01164539,
      "auxiliary_loss_mlp": 0.01023206,
      "balance_loss_clip": 1.04688239,
      "balance_loss_mlp": 1.01627707,
      "epoch": 0.8842662177598749,
      "flos": 24311523905280.0,
      "grad_norm": 2.3033963809454545,
      "language_loss": 0.83300853,
      "learning_rate": 1.387543620443492e-07,
      "loss": 0.85488594,
      "num_input_tokens_seen": 159101320,
      "step": 7354,
      "time_per_iteration": 2.418693780899048
    },
    {
      "auxiliary_loss_clip": 0.01164544,
      "auxiliary_loss_mlp": 0.01024318,
      "balance_loss_clip": 1.04751563,
      "balance_loss_mlp": 1.01733816,
      "epoch": 0.8843864606505141,
      "flos": 25007867942400.0,
      "grad_norm": 1.6967419519665319,
      "language_loss": 0.84051484,
      "learning_rate": 1.3846941538672606e-07,
      "loss": 0.86240351,
      "num_input_tokens_seen": 159120025,
      "step": 7355,
      "time_per_iteration": 2.4436275959014893
    },
    {
      "auxiliary_loss_clip": 0.01112892,
      "auxiliary_loss_mlp": 0.01023019,
      "balance_loss_clip": 1.04326117,
      "balance_loss_mlp": 1.01636386,
      "epoch": 0.8845067035411531,
      "flos": 28183915388160.0,
      "grad_norm": 2.1474783124500516,
      "language_loss": 0.80969095,
      "learning_rate": 1.3818475112181193e-07,
      "loss": 0.8310501,
      "num_input_tokens_seen": 159138820,
      "step": 7356,
      "time_per_iteration": 3.369537830352783
    },
    {
      "auxiliary_loss_clip": 0.01135514,
      "auxiliary_loss_mlp": 0.01023688,
      "balance_loss_clip": 1.04359782,
      "balance_loss_mlp": 1.01730442,
      "epoch": 0.8846269464317922,
      "flos": 12853219311360.0,
      "grad_norm": 2.0950300963596704,
      "language_loss": 0.80003589,
      "learning_rate": 1.3790036929279091e-07,
      "loss": 0.82162786,
      "num_input_tokens_seen": 159155975,
      "step": 7357,
      "time_per_iteration": 2.47066593170166
    },
    {
      "auxiliary_loss_clip": 0.01154914,
      "auxiliary_loss_mlp": 0.00762031,
      "balance_loss_clip": 1.04783964,
      "balance_loss_mlp": 1.00032699,
      "epoch": 0.8847471893224313,
      "flos": 18624351628800.0,
      "grad_norm": 2.2044592804261054,
      "language_loss": 0.5883671,
      "learning_rate": 1.3761626994280363e-07,
      "loss": 0.60753655,
      "num_input_tokens_seen": 159173445,
      "step": 7358,
      "time_per_iteration": 2.4363293647766113
    },
    {
      "auxiliary_loss_clip": 0.01129126,
      "auxiliary_loss_mlp": 0.01023427,
      "balance_loss_clip": 1.04412818,
      "balance_loss_mlp": 1.01628006,
      "epoch": 0.8848674322130704,
      "flos": 35769433449600.0,
      "grad_norm": 1.698533001841016,
      "language_loss": 0.73515761,
      "learning_rate": 1.3733245311494735e-07,
      "loss": 0.75668311,
      "num_input_tokens_seen": 159196100,
      "step": 7359,
      "time_per_iteration": 2.6968295574188232
    },
    {
      "auxiliary_loss_clip": 0.01153576,
      "auxiliary_loss_mlp": 0.01024562,
      "balance_loss_clip": 1.04804122,
      "balance_loss_mlp": 1.01734066,
      "epoch": 0.8849876751037095,
      "flos": 24243760897920.0,
      "grad_norm": 3.9645384812960938,
      "language_loss": 0.70654458,
      "learning_rate": 1.3704891885227676e-07,
      "loss": 0.72832602,
      "num_input_tokens_seen": 159216145,
      "step": 7360,
      "time_per_iteration": 2.474438428878784
    },
    {
      "auxiliary_loss_clip": 0.0112333,
      "auxiliary_loss_mlp": 0.01029616,
      "balance_loss_clip": 1.04123974,
      "balance_loss_mlp": 1.02166462,
      "epoch": 0.8851079179943486,
      "flos": 21500580251520.0,
      "grad_norm": 1.8429630333122524,
      "language_loss": 0.78026068,
      "learning_rate": 1.367656671978037e-07,
      "loss": 0.80179018,
      "num_input_tokens_seen": 159233610,
      "step": 7361,
      "time_per_iteration": 2.5289735794067383
    },
    {
      "auxiliary_loss_clip": 0.01142969,
      "auxiliary_loss_mlp": 0.01025879,
      "balance_loss_clip": 1.04348564,
      "balance_loss_mlp": 1.01915514,
      "epoch": 0.8852281608849877,
      "flos": 15300711198720.0,
      "grad_norm": 1.9542397882552758,
      "language_loss": 0.73432469,
      "learning_rate": 1.36482698194498e-07,
      "loss": 0.75601315,
      "num_input_tokens_seen": 159250155,
      "step": 7362,
      "time_per_iteration": 2.4433975219726562
    },
    {
      "auxiliary_loss_clip": 0.01136594,
      "auxiliary_loss_mlp": 0.01025033,
      "balance_loss_clip": 1.04306769,
      "balance_loss_mlp": 1.01749301,
      "epoch": 0.8853484037756267,
      "flos": 23295719283840.0,
      "grad_norm": 1.7797861929175878,
      "language_loss": 0.71842694,
      "learning_rate": 1.3620001188528506e-07,
      "loss": 0.74004328,
      "num_input_tokens_seen": 159270875,
      "step": 7363,
      "time_per_iteration": 2.519090414047241
    },
    {
      "auxiliary_loss_clip": 0.01153476,
      "auxiliary_loss_mlp": 0.01024235,
      "balance_loss_clip": 1.04402423,
      "balance_loss_mlp": 1.01638174,
      "epoch": 0.8854686466662659,
      "flos": 25114773795840.0,
      "grad_norm": 2.482945675780781,
      "language_loss": 0.73656273,
      "learning_rate": 1.3591760831304865e-07,
      "loss": 0.75833976,
      "num_input_tokens_seen": 159288565,
      "step": 7364,
      "time_per_iteration": 2.475139617919922
    },
    {
      "auxiliary_loss_clip": 0.01163315,
      "auxiliary_loss_mlp": 0.01022319,
      "balance_loss_clip": 1.04613352,
      "balance_loss_mlp": 1.01521993,
      "epoch": 0.885588889556905,
      "flos": 21390873137280.0,
      "grad_norm": 2.3943866196458363,
      "language_loss": 0.7963419,
      "learning_rate": 1.356354875206287e-07,
      "loss": 0.8181982,
      "num_input_tokens_seen": 159306400,
      "step": 7365,
      "time_per_iteration": 2.425713300704956
    },
    {
      "auxiliary_loss_clip": 0.01120097,
      "auxiliary_loss_mlp": 0.01021032,
      "balance_loss_clip": 1.04411805,
      "balance_loss_mlp": 1.01428771,
      "epoch": 0.885709132447544,
      "flos": 26906752431360.0,
      "grad_norm": 2.4027470743625323,
      "language_loss": 0.70194697,
      "learning_rate": 1.3535364955082296e-07,
      "loss": 0.72335827,
      "num_input_tokens_seen": 159326250,
      "step": 7366,
      "time_per_iteration": 2.545585870742798
    },
    {
      "auxiliary_loss_clip": 0.01161817,
      "auxiliary_loss_mlp": 0.01022819,
      "balance_loss_clip": 1.04649854,
      "balance_loss_mlp": 1.01589322,
      "epoch": 0.8858293753381832,
      "flos": 26103394800000.0,
      "grad_norm": 1.7741755039733227,
      "language_loss": 0.64465559,
      "learning_rate": 1.3507209444638613e-07,
      "loss": 0.66650188,
      "num_input_tokens_seen": 159348250,
      "step": 7367,
      "time_per_iteration": 2.4805407524108887
    },
    {
      "auxiliary_loss_clip": 0.01151003,
      "auxiliary_loss_mlp": 0.01023902,
      "balance_loss_clip": 1.04648042,
      "balance_loss_mlp": 1.01688075,
      "epoch": 0.8859496182288222,
      "flos": 23292810282240.0,
      "grad_norm": 2.0168229738428836,
      "language_loss": 0.74196815,
      "learning_rate": 1.347908222500298e-07,
      "loss": 0.76371723,
      "num_input_tokens_seen": 159368325,
      "step": 7368,
      "time_per_iteration": 2.452939510345459
    },
    {
      "auxiliary_loss_clip": 0.01112959,
      "auxiliary_loss_mlp": 0.01026246,
      "balance_loss_clip": 1.04356933,
      "balance_loss_mlp": 1.01949,
      "epoch": 0.8860698611194613,
      "flos": 16872916469760.0,
      "grad_norm": 1.9161946428212586,
      "language_loss": 0.69581848,
      "learning_rate": 1.3450983300442276e-07,
      "loss": 0.71721053,
      "num_input_tokens_seen": 159387555,
      "step": 7369,
      "time_per_iteration": 2.5182132720947266
    },
    {
      "auxiliary_loss_clip": 0.01151776,
      "auxiliary_loss_mlp": 0.0102322,
      "balance_loss_clip": 1.04609036,
      "balance_loss_mlp": 1.01636243,
      "epoch": 0.8861901040101005,
      "flos": 24681404206080.0,
      "grad_norm": 2.098912052055472,
      "language_loss": 0.73768014,
      "learning_rate": 1.3422912675219068e-07,
      "loss": 0.75943005,
      "num_input_tokens_seen": 159407310,
      "step": 7370,
      "time_per_iteration": 2.4848122596740723
    },
    {
      "auxiliary_loss_clip": 0.01161121,
      "auxiliary_loss_mlp": 0.01020819,
      "balance_loss_clip": 1.04725075,
      "balance_loss_mlp": 1.01467657,
      "epoch": 0.8863103469007395,
      "flos": 24423026699520.0,
      "grad_norm": 1.51897907599744,
      "language_loss": 0.79161704,
      "learning_rate": 1.339487035359166e-07,
      "loss": 0.81343639,
      "num_input_tokens_seen": 159427680,
      "step": 7371,
      "time_per_iteration": 2.488556385040283
    },
    {
      "auxiliary_loss_clip": 0.0113892,
      "auxiliary_loss_mlp": 0.00761439,
      "balance_loss_clip": 1.04668522,
      "balance_loss_mlp": 1.00043988,
      "epoch": 0.8864305897913786,
      "flos": 22053964158720.0,
      "grad_norm": 1.5583586568860697,
      "language_loss": 0.84675568,
      "learning_rate": 1.336685633981409e-07,
      "loss": 0.86575925,
      "num_input_tokens_seen": 159448765,
      "step": 7372,
      "time_per_iteration": 2.502751111984253
    },
    {
      "auxiliary_loss_clip": 0.01152084,
      "auxiliary_loss_mlp": 0.01023228,
      "balance_loss_clip": 1.04411387,
      "balance_loss_mlp": 1.01605463,
      "epoch": 0.8865508326820177,
      "flos": 19099449843840.0,
      "grad_norm": 1.759198767596091,
      "language_loss": 0.75007427,
      "learning_rate": 1.333887063813597e-07,
      "loss": 0.77182746,
      "num_input_tokens_seen": 159466870,
      "step": 7373,
      "time_per_iteration": 2.4545724391937256
    },
    {
      "auxiliary_loss_clip": 0.01136512,
      "auxiliary_loss_mlp": 0.01019296,
      "balance_loss_clip": 1.0419085,
      "balance_loss_mlp": 1.01268888,
      "epoch": 0.8866710755726568,
      "flos": 15414189240960.0,
      "grad_norm": 1.8081956465789568,
      "language_loss": 0.65971494,
      "learning_rate": 1.331091325280278e-07,
      "loss": 0.68127298,
      "num_input_tokens_seen": 159485840,
      "step": 7374,
      "time_per_iteration": 2.477057933807373
    },
    {
      "auxiliary_loss_clip": 0.01101656,
      "auxiliary_loss_mlp": 0.01024392,
      "balance_loss_clip": 1.04003167,
      "balance_loss_mlp": 1.01719809,
      "epoch": 0.8867913184632958,
      "flos": 20083689388800.0,
      "grad_norm": 1.629645969604547,
      "language_loss": 0.78244472,
      "learning_rate": 1.3282984188055625e-07,
      "loss": 0.80370522,
      "num_input_tokens_seen": 159505630,
      "step": 7375,
      "time_per_iteration": 4.1883156299591064
    },
    {
      "auxiliary_loss_clip": 0.01162492,
      "auxiliary_loss_mlp": 0.01029274,
      "balance_loss_clip": 1.04465771,
      "balance_loss_mlp": 1.02260971,
      "epoch": 0.8869115613539349,
      "flos": 23365852588800.0,
      "grad_norm": 1.8117834368326777,
      "language_loss": 0.79558909,
      "learning_rate": 1.3255083448131288e-07,
      "loss": 0.81750679,
      "num_input_tokens_seen": 159524675,
      "step": 7376,
      "time_per_iteration": 2.4247028827667236
    },
    {
      "auxiliary_loss_clip": 0.01152487,
      "auxiliary_loss_mlp": 0.010239,
      "balance_loss_clip": 1.04326463,
      "balance_loss_mlp": 1.01706338,
      "epoch": 0.8870318042445741,
      "flos": 21286840371840.0,
      "grad_norm": 1.9723421291595973,
      "language_loss": 0.78709328,
      "learning_rate": 1.3227211037262365e-07,
      "loss": 0.80885714,
      "num_input_tokens_seen": 159541915,
      "step": 7377,
      "time_per_iteration": 3.265744924545288
    },
    {
      "auxiliary_loss_clip": 0.01110145,
      "auxiliary_loss_mlp": 0.01027215,
      "balance_loss_clip": 1.03944755,
      "balance_loss_mlp": 1.01971054,
      "epoch": 0.8871520471352131,
      "flos": 20010862563840.0,
      "grad_norm": 2.6128971238669627,
      "language_loss": 0.85398757,
      "learning_rate": 1.319936695967696e-07,
      "loss": 0.8753612,
      "num_input_tokens_seen": 159559740,
      "step": 7378,
      "time_per_iteration": 2.5648369789123535
    },
    {
      "auxiliary_loss_clip": 0.01172738,
      "auxiliary_loss_mlp": 0.01023659,
      "balance_loss_clip": 1.04862738,
      "balance_loss_mlp": 1.01571369,
      "epoch": 0.8872722900258522,
      "flos": 22601422321920.0,
      "grad_norm": 2.4603420899076847,
      "language_loss": 0.82059735,
      "learning_rate": 1.3171551219599097e-07,
      "loss": 0.84256136,
      "num_input_tokens_seen": 159578265,
      "step": 7379,
      "time_per_iteration": 2.429309368133545
    },
    {
      "auxiliary_loss_clip": 0.01166356,
      "auxiliary_loss_mlp": 0.01023926,
      "balance_loss_clip": 1.04935789,
      "balance_loss_mlp": 1.01661849,
      "epoch": 0.8873925329164913,
      "flos": 22163276223360.0,
      "grad_norm": 10.800634261934333,
      "language_loss": 0.78081715,
      "learning_rate": 1.3143763821248377e-07,
      "loss": 0.80271995,
      "num_input_tokens_seen": 159595350,
      "step": 7380,
      "time_per_iteration": 2.423342704772949
    },
    {
      "auxiliary_loss_clip": 0.01162399,
      "auxiliary_loss_mlp": 0.01025604,
      "balance_loss_clip": 1.04613316,
      "balance_loss_mlp": 1.01890492,
      "epoch": 0.8875127758071304,
      "flos": 19208223204480.0,
      "grad_norm": 2.039672831047613,
      "language_loss": 0.72376847,
      "learning_rate": 1.3116004768840118e-07,
      "loss": 0.7456485,
      "num_input_tokens_seen": 159613725,
      "step": 7381,
      "time_per_iteration": 2.395111560821533
    },
    {
      "auxiliary_loss_clip": 0.01165582,
      "auxiliary_loss_mlp": 0.01027356,
      "balance_loss_clip": 1.04605055,
      "balance_loss_mlp": 1.02000403,
      "epoch": 0.8876330186977694,
      "flos": 18110900666880.0,
      "grad_norm": 1.605186338971091,
      "language_loss": 0.73751259,
      "learning_rate": 1.3088274066585348e-07,
      "loss": 0.75944197,
      "num_input_tokens_seen": 159631335,
      "step": 7382,
      "time_per_iteration": 2.3889708518981934
    },
    {
      "auxiliary_loss_clip": 0.01129574,
      "auxiliary_loss_mlp": 0.01022226,
      "balance_loss_clip": 1.04249358,
      "balance_loss_mlp": 1.01566339,
      "epoch": 0.8877532615884086,
      "flos": 22009434272640.0,
      "grad_norm": 2.141240953957513,
      "language_loss": 0.90490758,
      "learning_rate": 1.3060571718690749e-07,
      "loss": 0.92642558,
      "num_input_tokens_seen": 159648830,
      "step": 7383,
      "time_per_iteration": 3.247978687286377
    },
    {
      "auxiliary_loss_clip": 0.01030585,
      "auxiliary_loss_mlp": 0.00753116,
      "balance_loss_clip": 1.00662804,
      "balance_loss_mlp": 0.99996281,
      "epoch": 0.8878735044790477,
      "flos": 72136924346880.0,
      "grad_norm": 0.740767475928398,
      "language_loss": 0.56882465,
      "learning_rate": 1.3032897729358805e-07,
      "loss": 0.5866617,
      "num_input_tokens_seen": 159709785,
      "step": 7384,
      "time_per_iteration": 3.098950147628784
    },
    {
      "auxiliary_loss_clip": 0.01083905,
      "auxiliary_loss_mlp": 0.00762132,
      "balance_loss_clip": 1.03628862,
      "balance_loss_mlp": 1.00048435,
      "epoch": 0.8879937473696867,
      "flos": 27526355061120.0,
      "grad_norm": 1.8490682792470905,
      "language_loss": 0.799389,
      "learning_rate": 1.3005252102787645e-07,
      "loss": 0.81784934,
      "num_input_tokens_seen": 159728725,
      "step": 7385,
      "time_per_iteration": 2.620107412338257
    },
    {
      "auxiliary_loss_clip": 0.01153436,
      "auxiliary_loss_mlp": 0.01022024,
      "balance_loss_clip": 1.04546702,
      "balance_loss_mlp": 1.01478839,
      "epoch": 0.8881139902603259,
      "flos": 22234091886720.0,
      "grad_norm": 1.5884687384779326,
      "language_loss": 0.73328096,
      "learning_rate": 1.297763484317105e-07,
      "loss": 0.75503558,
      "num_input_tokens_seen": 159747020,
      "step": 7386,
      "time_per_iteration": 2.456625461578369
    },
    {
      "auxiliary_loss_clip": 0.0110674,
      "auxiliary_loss_mlp": 0.00762024,
      "balance_loss_clip": 1.03916097,
      "balance_loss_mlp": 1.00045383,
      "epoch": 0.888234233150965,
      "flos": 20299548170880.0,
      "grad_norm": 2.4678131581554066,
      "language_loss": 0.7072646,
      "learning_rate": 1.2950045954698551e-07,
      "loss": 0.72595221,
      "num_input_tokens_seen": 159764855,
      "step": 7387,
      "time_per_iteration": 2.5199553966522217
    },
    {
      "auxiliary_loss_clip": 0.01116769,
      "auxiliary_loss_mlp": 0.01023508,
      "balance_loss_clip": 1.04328775,
      "balance_loss_mlp": 1.0166651,
      "epoch": 0.888354476041604,
      "flos": 18147996437760.0,
      "grad_norm": 2.566538141348987,
      "language_loss": 0.75402319,
      "learning_rate": 1.2922485441555343e-07,
      "loss": 0.77542597,
      "num_input_tokens_seen": 159783935,
      "step": 7388,
      "time_per_iteration": 2.495147705078125
    },
    {
      "auxiliary_loss_clip": 0.01162016,
      "auxiliary_loss_mlp": 0.0102299,
      "balance_loss_clip": 1.04497838,
      "balance_loss_mlp": 1.01596832,
      "epoch": 0.8884747189322432,
      "flos": 22014282608640.0,
      "grad_norm": 1.746468268030024,
      "language_loss": 0.81769919,
      "learning_rate": 1.2894953307922363e-07,
      "loss": 0.8395493,
      "num_input_tokens_seen": 159802895,
      "step": 7389,
      "time_per_iteration": 2.4161171913146973
    },
    {
      "auxiliary_loss_clip": 0.01118811,
      "auxiliary_loss_mlp": 0.01023653,
      "balance_loss_clip": 1.04299176,
      "balance_loss_mlp": 1.01690578,
      "epoch": 0.8885949618228822,
      "flos": 19786779567360.0,
      "grad_norm": 2.2308530063595366,
      "language_loss": 0.84037381,
      "learning_rate": 1.2867449557976208e-07,
      "loss": 0.86179847,
      "num_input_tokens_seen": 159820995,
      "step": 7390,
      "time_per_iteration": 2.4809858798980713
    },
    {
      "auxiliary_loss_clip": 0.01150604,
      "auxiliary_loss_mlp": 0.01023462,
      "balance_loss_clip": 1.04704762,
      "balance_loss_mlp": 1.01650918,
      "epoch": 0.8887152047135213,
      "flos": 20047599198720.0,
      "grad_norm": 2.4767584532694107,
      "language_loss": 0.7560463,
      "learning_rate": 1.283997419588916e-07,
      "loss": 0.77778697,
      "num_input_tokens_seen": 159840465,
      "step": 7391,
      "time_per_iteration": 2.4775102138519287
    },
    {
      "auxiliary_loss_clip": 0.01154936,
      "auxiliary_loss_mlp": 0.01024917,
      "balance_loss_clip": 1.04622376,
      "balance_loss_mlp": 1.01793742,
      "epoch": 0.8888354476041604,
      "flos": 18588117784320.0,
      "grad_norm": 2.9011225206470748,
      "language_loss": 0.61898756,
      "learning_rate": 1.2812527225829216e-07,
      "loss": 0.64078605,
      "num_input_tokens_seen": 159858690,
      "step": 7392,
      "time_per_iteration": 2.428845167160034
    },
    {
      "auxiliary_loss_clip": 0.01158261,
      "auxiliary_loss_mlp": 0.01022825,
      "balance_loss_clip": 1.04874802,
      "balance_loss_mlp": 1.01490343,
      "epoch": 0.8889556904947995,
      "flos": 21689794120320.0,
      "grad_norm": 1.8064572492466093,
      "language_loss": 0.76296711,
      "learning_rate": 1.2785108651960052e-07,
      "loss": 0.784778,
      "num_input_tokens_seen": 159880325,
      "step": 7393,
      "time_per_iteration": 2.478649616241455
    },
    {
      "auxiliary_loss_clip": 0.01154083,
      "auxiliary_loss_mlp": 0.01024215,
      "balance_loss_clip": 1.0462513,
      "balance_loss_mlp": 1.01708913,
      "epoch": 0.8890759333854386,
      "flos": 27381204201600.0,
      "grad_norm": 1.8777480353814533,
      "language_loss": 0.80651295,
      "learning_rate": 1.2757718478441094e-07,
      "loss": 0.82829589,
      "num_input_tokens_seen": 159901070,
      "step": 7394,
      "time_per_iteration": 2.494816303253174
    },
    {
      "auxiliary_loss_clip": 0.01135999,
      "auxiliary_loss_mlp": 0.01020188,
      "balance_loss_clip": 1.04309082,
      "balance_loss_mlp": 1.01359856,
      "epoch": 0.8891961762760777,
      "flos": 24498834353280.0,
      "grad_norm": 1.698131686081044,
      "language_loss": 0.77360392,
      "learning_rate": 1.2730356709427302e-07,
      "loss": 0.79516578,
      "num_input_tokens_seen": 159919750,
      "step": 7395,
      "time_per_iteration": 2.5131702423095703
    },
    {
      "auxiliary_loss_clip": 0.01151857,
      "auxiliary_loss_mlp": 0.01030085,
      "balance_loss_clip": 1.04872251,
      "balance_loss_mlp": 1.02248871,
      "epoch": 0.8893164191667168,
      "flos": 41499770895360.0,
      "grad_norm": 1.5232726930304756,
      "language_loss": 0.59716618,
      "learning_rate": 1.2703023349069542e-07,
      "loss": 0.61898559,
      "num_input_tokens_seen": 159944600,
      "step": 7396,
      "time_per_iteration": 2.6196460723876953
    },
    {
      "auxiliary_loss_clip": 0.01149084,
      "auxiliary_loss_mlp": 0.01022189,
      "balance_loss_clip": 1.04570353,
      "balance_loss_mlp": 1.0152247,
      "epoch": 0.8894366620573558,
      "flos": 33583623120000.0,
      "grad_norm": 2.018056443920704,
      "language_loss": 0.61910081,
      "learning_rate": 1.2675718401514223e-07,
      "loss": 0.64081353,
      "num_input_tokens_seen": 159968780,
      "step": 7397,
      "time_per_iteration": 2.5523598194122314
    },
    {
      "auxiliary_loss_clip": 0.01136144,
      "auxiliary_loss_mlp": 0.01023348,
      "balance_loss_clip": 1.04382122,
      "balance_loss_mlp": 1.01604617,
      "epoch": 0.889556904947995,
      "flos": 16909832672640.0,
      "grad_norm": 1.876765995548863,
      "language_loss": 0.74329406,
      "learning_rate": 1.264844187090346e-07,
      "loss": 0.764889,
      "num_input_tokens_seen": 159985905,
      "step": 7398,
      "time_per_iteration": 2.4495253562927246
    },
    {
      "auxiliary_loss_clip": 0.01130571,
      "auxiliary_loss_mlp": 0.01022667,
      "balance_loss_clip": 1.0420146,
      "balance_loss_mlp": 1.01581502,
      "epoch": 0.889677147838634,
      "flos": 26030855283840.0,
      "grad_norm": 1.6358880314610367,
      "language_loss": 0.75075173,
      "learning_rate": 1.262119376137516e-07,
      "loss": 0.77228409,
      "num_input_tokens_seen": 160006965,
      "step": 7399,
      "time_per_iteration": 2.5144355297088623
    },
    {
      "auxiliary_loss_clip": 0.01140077,
      "auxiliary_loss_mlp": 0.0102124,
      "balance_loss_clip": 1.04162705,
      "balance_loss_mlp": 1.01450479,
      "epoch": 0.8897973907292731,
      "flos": 26468283110400.0,
      "grad_norm": 1.7458554696152109,
      "language_loss": 0.85329747,
      "learning_rate": 1.2593974077062707e-07,
      "loss": 0.87491059,
      "num_input_tokens_seen": 160028585,
      "step": 7400,
      "time_per_iteration": 2.476206064224243
    },
    {
      "auxiliary_loss_clip": 0.01113747,
      "auxiliary_loss_mlp": 0.01024551,
      "balance_loss_clip": 1.04120493,
      "balance_loss_mlp": 1.01775885,
      "epoch": 0.8899176336199123,
      "flos": 26249694894720.0,
      "grad_norm": 7.562010722876379,
      "language_loss": 0.63592076,
      "learning_rate": 1.2566782822095423e-07,
      "loss": 0.65730375,
      "num_input_tokens_seen": 160048840,
      "step": 7401,
      "time_per_iteration": 4.126861333847046
    },
    {
      "auxiliary_loss_clip": 0.01129818,
      "auxiliary_loss_mlp": 0.0102283,
      "balance_loss_clip": 1.04625154,
      "balance_loss_mlp": 1.01578498,
      "epoch": 0.8900378765105513,
      "flos": 20811742156800.0,
      "grad_norm": 2.8760480489099813,
      "language_loss": 0.71339417,
      "learning_rate": 1.2539620000598162e-07,
      "loss": 0.73492062,
      "num_input_tokens_seen": 160068175,
      "step": 7402,
      "time_per_iteration": 2.5083558559417725
    },
    {
      "auxiliary_loss_clip": 0.01164046,
      "auxiliary_loss_mlp": 0.01025839,
      "balance_loss_clip": 1.04726446,
      "balance_loss_mlp": 1.01882672,
      "epoch": 0.8901581194011904,
      "flos": 16472333018880.0,
      "grad_norm": 1.883859289469901,
      "language_loss": 0.79626667,
      "learning_rate": 1.2512485616691492e-07,
      "loss": 0.81816554,
      "num_input_tokens_seen": 160085230,
      "step": 7403,
      "time_per_iteration": 3.2173001766204834
    },
    {
      "auxiliary_loss_clip": 0.01125272,
      "auxiliary_loss_mlp": 0.01030064,
      "balance_loss_clip": 1.04240394,
      "balance_loss_mlp": 1.02227354,
      "epoch": 0.8902783622918296,
      "flos": 35155253773440.0,
      "grad_norm": 1.5119369748728388,
      "language_loss": 0.81071472,
      "learning_rate": 1.2485379674491681e-07,
      "loss": 0.83226812,
      "num_input_tokens_seen": 160111425,
      "step": 7404,
      "time_per_iteration": 2.6411101818084717
    },
    {
      "auxiliary_loss_clip": 0.01135224,
      "auxiliary_loss_mlp": 0.01026599,
      "balance_loss_clip": 1.0448761,
      "balance_loss_mlp": 1.01967311,
      "epoch": 0.8903986051824686,
      "flos": 17201068145280.0,
      "grad_norm": 2.6336134624395586,
      "language_loss": 0.79408747,
      "learning_rate": 1.2458302178110657e-07,
      "loss": 0.81570572,
      "num_input_tokens_seen": 160129790,
      "step": 7405,
      "time_per_iteration": 2.427273988723755
    },
    {
      "auxiliary_loss_clip": 0.01111209,
      "auxiliary_loss_mlp": 0.01019656,
      "balance_loss_clip": 1.04037714,
      "balance_loss_mlp": 1.01325107,
      "epoch": 0.8905188480731077,
      "flos": 25483863997440.0,
      "grad_norm": 1.941480803882672,
      "language_loss": 0.82519531,
      "learning_rate": 1.2431253131656118e-07,
      "loss": 0.84650397,
      "num_input_tokens_seen": 160149265,
      "step": 7406,
      "time_per_iteration": 2.5330002307891846
    },
    {
      "auxiliary_loss_clip": 0.01129471,
      "auxiliary_loss_mlp": 0.01024975,
      "balance_loss_clip": 1.04289412,
      "balance_loss_mlp": 1.01754248,
      "epoch": 0.8906390909637467,
      "flos": 23365888502400.0,
      "grad_norm": 3.7995477910424658,
      "language_loss": 0.76877761,
      "learning_rate": 1.240423253923133e-07,
      "loss": 0.79032207,
      "num_input_tokens_seen": 160168870,
      "step": 7407,
      "time_per_iteration": 2.478048086166382
    },
    {
      "auxiliary_loss_clip": 0.01150761,
      "auxiliary_loss_mlp": 0.01026083,
      "balance_loss_clip": 1.0448072,
      "balance_loss_mlp": 1.01868892,
      "epoch": 0.8907593338543859,
      "flos": 21068790860160.0,
      "grad_norm": 1.8226502738910912,
      "language_loss": 0.69473112,
      "learning_rate": 1.237724040493533e-07,
      "loss": 0.71649951,
      "num_input_tokens_seen": 160187495,
      "step": 7408,
      "time_per_iteration": 2.421055793762207
    },
    {
      "auxiliary_loss_clip": 0.01170912,
      "auxiliary_loss_mlp": 0.01028792,
      "balance_loss_clip": 1.05000138,
      "balance_loss_mlp": 1.02120709,
      "epoch": 0.8908795767450249,
      "flos": 21869562712320.0,
      "grad_norm": 3.31065782897963,
      "language_loss": 0.73056811,
      "learning_rate": 1.2350276732862773e-07,
      "loss": 0.75256515,
      "num_input_tokens_seen": 160208520,
      "step": 7409,
      "time_per_iteration": 2.4612503051757812
    },
    {
      "auxiliary_loss_clip": 0.01052676,
      "auxiliary_loss_mlp": 0.01000968,
      "balance_loss_clip": 1.0082022,
      "balance_loss_mlp": 1.00007427,
      "epoch": 0.890999819635664,
      "flos": 66307869348480.0,
      "grad_norm": 0.8374006158886216,
      "language_loss": 0.56670988,
      "learning_rate": 1.2323341527103993e-07,
      "loss": 0.5872463,
      "num_input_tokens_seen": 160263720,
      "step": 7410,
      "time_per_iteration": 3.659691572189331
    },
    {
      "auxiliary_loss_clip": 0.01162823,
      "auxiliary_loss_mlp": 0.01023182,
      "balance_loss_clip": 1.04615211,
      "balance_loss_mlp": 1.01632476,
      "epoch": 0.8911200625263032,
      "flos": 26869908055680.0,
      "grad_norm": 2.1003972524876815,
      "language_loss": 0.85460377,
      "learning_rate": 1.2296434791745135e-07,
      "loss": 0.87646377,
      "num_input_tokens_seen": 160282170,
      "step": 7411,
      "time_per_iteration": 2.4414825439453125
    },
    {
      "auxiliary_loss_clip": 0.01154789,
      "auxiliary_loss_mlp": 0.01026548,
      "balance_loss_clip": 1.04659259,
      "balance_loss_mlp": 1.01940739,
      "epoch": 0.8912403054169422,
      "flos": 20885825957760.0,
      "grad_norm": 2.557252075666105,
      "language_loss": 0.76739734,
      "learning_rate": 1.2269556530867875e-07,
      "loss": 0.78921074,
      "num_input_tokens_seen": 160300725,
      "step": 7412,
      "time_per_iteration": 2.450429677963257
    },
    {
      "auxiliary_loss_clip": 0.01170689,
      "auxiliary_loss_mlp": 0.01025854,
      "balance_loss_clip": 1.04911482,
      "balance_loss_mlp": 1.01763499,
      "epoch": 0.8913605483075813,
      "flos": 27016567286400.0,
      "grad_norm": 2.022584928617387,
      "language_loss": 0.82289821,
      "learning_rate": 1.2242706748549614e-07,
      "loss": 0.84486365,
      "num_input_tokens_seen": 160318720,
      "step": 7413,
      "time_per_iteration": 2.451805830001831
    },
    {
      "auxiliary_loss_clip": 0.01136557,
      "auxiliary_loss_mlp": 0.01019489,
      "balance_loss_clip": 1.04030132,
      "balance_loss_mlp": 1.01242292,
      "epoch": 0.8914807911982204,
      "flos": 23621500661760.0,
      "grad_norm": 1.7612669472426261,
      "language_loss": 0.82317173,
      "learning_rate": 1.2215885448863473e-07,
      "loss": 0.84473217,
      "num_input_tokens_seen": 160339595,
      "step": 7414,
      "time_per_iteration": 2.521181106567383
    },
    {
      "auxiliary_loss_clip": 0.01136878,
      "auxiliary_loss_mlp": 0.01025779,
      "balance_loss_clip": 1.04524505,
      "balance_loss_mlp": 1.01901364,
      "epoch": 0.8916010340888595,
      "flos": 24462277286400.0,
      "grad_norm": 1.9553151334583236,
      "language_loss": 0.8039254,
      "learning_rate": 1.2189092635878152e-07,
      "loss": 0.82555199,
      "num_input_tokens_seen": 160361045,
      "step": 7415,
      "time_per_iteration": 2.5448262691497803
    },
    {
      "auxiliary_loss_clip": 0.01113019,
      "auxiliary_loss_mlp": 0.01023014,
      "balance_loss_clip": 1.0408926,
      "balance_loss_mlp": 1.01594472,
      "epoch": 0.8917212769794985,
      "flos": 21215773313280.0,
      "grad_norm": 1.9104440364754791,
      "language_loss": 0.7771312,
      "learning_rate": 1.216232831365822e-07,
      "loss": 0.7984916,
      "num_input_tokens_seen": 160379990,
      "step": 7416,
      "time_per_iteration": 2.550075054168701
    },
    {
      "auxiliary_loss_clip": 0.01143079,
      "auxiliary_loss_mlp": 0.01026576,
      "balance_loss_clip": 1.04576087,
      "balance_loss_mlp": 1.01917338,
      "epoch": 0.8918415198701377,
      "flos": 25513992529920.0,
      "grad_norm": 1.9466832616763416,
      "language_loss": 0.80975842,
      "learning_rate": 1.2135592486263678e-07,
      "loss": 0.83145499,
      "num_input_tokens_seen": 160399240,
      "step": 7417,
      "time_per_iteration": 2.5348081588745117
    },
    {
      "auxiliary_loss_clip": 0.01135294,
      "auxiliary_loss_mlp": 0.01025101,
      "balance_loss_clip": 1.042979,
      "balance_loss_mlp": 1.01838028,
      "epoch": 0.8919617627607768,
      "flos": 37853006693760.0,
      "grad_norm": 1.6980455305116617,
      "language_loss": 0.61257374,
      "learning_rate": 1.2108885157750415e-07,
      "loss": 0.63417768,
      "num_input_tokens_seen": 160421600,
      "step": 7418,
      "time_per_iteration": 2.607163190841675
    },
    {
      "auxiliary_loss_clip": 0.01118372,
      "auxiliary_loss_mlp": 0.00761162,
      "balance_loss_clip": 1.04372573,
      "balance_loss_mlp": 1.00039625,
      "epoch": 0.8920820056514158,
      "flos": 26213676531840.0,
      "grad_norm": 1.8163406681216865,
      "language_loss": 0.79795241,
      "learning_rate": 1.2082206332169897e-07,
      "loss": 0.81674778,
      "num_input_tokens_seen": 160441695,
      "step": 7419,
      "time_per_iteration": 2.5493650436401367
    },
    {
      "auxiliary_loss_clip": 0.01133754,
      "auxiliary_loss_mlp": 0.01027934,
      "balance_loss_clip": 1.04627526,
      "balance_loss_mlp": 1.02039957,
      "epoch": 0.892202248542055,
      "flos": 17383135207680.0,
      "grad_norm": 4.941490706673702,
      "language_loss": 0.73491853,
      "learning_rate": 1.2055556013569225e-07,
      "loss": 0.75653541,
      "num_input_tokens_seen": 160457205,
      "step": 7420,
      "time_per_iteration": 2.424400568008423
    },
    {
      "auxiliary_loss_clip": 0.0113947,
      "auxiliary_loss_mlp": 0.01023454,
      "balance_loss_clip": 1.04440367,
      "balance_loss_mlp": 1.01654625,
      "epoch": 0.892322491432694,
      "flos": 21324223451520.0,
      "grad_norm": 1.8439221813257884,
      "language_loss": 0.82086712,
      "learning_rate": 1.2028934205991315e-07,
      "loss": 0.84249634,
      "num_input_tokens_seen": 160476525,
      "step": 7421,
      "time_per_iteration": 2.496135950088501
    },
    {
      "auxiliary_loss_clip": 0.01149567,
      "auxiliary_loss_mlp": 0.01021889,
      "balance_loss_clip": 1.04336703,
      "balance_loss_mlp": 1.01447713,
      "epoch": 0.8924427343233331,
      "flos": 24029374573440.0,
      "grad_norm": 1.7799896893532616,
      "language_loss": 0.76575053,
      "learning_rate": 1.2002340913474607e-07,
      "loss": 0.7874651,
      "num_input_tokens_seen": 160500160,
      "step": 7422,
      "time_per_iteration": 2.5181031227111816
    },
    {
      "auxiliary_loss_clip": 0.01165782,
      "auxiliary_loss_mlp": 0.01029137,
      "balance_loss_clip": 1.04667592,
      "balance_loss_mlp": 1.02164483,
      "epoch": 0.8925629772139723,
      "flos": 30008069631360.0,
      "grad_norm": 11.415048182270004,
      "language_loss": 0.74153155,
      "learning_rate": 1.1975776140053317e-07,
      "loss": 0.76348078,
      "num_input_tokens_seen": 160520130,
      "step": 7423,
      "time_per_iteration": 2.4831578731536865
    },
    {
      "auxiliary_loss_clip": 0.01111058,
      "auxiliary_loss_mlp": 0.01025745,
      "balance_loss_clip": 1.04205298,
      "balance_loss_mlp": 1.01827919,
      "epoch": 0.8926832201046113,
      "flos": 22601709630720.0,
      "grad_norm": 1.8767466371757187,
      "language_loss": 0.73635966,
      "learning_rate": 1.194923988975729e-07,
      "loss": 0.75772768,
      "num_input_tokens_seen": 160539730,
      "step": 7424,
      "time_per_iteration": 2.5377964973449707
    },
    {
      "auxiliary_loss_clip": 0.01119216,
      "auxiliary_loss_mlp": 0.01020395,
      "balance_loss_clip": 1.04280376,
      "balance_loss_mlp": 1.01238418,
      "epoch": 0.8928034629952504,
      "flos": 13297722117120.0,
      "grad_norm": 2.266639274637357,
      "language_loss": 0.74034125,
      "learning_rate": 1.192273216661206e-07,
      "loss": 0.76173735,
      "num_input_tokens_seen": 160557820,
      "step": 7425,
      "time_per_iteration": 2.517336130142212
    },
    {
      "auxiliary_loss_clip": 0.01011198,
      "auxiliary_loss_mlp": 0.01000785,
      "balance_loss_clip": 1.00823212,
      "balance_loss_mlp": 0.99978322,
      "epoch": 0.8929237058858895,
      "flos": 54854556744960.0,
      "grad_norm": 0.774418217978881,
      "language_loss": 0.57487988,
      "learning_rate": 1.189625297463881e-07,
      "loss": 0.59499979,
      "num_input_tokens_seen": 160619510,
      "step": 7426,
      "time_per_iteration": 3.2216598987579346
    },
    {
      "auxiliary_loss_clip": 0.01089477,
      "auxiliary_loss_mlp": 0.01025038,
      "balance_loss_clip": 1.03863728,
      "balance_loss_mlp": 1.01832664,
      "epoch": 0.8930439487765286,
      "flos": 28883850785280.0,
      "grad_norm": 1.5840368515953966,
      "language_loss": 0.79691482,
      "learning_rate": 1.1869802317854394e-07,
      "loss": 0.81805998,
      "num_input_tokens_seen": 160643295,
      "step": 7427,
      "time_per_iteration": 3.826235055923462
    },
    {
      "auxiliary_loss_clip": 0.01113734,
      "auxiliary_loss_mlp": 0.01023256,
      "balance_loss_clip": 1.04221451,
      "balance_loss_mlp": 1.0160886,
      "epoch": 0.8931641916671677,
      "flos": 22419283432320.0,
      "grad_norm": 1.9062757888674386,
      "language_loss": 0.71996498,
      "learning_rate": 1.1843380200271425e-07,
      "loss": 0.74133492,
      "num_input_tokens_seen": 160662495,
      "step": 7428,
      "time_per_iteration": 3.3114731311798096
    },
    {
      "auxiliary_loss_clip": 0.01115727,
      "auxiliary_loss_mlp": 0.01018786,
      "balance_loss_clip": 1.04142535,
      "balance_loss_mlp": 1.01155257,
      "epoch": 0.8932844345578068,
      "flos": 25843149786240.0,
      "grad_norm": 2.1047395284678694,
      "language_loss": 0.80183971,
      "learning_rate": 1.181698662589805e-07,
      "loss": 0.82318485,
      "num_input_tokens_seen": 160682080,
      "step": 7429,
      "time_per_iteration": 3.39693284034729
    },
    {
      "auxiliary_loss_clip": 0.01150058,
      "auxiliary_loss_mlp": 0.01025833,
      "balance_loss_clip": 1.04470217,
      "balance_loss_mlp": 1.01841831,
      "epoch": 0.8934046774484459,
      "flos": 22925803069440.0,
      "grad_norm": 2.0712411145507152,
      "language_loss": 0.7615875,
      "learning_rate": 1.1790621598738249e-07,
      "loss": 0.78334641,
      "num_input_tokens_seen": 160700395,
      "step": 7430,
      "time_per_iteration": 2.5288875102996826
    },
    {
      "auxiliary_loss_clip": 0.01162498,
      "auxiliary_loss_mlp": 0.01023594,
      "balance_loss_clip": 1.04783988,
      "balance_loss_mlp": 1.01703143,
      "epoch": 0.8935249203390849,
      "flos": 24462097718400.0,
      "grad_norm": 2.0024294129715092,
      "language_loss": 0.75163984,
      "learning_rate": 1.1764285122791461e-07,
      "loss": 0.77350074,
      "num_input_tokens_seen": 160721115,
      "step": 7431,
      "time_per_iteration": 2.4550232887268066
    },
    {
      "auxiliary_loss_clip": 0.01148549,
      "auxiliary_loss_mlp": 0.0102207,
      "balance_loss_clip": 1.04208839,
      "balance_loss_mlp": 1.01517642,
      "epoch": 0.8936451632297241,
      "flos": 15742735966080.0,
      "grad_norm": 2.513241487722935,
      "language_loss": 0.77320063,
      "learning_rate": 1.173797720205294e-07,
      "loss": 0.79490674,
      "num_input_tokens_seen": 160739150,
      "step": 7432,
      "time_per_iteration": 2.4107601642608643
    },
    {
      "auxiliary_loss_clip": 0.0115387,
      "auxiliary_loss_mlp": 0.0102771,
      "balance_loss_clip": 1.04704237,
      "balance_loss_mlp": 1.0200119,
      "epoch": 0.8937654061203631,
      "flos": 35115500396160.0,
      "grad_norm": 3.5361902286749323,
      "language_loss": 0.71541786,
      "learning_rate": 1.1711697840513602e-07,
      "loss": 0.73723364,
      "num_input_tokens_seen": 160758585,
      "step": 7433,
      "time_per_iteration": 2.547919273376465
    },
    {
      "auxiliary_loss_clip": 0.01142744,
      "auxiliary_loss_mlp": 0.01022179,
      "balance_loss_clip": 1.04243076,
      "balance_loss_mlp": 1.01492846,
      "epoch": 0.8938856490110022,
      "flos": 16107444708480.0,
      "grad_norm": 2.8434040265186833,
      "language_loss": 0.71168488,
      "learning_rate": 1.1685447042160012e-07,
      "loss": 0.73333406,
      "num_input_tokens_seen": 160776620,
      "step": 7434,
      "time_per_iteration": 2.403050184249878
    },
    {
      "auxiliary_loss_clip": 0.01166433,
      "auxiliary_loss_mlp": 0.01028259,
      "balance_loss_clip": 1.0472002,
      "balance_loss_mlp": 1.02115083,
      "epoch": 0.8940058919016414,
      "flos": 20704189858560.0,
      "grad_norm": 2.2754156304127213,
      "language_loss": 0.71512991,
      "learning_rate": 1.1659224810974367e-07,
      "loss": 0.73707676,
      "num_input_tokens_seen": 160796580,
      "step": 7435,
      "time_per_iteration": 2.403956413269043
    },
    {
      "auxiliary_loss_clip": 0.01138612,
      "auxiliary_loss_mlp": 0.0102581,
      "balance_loss_clip": 1.047207,
      "balance_loss_mlp": 1.01860118,
      "epoch": 0.8941261347922804,
      "flos": 25229041937280.0,
      "grad_norm": 2.186319305615022,
      "language_loss": 0.68190753,
      "learning_rate": 1.1633031150934591e-07,
      "loss": 0.70355177,
      "num_input_tokens_seen": 160819610,
      "step": 7436,
      "time_per_iteration": 2.550081491470337
    },
    {
      "auxiliary_loss_clip": 0.0115303,
      "auxiliary_loss_mlp": 0.01031275,
      "balance_loss_clip": 1.04694581,
      "balance_loss_mlp": 1.02384853,
      "epoch": 0.8942463776829195,
      "flos": 19537236806400.0,
      "grad_norm": 2.0891362424621764,
      "language_loss": 0.80209744,
      "learning_rate": 1.1606866066014176e-07,
      "loss": 0.82394052,
      "num_input_tokens_seen": 160838660,
      "step": 7437,
      "time_per_iteration": 3.192620277404785
    },
    {
      "auxiliary_loss_clip": 0.0111977,
      "auxiliary_loss_mlp": 0.01022264,
      "balance_loss_clip": 1.04246688,
      "balance_loss_mlp": 1.01493287,
      "epoch": 0.8943666205735585,
      "flos": 22301567585280.0,
      "grad_norm": 2.570170749072981,
      "language_loss": 0.75530273,
      "learning_rate": 1.1580729560182434e-07,
      "loss": 0.77672309,
      "num_input_tokens_seen": 160854515,
      "step": 7438,
      "time_per_iteration": 2.481064558029175
    },
    {
      "auxiliary_loss_clip": 0.01164953,
      "auxiliary_loss_mlp": 0.00761766,
      "balance_loss_clip": 1.0469346,
      "balance_loss_mlp": 1.0004034,
      "epoch": 0.8944868634641977,
      "flos": 18912893581440.0,
      "grad_norm": 2.0853067501221054,
      "language_loss": 0.70763063,
      "learning_rate": 1.1554621637404171e-07,
      "loss": 0.72689784,
      "num_input_tokens_seen": 160872605,
      "step": 7439,
      "time_per_iteration": 2.3844475746154785
    },
    {
      "auxiliary_loss_clip": 0.0115437,
      "auxiliary_loss_mlp": 0.01019545,
      "balance_loss_clip": 1.04668391,
      "balance_loss_mlp": 1.01252007,
      "epoch": 0.8946071063548368,
      "flos": 14460904241280.0,
      "grad_norm": 3.424561241713183,
      "language_loss": 0.61309856,
      "learning_rate": 1.1528542301639999e-07,
      "loss": 0.63483769,
      "num_input_tokens_seen": 160889395,
      "step": 7440,
      "time_per_iteration": 2.398780584335327
    },
    {
      "auxiliary_loss_clip": 0.01124041,
      "auxiliary_loss_mlp": 0.01020349,
      "balance_loss_clip": 1.04062712,
      "balance_loss_mlp": 1.01352429,
      "epoch": 0.8947273492454758,
      "flos": 20084084438400.0,
      "grad_norm": 2.939457369216438,
      "language_loss": 0.82780707,
      "learning_rate": 1.1502491556846105e-07,
      "loss": 0.84925097,
      "num_input_tokens_seen": 160907890,
      "step": 7441,
      "time_per_iteration": 2.5046281814575195
    },
    {
      "auxiliary_loss_clip": 0.01136078,
      "auxiliary_loss_mlp": 0.01023901,
      "balance_loss_clip": 1.0440073,
      "balance_loss_mlp": 1.01705587,
      "epoch": 0.894847592136115,
      "flos": 18550555136640.0,
      "grad_norm": 2.7401060281557363,
      "language_loss": 0.81608999,
      "learning_rate": 1.1476469406974331e-07,
      "loss": 0.83768976,
      "num_input_tokens_seen": 160923490,
      "step": 7442,
      "time_per_iteration": 2.4371020793914795
    },
    {
      "auxiliary_loss_clip": 0.01163353,
      "auxiliary_loss_mlp": 0.01025048,
      "balance_loss_clip": 1.04734921,
      "balance_loss_mlp": 1.01837802,
      "epoch": 0.894967835026754,
      "flos": 23478468704640.0,
      "grad_norm": 1.7182839172039632,
      "language_loss": 0.77063191,
      "learning_rate": 1.1450475855972341e-07,
      "loss": 0.79251593,
      "num_input_tokens_seen": 160944280,
      "step": 7443,
      "time_per_iteration": 2.4304957389831543
    },
    {
      "auxiliary_loss_clip": 0.01135476,
      "auxiliary_loss_mlp": 0.00762009,
      "balance_loss_clip": 1.04205918,
      "balance_loss_mlp": 1.00045168,
      "epoch": 0.8950880779173931,
      "flos": 15188310564480.0,
      "grad_norm": 3.074678388272662,
      "language_loss": 0.71294785,
      "learning_rate": 1.1424510907783158e-07,
      "loss": 0.73192275,
      "num_input_tokens_seen": 160961560,
      "step": 7444,
      "time_per_iteration": 2.4330248832702637
    },
    {
      "auxiliary_loss_clip": 0.01140329,
      "auxiliary_loss_mlp": 0.01026176,
      "balance_loss_clip": 1.04253697,
      "balance_loss_mlp": 1.01941133,
      "epoch": 0.8952083208080323,
      "flos": 22091957769600.0,
      "grad_norm": 1.5495937147365109,
      "language_loss": 0.82725537,
      "learning_rate": 1.1398574566345787e-07,
      "loss": 0.84892035,
      "num_input_tokens_seen": 160982195,
      "step": 7445,
      "time_per_iteration": 2.4831669330596924
    },
    {
      "auxiliary_loss_clip": 0.01141352,
      "auxiliary_loss_mlp": 0.01022911,
      "balance_loss_clip": 1.04245603,
      "balance_loss_mlp": 1.0154959,
      "epoch": 0.8953285636986713,
      "flos": 23254026572160.0,
      "grad_norm": 2.426159385548367,
      "language_loss": 0.82391077,
      "learning_rate": 1.1372666835594702e-07,
      "loss": 0.8455534,
      "num_input_tokens_seen": 161000520,
      "step": 7446,
      "time_per_iteration": 2.4716262817382812
    },
    {
      "auxiliary_loss_clip": 0.0113505,
      "auxiliary_loss_mlp": 0.0101974,
      "balance_loss_clip": 1.04381502,
      "balance_loss_mlp": 1.01338887,
      "epoch": 0.8954488065893104,
      "flos": 16362661818240.0,
      "grad_norm": 2.000137099003427,
      "language_loss": 0.71833777,
      "learning_rate": 1.1346787719460071e-07,
      "loss": 0.73988569,
      "num_input_tokens_seen": 161019405,
      "step": 7447,
      "time_per_iteration": 2.4457385540008545
    },
    {
      "auxiliary_loss_clip": 0.01136824,
      "auxiliary_loss_mlp": 0.01025025,
      "balance_loss_clip": 1.04591799,
      "balance_loss_mlp": 1.01820016,
      "epoch": 0.8955690494799495,
      "flos": 18257883120000.0,
      "grad_norm": 1.7287385889543345,
      "language_loss": 0.71899825,
      "learning_rate": 1.1320937221867732e-07,
      "loss": 0.7406168,
      "num_input_tokens_seen": 161036985,
      "step": 7448,
      "time_per_iteration": 2.464522361755371
    },
    {
      "auxiliary_loss_clip": 0.01134856,
      "auxiliary_loss_mlp": 0.01024953,
      "balance_loss_clip": 1.04284477,
      "balance_loss_mlp": 1.01863527,
      "epoch": 0.8956892923705886,
      "flos": 25447486498560.0,
      "grad_norm": 1.7557887997380401,
      "language_loss": 0.79429418,
      "learning_rate": 1.1295115346739192e-07,
      "loss": 0.81589228,
      "num_input_tokens_seen": 161056985,
      "step": 7449,
      "time_per_iteration": 2.5046324729919434
    },
    {
      "auxiliary_loss_clip": 0.01140015,
      "auxiliary_loss_mlp": 0.01025745,
      "balance_loss_clip": 1.04442084,
      "balance_loss_mlp": 1.01857412,
      "epoch": 0.8958095352612276,
      "flos": 52661883939840.0,
      "grad_norm": 5.6498850965287355,
      "language_loss": 0.72896147,
      "learning_rate": 1.1269322097991629e-07,
      "loss": 0.75061899,
      "num_input_tokens_seen": 161080270,
      "step": 7450,
      "time_per_iteration": 2.739476203918457
    },
    {
      "auxiliary_loss_clip": 0.0115662,
      "auxiliary_loss_mlp": 0.01025377,
      "balance_loss_clip": 1.0491457,
      "balance_loss_mlp": 1.0177176,
      "epoch": 0.8959297781518668,
      "flos": 23186335392000.0,
      "grad_norm": 2.5594880084113467,
      "language_loss": 0.67722237,
      "learning_rate": 1.1243557479537846e-07,
      "loss": 0.69904232,
      "num_input_tokens_seen": 161100160,
      "step": 7451,
      "time_per_iteration": 2.4521005153656006
    },
    {
      "auxiliary_loss_clip": 0.01162095,
      "auxiliary_loss_mlp": 0.01019687,
      "balance_loss_clip": 1.04424345,
      "balance_loss_mlp": 1.01251972,
      "epoch": 0.8960500210425059,
      "flos": 20334309557760.0,
      "grad_norm": 2.1946308636044383,
      "language_loss": 0.68573368,
      "learning_rate": 1.121782149528634e-07,
      "loss": 0.70755148,
      "num_input_tokens_seen": 161117260,
      "step": 7452,
      "time_per_iteration": 2.383960485458374
    },
    {
      "auxiliary_loss_clip": 0.01142041,
      "auxiliary_loss_mlp": 0.01020068,
      "balance_loss_clip": 1.04624486,
      "balance_loss_mlp": 1.01320195,
      "epoch": 0.8961702639331449,
      "flos": 19901694153600.0,
      "grad_norm": 2.1996692613356705,
      "language_loss": 0.78960574,
      "learning_rate": 1.1192114149141208e-07,
      "loss": 0.81122684,
      "num_input_tokens_seen": 161136895,
      "step": 7453,
      "time_per_iteration": 2.463416576385498
    },
    {
      "auxiliary_loss_clip": 0.01140509,
      "auxiliary_loss_mlp": 0.01027593,
      "balance_loss_clip": 1.04319167,
      "balance_loss_mlp": 1.01975524,
      "epoch": 0.8962905068237841,
      "flos": 12896348567040.0,
      "grad_norm": 3.0133785671726088,
      "language_loss": 0.6505838,
      "learning_rate": 1.1166435445002197e-07,
      "loss": 0.67226481,
      "num_input_tokens_seen": 161154565,
      "step": 7454,
      "time_per_iteration": 3.442310333251953
    },
    {
      "auxiliary_loss_clip": 0.0115446,
      "auxiliary_loss_mlp": 0.01026608,
      "balance_loss_clip": 1.04701924,
      "balance_loss_mlp": 1.01890111,
      "epoch": 0.8964107497144231,
      "flos": 23440331439360.0,
      "grad_norm": 2.124418998365666,
      "language_loss": 0.68864495,
      "learning_rate": 1.1140785386764818e-07,
      "loss": 0.71045566,
      "num_input_tokens_seen": 161173265,
      "step": 7455,
      "time_per_iteration": 3.191255807876587
    },
    {
      "auxiliary_loss_clip": 0.0114625,
      "auxiliary_loss_mlp": 0.01027711,
      "balance_loss_clip": 1.04445922,
      "balance_loss_mlp": 1.02039468,
      "epoch": 0.8965309926050622,
      "flos": 19500176949120.0,
      "grad_norm": 2.1232624020422164,
      "language_loss": 0.69580305,
      "learning_rate": 1.1115163978320153e-07,
      "loss": 0.71754265,
      "num_input_tokens_seen": 161191995,
      "step": 7456,
      "time_per_iteration": 3.2421813011169434
    },
    {
      "auxiliary_loss_clip": 0.01156426,
      "auxiliary_loss_mlp": 0.00762208,
      "balance_loss_clip": 1.04639804,
      "balance_loss_mlp": 1.00046539,
      "epoch": 0.8966512354957014,
      "flos": 28658008022400.0,
      "grad_norm": 1.8422661718315823,
      "language_loss": 0.82507384,
      "learning_rate": 1.1089571223554917e-07,
      "loss": 0.84426022,
      "num_input_tokens_seen": 161212880,
      "step": 7457,
      "time_per_iteration": 2.504429340362549
    },
    {
      "auxiliary_loss_clip": 0.01151697,
      "auxiliary_loss_mlp": 0.01024351,
      "balance_loss_clip": 1.04404759,
      "balance_loss_mlp": 1.01708531,
      "epoch": 0.8967714783863404,
      "flos": 23370916406400.0,
      "grad_norm": 2.0165449609390267,
      "language_loss": 0.85309386,
      "learning_rate": 1.1064007126351537e-07,
      "loss": 0.87485433,
      "num_input_tokens_seen": 161233595,
      "step": 7458,
      "time_per_iteration": 2.4564974308013916
    },
    {
      "auxiliary_loss_clip": 0.01131771,
      "auxiliary_loss_mlp": 0.01021595,
      "balance_loss_clip": 1.04406762,
      "balance_loss_mlp": 1.01449597,
      "epoch": 0.8968917212769795,
      "flos": 24535175938560.0,
      "grad_norm": 2.126901697992628,
      "language_loss": 0.76233041,
      "learning_rate": 1.1038471690588003e-07,
      "loss": 0.78386408,
      "num_input_tokens_seen": 161252740,
      "step": 7459,
      "time_per_iteration": 2.479123592376709
    },
    {
      "auxiliary_loss_clip": 0.01112005,
      "auxiliary_loss_mlp": 0.01025745,
      "balance_loss_clip": 1.04642045,
      "balance_loss_mlp": 1.01862216,
      "epoch": 0.8970119641676186,
      "flos": 23475416048640.0,
      "grad_norm": 2.3379079766910373,
      "language_loss": 0.79907155,
      "learning_rate": 1.1012964920138145e-07,
      "loss": 0.82044905,
      "num_input_tokens_seen": 161272325,
      "step": 7460,
      "time_per_iteration": 2.550973892211914
    },
    {
      "auxiliary_loss_clip": 0.01130086,
      "auxiliary_loss_mlp": 0.01024242,
      "balance_loss_clip": 1.03987694,
      "balance_loss_mlp": 1.01748919,
      "epoch": 0.8971322070582577,
      "flos": 24538192680960.0,
      "grad_norm": 1.8724472127650784,
      "language_loss": 0.75860697,
      "learning_rate": 1.0987486818871205e-07,
      "loss": 0.78015023,
      "num_input_tokens_seen": 161295915,
      "step": 7461,
      "time_per_iteration": 2.5437662601470947
    },
    {
      "auxiliary_loss_clip": 0.01151035,
      "auxiliary_loss_mlp": 0.00761777,
      "balance_loss_clip": 1.04521227,
      "balance_loss_mlp": 1.00043297,
      "epoch": 0.8972524499488967,
      "flos": 21797454159360.0,
      "grad_norm": 2.5814274669859403,
      "language_loss": 0.73589694,
      "learning_rate": 1.0962037390652245e-07,
      "loss": 0.75502509,
      "num_input_tokens_seen": 161314935,
      "step": 7462,
      "time_per_iteration": 2.455146551132202
    },
    {
      "auxiliary_loss_clip": 0.01138681,
      "auxiliary_loss_mlp": 0.01026467,
      "balance_loss_clip": 1.04517114,
      "balance_loss_mlp": 1.01901698,
      "epoch": 0.8973726928395359,
      "flos": 21726243446400.0,
      "grad_norm": 1.785756433750136,
      "language_loss": 0.71740371,
      "learning_rate": 1.0936616639341911e-07,
      "loss": 0.73905522,
      "num_input_tokens_seen": 161335225,
      "step": 7463,
      "time_per_iteration": 2.479999303817749
    },
    {
      "auxiliary_loss_clip": 0.01046846,
      "auxiliary_loss_mlp": 0.01002208,
      "balance_loss_clip": 1.0100919,
      "balance_loss_mlp": 1.00145423,
      "epoch": 0.897492935730175,
      "flos": 53837100097920.0,
      "grad_norm": 0.7421691002946087,
      "language_loss": 0.54748058,
      "learning_rate": 1.0911224568796473e-07,
      "loss": 0.56797123,
      "num_input_tokens_seen": 161393420,
      "step": 7464,
      "time_per_iteration": 3.8243470191955566
    },
    {
      "auxiliary_loss_clip": 0.01149507,
      "auxiliary_loss_mlp": 0.01027458,
      "balance_loss_clip": 1.04633892,
      "balance_loss_mlp": 1.02052569,
      "epoch": 0.897613178620814,
      "flos": 18290346036480.0,
      "grad_norm": 3.4304111658655776,
      "language_loss": 0.70918477,
      "learning_rate": 1.0885861182867984e-07,
      "loss": 0.73095441,
      "num_input_tokens_seen": 161411525,
      "step": 7465,
      "time_per_iteration": 2.428588628768921
    },
    {
      "auxiliary_loss_clip": 0.01140001,
      "auxiliary_loss_mlp": 0.01022244,
      "balance_loss_clip": 1.04350233,
      "balance_loss_mlp": 1.01524067,
      "epoch": 0.8977334215114532,
      "flos": 32993718059520.0,
      "grad_norm": 2.1091594877299604,
      "language_loss": 0.70983779,
      "learning_rate": 1.0860526485403942e-07,
      "loss": 0.73146021,
      "num_input_tokens_seen": 161432800,
      "step": 7466,
      "time_per_iteration": 2.5627036094665527
    },
    {
      "auxiliary_loss_clip": 0.01164526,
      "auxiliary_loss_mlp": 0.01024297,
      "balance_loss_clip": 1.0464983,
      "balance_loss_mlp": 1.01779079,
      "epoch": 0.8978536644020922,
      "flos": 15195636938880.0,
      "grad_norm": 1.668742687822235,
      "language_loss": 0.77024531,
      "learning_rate": 1.0835220480247675e-07,
      "loss": 0.79213357,
      "num_input_tokens_seen": 161451295,
      "step": 7467,
      "time_per_iteration": 2.377856492996216
    },
    {
      "auxiliary_loss_clip": 0.0113386,
      "auxiliary_loss_mlp": 0.01024681,
      "balance_loss_clip": 1.04409337,
      "balance_loss_mlp": 1.0176661,
      "epoch": 0.8979739072927313,
      "flos": 18004389863040.0,
      "grad_norm": 2.1487084897035733,
      "language_loss": 0.8361212,
      "learning_rate": 1.0809943171238067e-07,
      "loss": 0.85770667,
      "num_input_tokens_seen": 161469220,
      "step": 7468,
      "time_per_iteration": 2.4430463314056396
    },
    {
      "auxiliary_loss_clip": 0.01144633,
      "auxiliary_loss_mlp": 0.01029538,
      "balance_loss_clip": 1.04544282,
      "balance_loss_mlp": 1.02128887,
      "epoch": 0.8980941501833704,
      "flos": 22271546793600.0,
      "grad_norm": 2.3138926097217705,
      "language_loss": 0.62957704,
      "learning_rate": 1.078469456220965e-07,
      "loss": 0.65131873,
      "num_input_tokens_seen": 161489375,
      "step": 7469,
      "time_per_iteration": 2.4690537452697754
    },
    {
      "auxiliary_loss_clip": 0.01150439,
      "auxiliary_loss_mlp": 0.01024784,
      "balance_loss_clip": 1.04356158,
      "balance_loss_mlp": 1.01764941,
      "epoch": 0.8982143930740095,
      "flos": 37560729726720.0,
      "grad_norm": 2.2147316442639955,
      "language_loss": 0.69598716,
      "learning_rate": 1.0759474656992606e-07,
      "loss": 0.71773946,
      "num_input_tokens_seen": 161512145,
      "step": 7470,
      "time_per_iteration": 2.5764031410217285
    },
    {
      "auxiliary_loss_clip": 0.01140976,
      "auxiliary_loss_mlp": 0.0102563,
      "balance_loss_clip": 1.04351974,
      "balance_loss_mlp": 1.01827812,
      "epoch": 0.8983346359646486,
      "flos": 18076893465600.0,
      "grad_norm": 2.144831081515994,
      "language_loss": 0.77924454,
      "learning_rate": 1.0734283459412785e-07,
      "loss": 0.80091059,
      "num_input_tokens_seen": 161528995,
      "step": 7471,
      "time_per_iteration": 2.447983741760254
    },
    {
      "auxiliary_loss_clip": 0.01111957,
      "auxiliary_loss_mlp": 0.01026433,
      "balance_loss_clip": 1.04047716,
      "balance_loss_mlp": 1.01849985,
      "epoch": 0.8984548788552876,
      "flos": 20558895344640.0,
      "grad_norm": 1.855930994033701,
      "language_loss": 0.80479497,
      "learning_rate": 1.0709120973291707e-07,
      "loss": 0.82617891,
      "num_input_tokens_seen": 161548775,
      "step": 7472,
      "time_per_iteration": 2.5386130809783936
    },
    {
      "auxiliary_loss_clip": 0.01166358,
      "auxiliary_loss_mlp": 0.01026877,
      "balance_loss_clip": 1.04730916,
      "balance_loss_mlp": 1.01941466,
      "epoch": 0.8985751217459268,
      "flos": 17785442511360.0,
      "grad_norm": 2.0362015033547927,
      "language_loss": 0.78058845,
      "learning_rate": 1.0683987202446475e-07,
      "loss": 0.80252075,
      "num_input_tokens_seen": 161566960,
      "step": 7473,
      "time_per_iteration": 2.4053902626037598
    },
    {
      "auxiliary_loss_clip": 0.01153978,
      "auxiliary_loss_mlp": 0.01022581,
      "balance_loss_clip": 1.04515266,
      "balance_loss_mlp": 1.01513028,
      "epoch": 0.8986953646365659,
      "flos": 21617003208960.0,
      "grad_norm": 1.873774541766139,
      "language_loss": 0.69974357,
      "learning_rate": 1.0658882150689862e-07,
      "loss": 0.72150916,
      "num_input_tokens_seen": 161585820,
      "step": 7474,
      "time_per_iteration": 2.445111036300659
    },
    {
      "auxiliary_loss_clip": 0.01127501,
      "auxiliary_loss_mlp": 0.01021578,
      "balance_loss_clip": 1.04371154,
      "balance_loss_mlp": 1.01436627,
      "epoch": 0.8988156075272049,
      "flos": 14027355083520.0,
      "grad_norm": 2.9974118179394447,
      "language_loss": 0.78034985,
      "learning_rate": 1.0633805821830288e-07,
      "loss": 0.80184066,
      "num_input_tokens_seen": 161602505,
      "step": 7475,
      "time_per_iteration": 2.4873292446136475
    },
    {
      "auxiliary_loss_clip": 0.01139418,
      "auxiliary_loss_mlp": 0.01025284,
      "balance_loss_clip": 1.04479635,
      "balance_loss_mlp": 1.01764822,
      "epoch": 0.8989358504178441,
      "flos": 29059202004480.0,
      "grad_norm": 6.770366517517771,
      "language_loss": 0.82814127,
      "learning_rate": 1.0608758219671753e-07,
      "loss": 0.84978831,
      "num_input_tokens_seen": 161621545,
      "step": 7476,
      "time_per_iteration": 2.529716968536377
    },
    {
      "auxiliary_loss_clip": 0.01141771,
      "auxiliary_loss_mlp": 0.01021794,
      "balance_loss_clip": 1.04437053,
      "balance_loss_mlp": 1.01490045,
      "epoch": 0.8990560933084831,
      "flos": 20230420446720.0,
      "grad_norm": 1.6617175867881215,
      "language_loss": 0.70759833,
      "learning_rate": 1.0583739348014065e-07,
      "loss": 0.72923398,
      "num_input_tokens_seen": 161642630,
      "step": 7477,
      "time_per_iteration": 2.4870245456695557
    },
    {
      "auxiliary_loss_clip": 0.01168127,
      "auxiliary_loss_mlp": 0.01021755,
      "balance_loss_clip": 1.05057144,
      "balance_loss_mlp": 1.01488304,
      "epoch": 0.8991763361991222,
      "flos": 25520672459520.0,
      "grad_norm": 2.1802925975292387,
      "language_loss": 0.8453145,
      "learning_rate": 1.0558749210652518e-07,
      "loss": 0.86721325,
      "num_input_tokens_seen": 161662560,
      "step": 7478,
      "time_per_iteration": 2.4409868717193604
    },
    {
      "auxiliary_loss_clip": 0.01128589,
      "auxiliary_loss_mlp": 0.01024721,
      "balance_loss_clip": 1.04435384,
      "balance_loss_mlp": 1.01808703,
      "epoch": 0.8992965790897613,
      "flos": 25119191168640.0,
      "grad_norm": 1.8246832863465052,
      "language_loss": 0.85857689,
      "learning_rate": 1.053378781137808e-07,
      "loss": 0.88011003,
      "num_input_tokens_seen": 161683480,
      "step": 7479,
      "time_per_iteration": 2.546184778213501
    },
    {
      "auxiliary_loss_clip": 0.01140802,
      "auxiliary_loss_mlp": 0.01026357,
      "balance_loss_clip": 1.04472256,
      "balance_loss_mlp": 1.01917768,
      "epoch": 0.8994168219804004,
      "flos": 16070815814400.0,
      "grad_norm": 1.9466141023063341,
      "language_loss": 0.77657908,
      "learning_rate": 1.0508855153977392e-07,
      "loss": 0.79825068,
      "num_input_tokens_seen": 161699945,
      "step": 7480,
      "time_per_iteration": 2.4394664764404297
    },
    {
      "auxiliary_loss_clip": 0.01152279,
      "auxiliary_loss_mlp": 0.01026968,
      "balance_loss_clip": 1.04386854,
      "balance_loss_mlp": 1.01962423,
      "epoch": 0.8995370648710395,
      "flos": 24825764966400.0,
      "grad_norm": 2.3094500020391857,
      "language_loss": 0.67153329,
      "learning_rate": 1.0483951242232669e-07,
      "loss": 0.69332576,
      "num_input_tokens_seen": 161720420,
      "step": 7481,
      "time_per_iteration": 3.387246608734131
    },
    {
      "auxiliary_loss_clip": 0.01061128,
      "auxiliary_loss_mlp": 0.01000908,
      "balance_loss_clip": 1.00747323,
      "balance_loss_mlp": 1.00005579,
      "epoch": 0.8996573077616786,
      "flos": 63116238378240.0,
      "grad_norm": 0.9730907483693916,
      "language_loss": 0.57671976,
      "learning_rate": 1.0459076079921936e-07,
      "loss": 0.59734011,
      "num_input_tokens_seen": 161773080,
      "step": 7482,
      "time_per_iteration": 3.7595181465148926
    },
    {
      "auxiliary_loss_clip": 0.01132041,
      "auxiliary_loss_mlp": 0.0102983,
      "balance_loss_clip": 1.04411364,
      "balance_loss_mlp": 1.02226901,
      "epoch": 0.8997775506523177,
      "flos": 18219674027520.0,
      "grad_norm": 2.167037524681334,
      "language_loss": 0.85043657,
      "learning_rate": 1.0434229670818618e-07,
      "loss": 0.87205529,
      "num_input_tokens_seen": 161789755,
      "step": 7483,
      "time_per_iteration": 3.267090320587158
    },
    {
      "auxiliary_loss_clip": 0.01129703,
      "auxiliary_loss_mlp": 0.01022938,
      "balance_loss_clip": 1.04197001,
      "balance_loss_mlp": 1.01560652,
      "epoch": 0.8998977935429567,
      "flos": 24166768095360.0,
      "grad_norm": 1.8458140546701203,
      "language_loss": 0.79887593,
      "learning_rate": 1.0409412018691944e-07,
      "loss": 0.82040238,
      "num_input_tokens_seen": 161810220,
      "step": 7484,
      "time_per_iteration": 2.478732109069824
    },
    {
      "auxiliary_loss_clip": 0.01134222,
      "auxiliary_loss_mlp": 0.010253,
      "balance_loss_clip": 1.04471016,
      "balance_loss_mlp": 1.01820457,
      "epoch": 0.9000180364335959,
      "flos": 20773030273920.0,
      "grad_norm": 2.04131354953962,
      "language_loss": 0.74898016,
      "learning_rate": 1.0384623127306724e-07,
      "loss": 0.77057534,
      "num_input_tokens_seen": 161827565,
      "step": 7485,
      "time_per_iteration": 2.4748785495758057
    },
    {
      "auxiliary_loss_clip": 0.01120198,
      "auxiliary_loss_mlp": 0.01019777,
      "balance_loss_clip": 1.04142404,
      "balance_loss_mlp": 1.01309812,
      "epoch": 0.900138279324235,
      "flos": 19205745166080.0,
      "grad_norm": 1.7764018029115372,
      "language_loss": 0.79428732,
      "learning_rate": 1.0359863000423397e-07,
      "loss": 0.81568706,
      "num_input_tokens_seen": 161845700,
      "step": 7486,
      "time_per_iteration": 2.4852969646453857
    },
    {
      "auxiliary_loss_clip": 0.01166592,
      "auxiliary_loss_mlp": 0.01024787,
      "balance_loss_clip": 1.04826248,
      "balance_loss_mlp": 1.01786065,
      "epoch": 0.900258522214874,
      "flos": 28731158069760.0,
      "grad_norm": 1.7360435892239923,
      "language_loss": 0.71823859,
      "learning_rate": 1.0335131641798112e-07,
      "loss": 0.74015236,
      "num_input_tokens_seen": 161867660,
      "step": 7487,
      "time_per_iteration": 2.4582371711730957
    },
    {
      "auxiliary_loss_clip": 0.01041264,
      "auxiliary_loss_mlp": 0.01001337,
      "balance_loss_clip": 1.00787663,
      "balance_loss_mlp": 1.00045443,
      "epoch": 0.9003787651055132,
      "flos": 58280685655680.0,
      "grad_norm": 0.9647917465211733,
      "language_loss": 0.55606437,
      "learning_rate": 1.0310429055182512e-07,
      "loss": 0.57649028,
      "num_input_tokens_seen": 161921980,
      "step": 7488,
      "time_per_iteration": 2.888615131378174
    },
    {
      "auxiliary_loss_clip": 0.01123851,
      "auxiliary_loss_mlp": 0.01028221,
      "balance_loss_clip": 1.04195583,
      "balance_loss_mlp": 1.02096105,
      "epoch": 0.9004990079961522,
      "flos": 25556475340800.0,
      "grad_norm": 1.6885474998831453,
      "language_loss": 0.74030489,
      "learning_rate": 1.0285755244324024e-07,
      "loss": 0.76182562,
      "num_input_tokens_seen": 161942725,
      "step": 7489,
      "time_per_iteration": 2.5354793071746826
    },
    {
      "auxiliary_loss_clip": 0.01139014,
      "auxiliary_loss_mlp": 0.00761441,
      "balance_loss_clip": 1.04192805,
      "balance_loss_mlp": 1.0004847,
      "epoch": 0.9006192508867913,
      "flos": 23335185352320.0,
      "grad_norm": 2.9019924795602194,
      "language_loss": 0.68757826,
      "learning_rate": 1.0261110212965629e-07,
      "loss": 0.70658278,
      "num_input_tokens_seen": 161964520,
      "step": 7490,
      "time_per_iteration": 3.296309471130371
    },
    {
      "auxiliary_loss_clip": 0.01137709,
      "auxiliary_loss_mlp": 0.01025712,
      "balance_loss_clip": 1.04443598,
      "balance_loss_mlp": 1.01904547,
      "epoch": 0.9007394937774305,
      "flos": 18040300485120.0,
      "grad_norm": 1.889246220983641,
      "language_loss": 0.7922442,
      "learning_rate": 1.023649396484596e-07,
      "loss": 0.81387842,
      "num_input_tokens_seen": 161983575,
      "step": 7491,
      "time_per_iteration": 2.449601173400879
    },
    {
      "auxiliary_loss_clip": 0.01164072,
      "auxiliary_loss_mlp": 0.01024508,
      "balance_loss_clip": 1.04584169,
      "balance_loss_mlp": 1.01798701,
      "epoch": 0.9008597366680695,
      "flos": 43068456633600.0,
      "grad_norm": 1.8900207930832382,
      "language_loss": 0.67653573,
      "learning_rate": 1.0211906503699275e-07,
      "loss": 0.69842148,
      "num_input_tokens_seen": 162006550,
      "step": 7492,
      "time_per_iteration": 2.5934319496154785
    },
    {
      "auxiliary_loss_clip": 0.01154813,
      "auxiliary_loss_mlp": 0.01027674,
      "balance_loss_clip": 1.04832292,
      "balance_loss_mlp": 1.01970506,
      "epoch": 0.9009799795587086,
      "flos": 14939055112320.0,
      "grad_norm": 2.452430364775128,
      "language_loss": 0.82468253,
      "learning_rate": 1.0187347833255455e-07,
      "loss": 0.84650743,
      "num_input_tokens_seen": 162022455,
      "step": 7493,
      "time_per_iteration": 2.398282527923584
    },
    {
      "auxiliary_loss_clip": 0.01162646,
      "auxiliary_loss_mlp": 0.01024284,
      "balance_loss_clip": 1.04701686,
      "balance_loss_mlp": 1.0173043,
      "epoch": 0.9011002224493477,
      "flos": 21579584215680.0,
      "grad_norm": 1.635963031346961,
      "language_loss": 0.79221272,
      "learning_rate": 1.0162817957240056e-07,
      "loss": 0.81408203,
      "num_input_tokens_seen": 162042350,
      "step": 7494,
      "time_per_iteration": 2.407040596008301
    },
    {
      "auxiliary_loss_clip": 0.0105174,
      "auxiliary_loss_mlp": 0.01001423,
      "balance_loss_clip": 1.00783718,
      "balance_loss_mlp": 1.0005703,
      "epoch": 0.9012204653399868,
      "flos": 71166367883520.0,
      "grad_norm": 0.8835582868496925,
      "language_loss": 0.63025457,
      "learning_rate": 1.0138316879374253e-07,
      "loss": 0.65078616,
      "num_input_tokens_seen": 162111640,
      "step": 7495,
      "time_per_iteration": 3.1716079711914062
    },
    {
      "auxiliary_loss_clip": 0.01141133,
      "auxiliary_loss_mlp": 0.01021399,
      "balance_loss_clip": 1.04807615,
      "balance_loss_mlp": 1.01397514,
      "epoch": 0.9013407082306258,
      "flos": 15594963413760.0,
      "grad_norm": 3.6988235192551513,
      "language_loss": 0.74481285,
      "learning_rate": 1.0113844603374833e-07,
      "loss": 0.76643813,
      "num_input_tokens_seen": 162128165,
      "step": 7496,
      "time_per_iteration": 2.4523158073425293
    },
    {
      "auxiliary_loss_clip": 0.01136963,
      "auxiliary_loss_mlp": 0.01024187,
      "balance_loss_clip": 1.04221725,
      "balance_loss_mlp": 1.01631284,
      "epoch": 0.901460951121265,
      "flos": 15049157276160.0,
      "grad_norm": 3.226879365709535,
      "language_loss": 0.72016931,
      "learning_rate": 1.0089401132954178e-07,
      "loss": 0.74178082,
      "num_input_tokens_seen": 162146145,
      "step": 7497,
      "time_per_iteration": 2.4674620628356934
    },
    {
      "auxiliary_loss_clip": 0.01139322,
      "auxiliary_loss_mlp": 0.0102361,
      "balance_loss_clip": 1.04646587,
      "balance_loss_mlp": 1.01690114,
      "epoch": 0.9015811940119041,
      "flos": 22236857233920.0,
      "grad_norm": 1.580516604419315,
      "language_loss": 0.72286189,
      "learning_rate": 1.006498647182037e-07,
      "loss": 0.74449122,
      "num_input_tokens_seen": 162164800,
      "step": 7498,
      "time_per_iteration": 2.483145236968994
    },
    {
      "auxiliary_loss_clip": 0.01092436,
      "auxiliary_loss_mlp": 0.01027796,
      "balance_loss_clip": 1.03913879,
      "balance_loss_mlp": 1.0205512,
      "epoch": 0.9017014369025431,
      "flos": 24973824827520.0,
      "grad_norm": 2.358434843356064,
      "language_loss": 0.7148248,
      "learning_rate": 1.004060062367713e-07,
      "loss": 0.73602712,
      "num_input_tokens_seen": 162185895,
      "step": 7499,
      "time_per_iteration": 2.6162989139556885
    },
    {
      "auxiliary_loss_clip": 0.0115358,
      "auxiliary_loss_mlp": 0.01023289,
      "balance_loss_clip": 1.04602504,
      "balance_loss_mlp": 1.01587462,
      "epoch": 0.9018216797931822,
      "flos": 18114168804480.0,
      "grad_norm": 1.739161401098802,
      "language_loss": 0.69203043,
      "learning_rate": 1.0016243592223728e-07,
      "loss": 0.71379912,
      "num_input_tokens_seen": 162206295,
      "step": 7500,
      "time_per_iteration": 2.4434964656829834
    },
    {
      "auxiliary_loss_clip": 0.01094081,
      "auxiliary_loss_mlp": 0.01024004,
      "balance_loss_clip": 1.04049206,
      "balance_loss_mlp": 1.01689661,
      "epoch": 0.9019419226838213,
      "flos": 37268452759680.0,
      "grad_norm": 1.856609138793378,
      "language_loss": 0.65798825,
      "learning_rate": 9.991915381155114e-08,
      "loss": 0.67916918,
      "num_input_tokens_seen": 162229275,
      "step": 7501,
      "time_per_iteration": 2.699352502822876
    },
    {
      "auxiliary_loss_clip": 0.01154714,
      "auxiliary_loss_mlp": 0.01023093,
      "balance_loss_clip": 1.04588509,
      "balance_loss_mlp": 1.01587462,
      "epoch": 0.9020621655744604,
      "flos": 23441121538560.0,
      "grad_norm": 2.0112876345868647,
      "language_loss": 0.74872202,
      "learning_rate": 9.967615994161871e-08,
      "loss": 0.77050006,
      "num_input_tokens_seen": 162248935,
      "step": 7502,
      "time_per_iteration": 2.4711592197418213
    },
    {
      "auxiliary_loss_clip": 0.0116393,
      "auxiliary_loss_mlp": 0.01021335,
      "balance_loss_clip": 1.04577315,
      "balance_loss_mlp": 1.01435566,
      "epoch": 0.9021824084650995,
      "flos": 22857465444480.0,
      "grad_norm": 1.7030176325353608,
      "language_loss": 0.78332675,
      "learning_rate": 9.943345434930161e-08,
      "loss": 0.80517936,
      "num_input_tokens_seen": 162269185,
      "step": 7503,
      "time_per_iteration": 2.4451146125793457
    },
    {
      "auxiliary_loss_clip": 0.01123429,
      "auxiliary_loss_mlp": 0.01023997,
      "balance_loss_clip": 1.04454303,
      "balance_loss_mlp": 1.01692176,
      "epoch": 0.9023026513557386,
      "flos": 22127581082880.0,
      "grad_norm": 1.8910205186235143,
      "language_loss": 0.68888879,
      "learning_rate": 9.919103707141885e-08,
      "loss": 0.71036303,
      "num_input_tokens_seen": 162288065,
      "step": 7504,
      "time_per_iteration": 2.504936456680298
    },
    {
      "auxiliary_loss_clip": 0.01149092,
      "auxiliary_loss_mlp": 0.01025003,
      "balance_loss_clip": 1.04483533,
      "balance_loss_mlp": 1.01733744,
      "epoch": 0.9024228942463777,
      "flos": 24199087357440.0,
      "grad_norm": 1.8683829897163502,
      "language_loss": 0.76707155,
      "learning_rate": 9.89489081447441e-08,
      "loss": 0.78881252,
      "num_input_tokens_seen": 162305265,
      "step": 7505,
      "time_per_iteration": 2.436614751815796
    },
    {
      "auxiliary_loss_clip": 0.01136553,
      "auxiliary_loss_mlp": 0.01021817,
      "balance_loss_clip": 1.04272437,
      "balance_loss_mlp": 1.0144136,
      "epoch": 0.9025431371370167,
      "flos": 25008262992000.0,
      "grad_norm": 1.8462170707015797,
      "language_loss": 0.83278984,
      "learning_rate": 9.870706760600844e-08,
      "loss": 0.85437357,
      "num_input_tokens_seen": 162325215,
      "step": 7506,
      "time_per_iteration": 2.4989330768585205
    },
    {
      "auxiliary_loss_clip": 0.01118712,
      "auxiliary_loss_mlp": 0.01026842,
      "balance_loss_clip": 1.04765582,
      "balance_loss_mlp": 1.01949906,
      "epoch": 0.9026633800276559,
      "flos": 18952862440320.0,
      "grad_norm": 2.492504751861586,
      "language_loss": 0.72674954,
      "learning_rate": 9.846551549189918e-08,
      "loss": 0.74820513,
      "num_input_tokens_seen": 162344820,
      "step": 7507,
      "time_per_iteration": 3.3484420776367188
    },
    {
      "auxiliary_loss_clip": 0.01135015,
      "auxiliary_loss_mlp": 0.01023458,
      "balance_loss_clip": 1.04397464,
      "balance_loss_mlp": 1.01621294,
      "epoch": 0.902783622918295,
      "flos": 32416059536640.0,
      "grad_norm": 1.9638969202479812,
      "language_loss": 0.68712258,
      "learning_rate": 9.822425183905902e-08,
      "loss": 0.70870733,
      "num_input_tokens_seen": 162365345,
      "step": 7508,
      "time_per_iteration": 2.557969808578491
    },
    {
      "auxiliary_loss_clip": 0.01032407,
      "auxiliary_loss_mlp": 0.01001098,
      "balance_loss_clip": 1.00843966,
      "balance_loss_mlp": 1.00021541,
      "epoch": 0.902903865808934,
      "flos": 63717453244800.0,
      "grad_norm": 0.9300813243559274,
      "language_loss": 0.7522254,
      "learning_rate": 9.798327668408823e-08,
      "loss": 0.77256048,
      "num_input_tokens_seen": 162426980,
      "step": 7509,
      "time_per_iteration": 3.8911752700805664
    },
    {
      "auxiliary_loss_clip": 0.0116887,
      "auxiliary_loss_mlp": 0.01025138,
      "balance_loss_clip": 1.0476203,
      "balance_loss_mlp": 1.01756859,
      "epoch": 0.9030241086995732,
      "flos": 23804034600960.0,
      "grad_norm": 2.3033024802018858,
      "language_loss": 0.68901622,
      "learning_rate": 9.774259006354158e-08,
      "loss": 0.71095628,
      "num_input_tokens_seen": 162447050,
      "step": 7510,
      "time_per_iteration": 3.165647268295288
    },
    {
      "auxiliary_loss_clip": 0.0114119,
      "auxiliary_loss_mlp": 0.01025986,
      "balance_loss_clip": 1.0435791,
      "balance_loss_mlp": 1.01893759,
      "epoch": 0.9031443515902122,
      "flos": 26395887248640.0,
      "grad_norm": 1.9847206849650965,
      "language_loss": 0.75995255,
      "learning_rate": 9.750219201393184e-08,
      "loss": 0.78162432,
      "num_input_tokens_seen": 162467015,
      "step": 7511,
      "time_per_iteration": 2.5126726627349854
    },
    {
      "auxiliary_loss_clip": 0.0114898,
      "auxiliary_loss_mlp": 0.01019428,
      "balance_loss_clip": 1.0442965,
      "balance_loss_mlp": 1.0123018,
      "epoch": 0.9032645944808513,
      "flos": 24939350749440.0,
      "grad_norm": 1.922756282202372,
      "language_loss": 0.77448583,
      "learning_rate": 9.726208257172697e-08,
      "loss": 0.79616994,
      "num_input_tokens_seen": 162488710,
      "step": 7512,
      "time_per_iteration": 2.4924213886260986
    },
    {
      "auxiliary_loss_clip": 0.01164692,
      "auxiliary_loss_mlp": 0.01022105,
      "balance_loss_clip": 1.04710519,
      "balance_loss_mlp": 1.0151639,
      "epoch": 0.9033848373714904,
      "flos": 21178821196800.0,
      "grad_norm": 2.6589378993722392,
      "language_loss": 0.74869305,
      "learning_rate": 9.702226177335115e-08,
      "loss": 0.77056104,
      "num_input_tokens_seen": 162507205,
      "step": 7513,
      "time_per_iteration": 2.4249424934387207
    },
    {
      "auxiliary_loss_clip": 0.01136297,
      "auxiliary_loss_mlp": 0.0103205,
      "balance_loss_clip": 1.04524136,
      "balance_loss_mlp": 1.02432549,
      "epoch": 0.9035050802621295,
      "flos": 26286359702400.0,
      "grad_norm": 1.7505175252544227,
      "language_loss": 0.72428823,
      "learning_rate": 9.67827296551853e-08,
      "loss": 0.7459718,
      "num_input_tokens_seen": 162528490,
      "step": 7514,
      "time_per_iteration": 2.5159690380096436
    },
    {
      "auxiliary_loss_clip": 0.01129064,
      "auxiliary_loss_mlp": 0.00761706,
      "balance_loss_clip": 1.040609,
      "balance_loss_mlp": 1.00048304,
      "epoch": 0.9036253231527686,
      "flos": 24204546224640.0,
      "grad_norm": 2.1967319204783156,
      "language_loss": 0.68626791,
      "learning_rate": 9.65434862535659e-08,
      "loss": 0.70517558,
      "num_input_tokens_seen": 162547860,
      "step": 7515,
      "time_per_iteration": 2.486316204071045
    },
    {
      "auxiliary_loss_clip": 0.01138782,
      "auxiliary_loss_mlp": 0.01029733,
      "balance_loss_clip": 1.04319763,
      "balance_loss_mlp": 1.0222944,
      "epoch": 0.9037455660434077,
      "flos": 18072655660800.0,
      "grad_norm": 3.026203917082258,
      "language_loss": 0.64671063,
      "learning_rate": 9.630453160478635e-08,
      "loss": 0.66839582,
      "num_input_tokens_seen": 162563215,
      "step": 7516,
      "time_per_iteration": 2.4332194328308105
    },
    {
      "auxiliary_loss_clip": 0.01111823,
      "auxiliary_loss_mlp": 0.01022925,
      "balance_loss_clip": 1.04214573,
      "balance_loss_mlp": 1.01568878,
      "epoch": 0.9038658089340468,
      "flos": 24060795995520.0,
      "grad_norm": 1.6804954566547874,
      "language_loss": 0.82410437,
      "learning_rate": 9.60658657450959e-08,
      "loss": 0.84545183,
      "num_input_tokens_seen": 162583515,
      "step": 7517,
      "time_per_iteration": 3.349963426589966
    },
    {
      "auxiliary_loss_clip": 0.01122738,
      "auxiliary_loss_mlp": 0.01024807,
      "balance_loss_clip": 1.04015434,
      "balance_loss_mlp": 1.01776767,
      "epoch": 0.9039860518246858,
      "flos": 21834298535040.0,
      "grad_norm": 1.5632127066299444,
      "language_loss": 0.79289317,
      "learning_rate": 9.582748871069979e-08,
      "loss": 0.81436861,
      "num_input_tokens_seen": 162602955,
      "step": 7518,
      "time_per_iteration": 2.5092217922210693
    },
    {
      "auxiliary_loss_clip": 0.01138744,
      "auxiliary_loss_mlp": 0.00761296,
      "balance_loss_clip": 1.04273462,
      "balance_loss_mlp": 1.00042844,
      "epoch": 0.904106294715325,
      "flos": 26614870513920.0,
      "grad_norm": 2.177561591943742,
      "language_loss": 0.82648271,
      "learning_rate": 9.558940053775954e-08,
      "loss": 0.84548306,
      "num_input_tokens_seen": 162621595,
      "step": 7519,
      "time_per_iteration": 2.578040599822998
    },
    {
      "auxiliary_loss_clip": 0.01149752,
      "auxiliary_loss_mlp": 0.01026148,
      "balance_loss_clip": 1.04571056,
      "balance_loss_mlp": 1.01901889,
      "epoch": 0.904226537605964,
      "flos": 17785693906560.0,
      "grad_norm": 6.918367209100176,
      "language_loss": 0.68074358,
      "learning_rate": 9.535160126239294e-08,
      "loss": 0.70250249,
      "num_input_tokens_seen": 162638220,
      "step": 7520,
      "time_per_iteration": 2.4742648601531982
    },
    {
      "auxiliary_loss_clip": 0.01149248,
      "auxiliary_loss_mlp": 0.01025002,
      "balance_loss_clip": 1.04616082,
      "balance_loss_mlp": 1.01804328,
      "epoch": 0.9043467804966031,
      "flos": 24790428961920.0,
      "grad_norm": 1.5518936283260256,
      "language_loss": 0.70606309,
      "learning_rate": 9.511409092067424e-08,
      "loss": 0.72780561,
      "num_input_tokens_seen": 162658575,
      "step": 7521,
      "time_per_iteration": 2.55526065826416
    },
    {
      "auxiliary_loss_clip": 0.01139797,
      "auxiliary_loss_mlp": 0.01022345,
      "balance_loss_clip": 1.0469507,
      "balance_loss_mlp": 1.01537108,
      "epoch": 0.9044670233872423,
      "flos": 22632125472000.0,
      "grad_norm": 1.7966911876723506,
      "language_loss": 0.67536676,
      "learning_rate": 9.487686954863327e-08,
      "loss": 0.69698817,
      "num_input_tokens_seen": 162678295,
      "step": 7522,
      "time_per_iteration": 2.4851834774017334
    },
    {
      "auxiliary_loss_clip": 0.01149355,
      "auxiliary_loss_mlp": 0.01023135,
      "balance_loss_clip": 1.04549479,
      "balance_loss_mlp": 1.01616693,
      "epoch": 0.9045872662778813,
      "flos": 23771320289280.0,
      "grad_norm": 5.003024867350654,
      "language_loss": 0.76886868,
      "learning_rate": 9.46399371822566e-08,
      "loss": 0.79059362,
      "num_input_tokens_seen": 162698070,
      "step": 7523,
      "time_per_iteration": 2.4811840057373047
    },
    {
      "auxiliary_loss_clip": 0.01165438,
      "auxiliary_loss_mlp": 0.01023687,
      "balance_loss_clip": 1.04738045,
      "balance_loss_mlp": 1.01633167,
      "epoch": 0.9047075091685204,
      "flos": 15191039998080.0,
      "grad_norm": 3.0669154921102857,
      "language_loss": 0.72337472,
      "learning_rate": 9.440329385748657e-08,
      "loss": 0.74526596,
      "num_input_tokens_seen": 162715140,
      "step": 7524,
      "time_per_iteration": 2.378437042236328
    },
    {
      "auxiliary_loss_clip": 0.0112303,
      "auxiliary_loss_mlp": 0.01017748,
      "balance_loss_clip": 1.04415119,
      "balance_loss_mlp": 1.01148367,
      "epoch": 0.9048277520591596,
      "flos": 18003707504640.0,
      "grad_norm": 1.6884552421013848,
      "language_loss": 0.70513642,
      "learning_rate": 9.416693961022137e-08,
      "loss": 0.72654414,
      "num_input_tokens_seen": 162733390,
      "step": 7525,
      "time_per_iteration": 2.4754233360290527
    },
    {
      "auxiliary_loss_clip": 0.01083976,
      "auxiliary_loss_mlp": 0.0102422,
      "balance_loss_clip": 1.03895688,
      "balance_loss_mlp": 1.01720452,
      "epoch": 0.9049479949497986,
      "flos": 21872471713920.0,
      "grad_norm": 1.6828772868579742,
      "language_loss": 0.77079493,
      "learning_rate": 9.393087447631654e-08,
      "loss": 0.79187691,
      "num_input_tokens_seen": 162751670,
      "step": 7526,
      "time_per_iteration": 2.560612678527832
    },
    {
      "auxiliary_loss_clip": 0.01137852,
      "auxiliary_loss_mlp": 0.01019729,
      "balance_loss_clip": 1.04315889,
      "balance_loss_mlp": 1.01331544,
      "epoch": 0.9050682378404377,
      "flos": 20773928113920.0,
      "grad_norm": 1.8653675147424515,
      "language_loss": 0.72918928,
      "learning_rate": 9.36950984915823e-08,
      "loss": 0.75076509,
      "num_input_tokens_seen": 162770025,
      "step": 7527,
      "time_per_iteration": 2.4717085361480713
    },
    {
      "auxiliary_loss_clip": 0.01167262,
      "auxiliary_loss_mlp": 0.01025271,
      "balance_loss_clip": 1.0491178,
      "balance_loss_mlp": 1.01799929,
      "epoch": 0.9051884807310768,
      "flos": 21580015178880.0,
      "grad_norm": 1.7162262224571077,
      "language_loss": 0.6927138,
      "learning_rate": 9.345961169178607e-08,
      "loss": 0.71463913,
      "num_input_tokens_seen": 162789710,
      "step": 7528,
      "time_per_iteration": 2.4101996421813965
    },
    {
      "auxiliary_loss_clip": 0.01113461,
      "auxiliary_loss_mlp": 0.01026406,
      "balance_loss_clip": 1.04762328,
      "balance_loss_mlp": 1.01943541,
      "epoch": 0.9053087236217159,
      "flos": 21908059113600.0,
      "grad_norm": 1.4055299782988353,
      "language_loss": 0.7284615,
      "learning_rate": 9.322441411265081e-08,
      "loss": 0.74986017,
      "num_input_tokens_seen": 162810695,
      "step": 7529,
      "time_per_iteration": 2.531101703643799
    },
    {
      "auxiliary_loss_clip": 0.01133867,
      "auxiliary_loss_mlp": 0.01024464,
      "balance_loss_clip": 1.04462218,
      "balance_loss_mlp": 1.01745772,
      "epoch": 0.9054289665123549,
      "flos": 17055809544960.0,
      "grad_norm": 1.9134742899752655,
      "language_loss": 0.72953081,
      "learning_rate": 9.298950578985554e-08,
      "loss": 0.75111413,
      "num_input_tokens_seen": 162827770,
      "step": 7530,
      "time_per_iteration": 2.532728672027588
    },
    {
      "auxiliary_loss_clip": 0.01146184,
      "auxiliary_loss_mlp": 0.00762121,
      "balance_loss_clip": 1.04526711,
      "balance_loss_mlp": 1.00035715,
      "epoch": 0.905549209402994,
      "flos": 20777268078720.0,
      "grad_norm": 1.6917842593366994,
      "language_loss": 0.70836216,
      "learning_rate": 9.275488675903665e-08,
      "loss": 0.72744524,
      "num_input_tokens_seen": 162846715,
      "step": 7531,
      "time_per_iteration": 2.448601722717285
    },
    {
      "auxiliary_loss_clip": 0.01109131,
      "auxiliary_loss_mlp": 0.0102543,
      "balance_loss_clip": 1.04351771,
      "balance_loss_mlp": 1.01837349,
      "epoch": 0.9056694522936332,
      "flos": 21686813291520.0,
      "grad_norm": 1.9192459311768664,
      "language_loss": 0.73923755,
      "learning_rate": 9.252055705578454e-08,
      "loss": 0.76058316,
      "num_input_tokens_seen": 162866215,
      "step": 7532,
      "time_per_iteration": 2.5558061599731445
    },
    {
      "auxiliary_loss_clip": 0.01149128,
      "auxiliary_loss_mlp": 0.01024256,
      "balance_loss_clip": 1.04382646,
      "balance_loss_mlp": 1.017717,
      "epoch": 0.9057896951842722,
      "flos": 29569133433600.0,
      "grad_norm": 1.6143226344728039,
      "language_loss": 0.71910238,
      "learning_rate": 9.228651671564747e-08,
      "loss": 0.74083626,
      "num_input_tokens_seen": 162888245,
      "step": 7533,
      "time_per_iteration": 2.512439012527466
    },
    {
      "auxiliary_loss_clip": 0.01104851,
      "auxiliary_loss_mlp": 0.01024418,
      "balance_loss_clip": 1.04348254,
      "balance_loss_mlp": 1.01763821,
      "epoch": 0.9059099380749113,
      "flos": 27892248952320.0,
      "grad_norm": 1.567067747710581,
      "language_loss": 0.77892238,
      "learning_rate": 9.205276577412901e-08,
      "loss": 0.80021507,
      "num_input_tokens_seen": 162911025,
      "step": 7534,
      "time_per_iteration": 3.405759572982788
    },
    {
      "auxiliary_loss_clip": 0.0114276,
      "auxiliary_loss_mlp": 0.00761577,
      "balance_loss_clip": 1.0430876,
      "balance_loss_mlp": 1.00043666,
      "epoch": 0.9060301809655504,
      "flos": 17748993185280.0,
      "grad_norm": 2.6138948860132687,
      "language_loss": 0.77097189,
      "learning_rate": 9.181930426668905e-08,
      "loss": 0.79001522,
      "num_input_tokens_seen": 162927820,
      "step": 7535,
      "time_per_iteration": 2.4517784118652344
    },
    {
      "auxiliary_loss_clip": 0.01108211,
      "auxiliary_loss_mlp": 0.01023992,
      "balance_loss_clip": 1.04355836,
      "balance_loss_mlp": 1.01737297,
      "epoch": 0.9061504238561895,
      "flos": 31759432963200.0,
      "grad_norm": 1.5695747644015576,
      "language_loss": 0.67939436,
      "learning_rate": 9.158613222874346e-08,
      "loss": 0.70071638,
      "num_input_tokens_seen": 162949445,
      "step": 7536,
      "time_per_iteration": 4.1521477699279785
    },
    {
      "auxiliary_loss_clip": 0.01134582,
      "auxiliary_loss_mlp": 0.01020546,
      "balance_loss_clip": 1.04255736,
      "balance_loss_mlp": 1.0136584,
      "epoch": 0.9062706667468285,
      "flos": 20048066075520.0,
      "grad_norm": 1.625523697874035,
      "language_loss": 0.81722605,
      "learning_rate": 9.135324969566394e-08,
      "loss": 0.8387773,
      "num_input_tokens_seen": 162968945,
      "step": 7537,
      "time_per_iteration": 2.476557493209839
    },
    {
      "auxiliary_loss_clip": 0.0115614,
      "auxiliary_loss_mlp": 0.01025756,
      "balance_loss_clip": 1.0471549,
      "balance_loss_mlp": 1.01867819,
      "epoch": 0.9063909096374677,
      "flos": 18437292576000.0,
      "grad_norm": 1.785844906890705,
      "language_loss": 0.75357199,
      "learning_rate": 9.112065670277913e-08,
      "loss": 0.77539092,
      "num_input_tokens_seen": 162985310,
      "step": 7538,
      "time_per_iteration": 2.420950174331665
    },
    {
      "auxiliary_loss_clip": 0.01134283,
      "auxiliary_loss_mlp": 0.01022287,
      "balance_loss_clip": 1.04285014,
      "balance_loss_mlp": 1.01562071,
      "epoch": 0.9065111525281068,
      "flos": 33547353361920.0,
      "grad_norm": 2.1834346437163754,
      "language_loss": 0.72660363,
      "learning_rate": 9.088835328537303e-08,
      "loss": 0.7481693,
      "num_input_tokens_seen": 163006900,
      "step": 7539,
      "time_per_iteration": 2.586735248565674
    },
    {
      "auxiliary_loss_clip": 0.0114094,
      "auxiliary_loss_mlp": 0.01023267,
      "balance_loss_clip": 1.04530478,
      "balance_loss_mlp": 1.01602483,
      "epoch": 0.9066313954187458,
      "flos": 23367863750400.0,
      "grad_norm": 11.837360681386912,
      "language_loss": 0.71457005,
      "learning_rate": 9.065633947868568e-08,
      "loss": 0.73621213,
      "num_input_tokens_seen": 163026505,
      "step": 7540,
      "time_per_iteration": 2.5033652782440186
    },
    {
      "auxiliary_loss_clip": 0.01123452,
      "auxiliary_loss_mlp": 0.00761718,
      "balance_loss_clip": 1.04577088,
      "balance_loss_mlp": 1.00037241,
      "epoch": 0.906751638309385,
      "flos": 26249623067520.0,
      "grad_norm": 2.3554558908120646,
      "language_loss": 0.79849142,
      "learning_rate": 9.042461531791379e-08,
      "loss": 0.81734312,
      "num_input_tokens_seen": 163044925,
      "step": 7541,
      "time_per_iteration": 2.55106520652771
    },
    {
      "auxiliary_loss_clip": 0.01161096,
      "auxiliary_loss_mlp": 0.01022355,
      "balance_loss_clip": 1.04542935,
      "balance_loss_mlp": 1.01575983,
      "epoch": 0.906871881200024,
      "flos": 16544477485440.0,
      "grad_norm": 1.7459686267517114,
      "language_loss": 0.78215909,
      "learning_rate": 9.019318083820903e-08,
      "loss": 0.80399358,
      "num_input_tokens_seen": 163063505,
      "step": 7542,
      "time_per_iteration": 2.4267797470092773
    },
    {
      "auxiliary_loss_clip": 0.01150141,
      "auxiliary_loss_mlp": 0.01029084,
      "balance_loss_clip": 1.04700756,
      "balance_loss_mlp": 1.02212477,
      "epoch": 0.9069921240906631,
      "flos": 24605129675520.0,
      "grad_norm": 1.6561832999381751,
      "language_loss": 0.85298896,
      "learning_rate": 8.996203607468045e-08,
      "loss": 0.87478125,
      "num_input_tokens_seen": 163082505,
      "step": 7543,
      "time_per_iteration": 2.4794650077819824
    },
    {
      "auxiliary_loss_clip": 0.01147579,
      "auxiliary_loss_mlp": 0.01025188,
      "balance_loss_clip": 1.04332185,
      "balance_loss_mlp": 1.01804733,
      "epoch": 0.9071123669813023,
      "flos": 25374731500800.0,
      "grad_norm": 1.503951133656035,
      "language_loss": 0.75568414,
      "learning_rate": 8.973118106239241e-08,
      "loss": 0.77741194,
      "num_input_tokens_seen": 163105110,
      "step": 7544,
      "time_per_iteration": 3.316143274307251
    },
    {
      "auxiliary_loss_clip": 0.01092663,
      "auxiliary_loss_mlp": 0.01027414,
      "balance_loss_clip": 1.03725302,
      "balance_loss_mlp": 1.02018106,
      "epoch": 0.9072326098719413,
      "flos": 26725798690560.0,
      "grad_norm": 1.9916631437492198,
      "language_loss": 0.94762319,
      "learning_rate": 8.95006158363656e-08,
      "loss": 0.96882391,
      "num_input_tokens_seen": 163125295,
      "step": 7545,
      "time_per_iteration": 2.6222267150878906
    },
    {
      "auxiliary_loss_clip": 0.01152857,
      "auxiliary_loss_mlp": 0.0102901,
      "balance_loss_clip": 1.04871011,
      "balance_loss_mlp": 1.02144623,
      "epoch": 0.9073528527625804,
      "flos": 23878800760320.0,
      "grad_norm": 1.8083816990157908,
      "language_loss": 0.77342892,
      "learning_rate": 8.9270340431576e-08,
      "loss": 0.79524761,
      "num_input_tokens_seen": 163144385,
      "step": 7546,
      "time_per_iteration": 2.461282253265381
    },
    {
      "auxiliary_loss_clip": 0.01153047,
      "auxiliary_loss_mlp": 0.01025931,
      "balance_loss_clip": 1.04460895,
      "balance_loss_mlp": 1.01890039,
      "epoch": 0.9074730956532195,
      "flos": 37852144767360.0,
      "grad_norm": 2.0768832205644254,
      "language_loss": 0.7363615,
      "learning_rate": 8.904035488295658e-08,
      "loss": 0.75815123,
      "num_input_tokens_seen": 163163885,
      "step": 7547,
      "time_per_iteration": 2.592989683151245
    },
    {
      "auxiliary_loss_clip": 0.01051382,
      "auxiliary_loss_mlp": 0.00752984,
      "balance_loss_clip": 1.00768936,
      "balance_loss_mlp": 0.99991649,
      "epoch": 0.9075933385438586,
      "flos": 65173307385600.0,
      "grad_norm": 0.7052276313477072,
      "language_loss": 0.53276253,
      "learning_rate": 8.881065922539632e-08,
      "loss": 0.55080622,
      "num_input_tokens_seen": 163224325,
      "step": 7548,
      "time_per_iteration": 2.9635889530181885
    },
    {
      "auxiliary_loss_clip": 0.01115727,
      "auxiliary_loss_mlp": 0.01018214,
      "balance_loss_clip": 1.04232168,
      "balance_loss_mlp": 1.0118897,
      "epoch": 0.9077135814344977,
      "flos": 19931571290880.0,
      "grad_norm": 2.2110591409263756,
      "language_loss": 0.73235798,
      "learning_rate": 8.85812534937389e-08,
      "loss": 0.7536974,
      "num_input_tokens_seen": 163242425,
      "step": 7549,
      "time_per_iteration": 2.503549337387085
    },
    {
      "auxiliary_loss_clip": 0.01157539,
      "auxiliary_loss_mlp": 0.01026804,
      "balance_loss_clip": 1.04628873,
      "balance_loss_mlp": 1.01960421,
      "epoch": 0.9078338243251368,
      "flos": 17529650784000.0,
      "grad_norm": 2.5991638556843792,
      "language_loss": 0.67745119,
      "learning_rate": 8.835213772278583e-08,
      "loss": 0.69929463,
      "num_input_tokens_seen": 163259280,
      "step": 7550,
      "time_per_iteration": 2.405254602432251
    },
    {
      "auxiliary_loss_clip": 0.01114218,
      "auxiliary_loss_mlp": 0.01021718,
      "balance_loss_clip": 1.04352665,
      "balance_loss_mlp": 1.0150156,
      "epoch": 0.9079540672157759,
      "flos": 28803410277120.0,
      "grad_norm": 1.7372053746226181,
      "language_loss": 0.78826368,
      "learning_rate": 8.812331194729373e-08,
      "loss": 0.809623,
      "num_input_tokens_seen": 163278925,
      "step": 7551,
      "time_per_iteration": 2.5577621459960938
    },
    {
      "auxiliary_loss_clip": 0.01172598,
      "auxiliary_loss_mlp": 0.01024896,
      "balance_loss_clip": 1.05264294,
      "balance_loss_mlp": 1.01746345,
      "epoch": 0.9080743101064149,
      "flos": 23513840622720.0,
      "grad_norm": 1.8450358039537986,
      "language_loss": 0.72374737,
      "learning_rate": 8.789477620197461e-08,
      "loss": 0.74572229,
      "num_input_tokens_seen": 163298450,
      "step": 7552,
      "time_per_iteration": 2.4289121627807617
    },
    {
      "auxiliary_loss_clip": 0.0113478,
      "auxiliary_loss_mlp": 0.01025533,
      "balance_loss_clip": 1.04257774,
      "balance_loss_mlp": 1.01837099,
      "epoch": 0.9081945529970541,
      "flos": 22778102344320.0,
      "grad_norm": 2.3035225049834023,
      "language_loss": 0.79151344,
      "learning_rate": 8.766653052149831e-08,
      "loss": 0.81311661,
      "num_input_tokens_seen": 163313635,
      "step": 7553,
      "time_per_iteration": 2.4660696983337402
    },
    {
      "auxiliary_loss_clip": 0.0113367,
      "auxiliary_loss_mlp": 0.01026882,
      "balance_loss_clip": 1.04270899,
      "balance_loss_mlp": 1.01934838,
      "epoch": 0.9083147958876931,
      "flos": 18873714821760.0,
      "grad_norm": 2.1813363136202804,
      "language_loss": 0.74388784,
      "learning_rate": 8.743857494048823e-08,
      "loss": 0.76549333,
      "num_input_tokens_seen": 163330450,
      "step": 7554,
      "time_per_iteration": 2.4595232009887695
    },
    {
      "auxiliary_loss_clip": 0.01122732,
      "auxiliary_loss_mlp": 0.01027159,
      "balance_loss_clip": 1.04260564,
      "balance_loss_mlp": 1.01972318,
      "epoch": 0.9084350387783322,
      "flos": 18909374048640.0,
      "grad_norm": 2.3014308346613634,
      "language_loss": 0.62604713,
      "learning_rate": 8.721090949352605e-08,
      "loss": 0.64754605,
      "num_input_tokens_seen": 163346690,
      "step": 7555,
      "time_per_iteration": 2.4897663593292236
    },
    {
      "auxiliary_loss_clip": 0.01160267,
      "auxiliary_loss_mlp": 0.01025307,
      "balance_loss_clip": 1.04845309,
      "balance_loss_mlp": 1.0175879,
      "epoch": 0.9085552816689714,
      "flos": 20595488325120.0,
      "grad_norm": 2.959404170864298,
      "language_loss": 0.73229825,
      "learning_rate": 8.698353421514793e-08,
      "loss": 0.75415397,
      "num_input_tokens_seen": 163365065,
      "step": 7556,
      "time_per_iteration": 2.4372453689575195
    },
    {
      "auxiliary_loss_clip": 0.01152229,
      "auxiliary_loss_mlp": 0.01025786,
      "balance_loss_clip": 1.04782963,
      "balance_loss_mlp": 1.01907396,
      "epoch": 0.9086755245596104,
      "flos": 18113163223680.0,
      "grad_norm": 2.5466287973693644,
      "language_loss": 0.80263793,
      "learning_rate": 8.67564491398467e-08,
      "loss": 0.82441807,
      "num_input_tokens_seen": 163382070,
      "step": 7557,
      "time_per_iteration": 2.4207801818847656
    },
    {
      "auxiliary_loss_clip": 0.01152376,
      "auxiliary_loss_mlp": 0.01025431,
      "balance_loss_clip": 1.04449224,
      "balance_loss_mlp": 1.01765227,
      "epoch": 0.9087957674502495,
      "flos": 19129793857920.0,
      "grad_norm": 1.829358981837066,
      "language_loss": 0.74024117,
      "learning_rate": 8.652965430207104e-08,
      "loss": 0.76201928,
      "num_input_tokens_seen": 163399975,
      "step": 7558,
      "time_per_iteration": 2.4300026893615723
    },
    {
      "auxiliary_loss_clip": 0.01154412,
      "auxiliary_loss_mlp": 0.01027573,
      "balance_loss_clip": 1.04613686,
      "balance_loss_mlp": 1.01985383,
      "epoch": 0.9089160103408886,
      "flos": 18109930999680.0,
      "grad_norm": 2.7621351215237477,
      "language_loss": 0.65728498,
      "learning_rate": 8.630314973622521e-08,
      "loss": 0.6791048,
      "num_input_tokens_seen": 163417520,
      "step": 7559,
      "time_per_iteration": 2.4251675605773926
    },
    {
      "auxiliary_loss_clip": 0.01148022,
      "auxiliary_loss_mlp": 0.01023092,
      "balance_loss_clip": 1.0470233,
      "balance_loss_mlp": 1.01640701,
      "epoch": 0.9090362532315277,
      "flos": 33364855336320.0,
      "grad_norm": 6.0647576622088,
      "language_loss": 0.7082597,
      "learning_rate": 8.607693547666995e-08,
      "loss": 0.72997075,
      "num_input_tokens_seen": 163440060,
      "step": 7560,
      "time_per_iteration": 3.3738601207733154
    },
    {
      "auxiliary_loss_clip": 0.01032142,
      "auxiliary_loss_mlp": 0.01002092,
      "balance_loss_clip": 1.00794005,
      "balance_loss_mlp": 1.00115621,
      "epoch": 0.9091564961221668,
      "flos": 71480585082240.0,
      "grad_norm": 0.945148805249054,
      "language_loss": 0.58006293,
      "learning_rate": 8.585101155772201e-08,
      "loss": 0.60040528,
      "num_input_tokens_seen": 163502180,
      "step": 7561,
      "time_per_iteration": 3.174197196960449
    },
    {
      "auxiliary_loss_clip": 0.01128535,
      "auxiliary_loss_mlp": 0.01025505,
      "balance_loss_clip": 1.03911686,
      "balance_loss_mlp": 1.01817942,
      "epoch": 0.9092767390128058,
      "flos": 24712574232960.0,
      "grad_norm": 1.871363762117446,
      "language_loss": 0.68608797,
      "learning_rate": 8.562537801365377e-08,
      "loss": 0.70762837,
      "num_input_tokens_seen": 163521915,
      "step": 7562,
      "time_per_iteration": 3.236593246459961
    },
    {
      "auxiliary_loss_clip": 0.0116584,
      "auxiliary_loss_mlp": 0.01027418,
      "balance_loss_clip": 1.04670584,
      "balance_loss_mlp": 1.02020562,
      "epoch": 0.909396981903445,
      "flos": 23586487879680.0,
      "grad_norm": 1.7726976425605123,
      "language_loss": 0.70005274,
      "learning_rate": 8.540003487869362e-08,
      "loss": 0.72198522,
      "num_input_tokens_seen": 163543585,
      "step": 7563,
      "time_per_iteration": 3.2606747150421143
    },
    {
      "auxiliary_loss_clip": 0.0111132,
      "auxiliary_loss_mlp": 0.0102229,
      "balance_loss_clip": 1.04052711,
      "balance_loss_mlp": 1.01507771,
      "epoch": 0.909517224794084,
      "flos": 23404169422080.0,
      "grad_norm": 1.758883586674906,
      "language_loss": 0.80005574,
      "learning_rate": 8.517498218702557e-08,
      "loss": 0.82139182,
      "num_input_tokens_seen": 163561515,
      "step": 7564,
      "time_per_iteration": 2.5211315155029297
    },
    {
      "auxiliary_loss_clip": 0.0111875,
      "auxiliary_loss_mlp": 0.01020598,
      "balance_loss_clip": 1.0425452,
      "balance_loss_mlp": 1.01372004,
      "epoch": 0.9096374676847231,
      "flos": 19208618254080.0,
      "grad_norm": 1.602903414177587,
      "language_loss": 0.69473499,
      "learning_rate": 8.49502199727905e-08,
      "loss": 0.71612847,
      "num_input_tokens_seen": 163579540,
      "step": 7565,
      "time_per_iteration": 2.50368070602417
    },
    {
      "auxiliary_loss_clip": 0.01146143,
      "auxiliary_loss_mlp": 0.01025401,
      "balance_loss_clip": 1.04178071,
      "balance_loss_mlp": 1.01820064,
      "epoch": 0.9097577105753623,
      "flos": 33292495388160.0,
      "grad_norm": 2.8865628899563283,
      "language_loss": 0.66459376,
      "learning_rate": 8.472574827008428e-08,
      "loss": 0.68630922,
      "num_input_tokens_seen": 163600425,
      "step": 7566,
      "time_per_iteration": 2.5603184700012207
    },
    {
      "auxiliary_loss_clip": 0.01149132,
      "auxiliary_loss_mlp": 0.01023433,
      "balance_loss_clip": 1.04352951,
      "balance_loss_mlp": 1.01647758,
      "epoch": 0.9098779534660013,
      "flos": 21906443001600.0,
      "grad_norm": 1.6950286568803576,
      "language_loss": 0.84148777,
      "learning_rate": 8.450156711295942e-08,
      "loss": 0.86321342,
      "num_input_tokens_seen": 163620595,
      "step": 7567,
      "time_per_iteration": 2.449101448059082
    },
    {
      "auxiliary_loss_clip": 0.01137527,
      "auxiliary_loss_mlp": 0.01025478,
      "balance_loss_clip": 1.04732513,
      "balance_loss_mlp": 1.01843262,
      "epoch": 0.9099981963566404,
      "flos": 25730354102400.0,
      "grad_norm": 1.9698055534687635,
      "language_loss": 0.86512679,
      "learning_rate": 8.427767653542383e-08,
      "loss": 0.88675684,
      "num_input_tokens_seen": 163635765,
      "step": 7568,
      "time_per_iteration": 2.5017755031585693
    },
    {
      "auxiliary_loss_clip": 0.01102267,
      "auxiliary_loss_mlp": 0.01025372,
      "balance_loss_clip": 1.03823709,
      "balance_loss_mlp": 1.0188688,
      "epoch": 0.9101184392472795,
      "flos": 21069437304960.0,
      "grad_norm": 2.4978980834833613,
      "language_loss": 0.70341295,
      "learning_rate": 8.405407657144125e-08,
      "loss": 0.72468925,
      "num_input_tokens_seen": 163654925,
      "step": 7569,
      "time_per_iteration": 2.5462863445281982
    },
    {
      "auxiliary_loss_clip": 0.01132208,
      "auxiliary_loss_mlp": 0.01023143,
      "balance_loss_clip": 1.04384458,
      "balance_loss_mlp": 1.016011,
      "epoch": 0.9102386821379186,
      "flos": 24752614919040.0,
      "grad_norm": 1.688640509720927,
      "language_loss": 0.72256434,
      "learning_rate": 8.383076725493232e-08,
      "loss": 0.74411786,
      "num_input_tokens_seen": 163672245,
      "step": 7570,
      "time_per_iteration": 3.277507781982422
    },
    {
      "auxiliary_loss_clip": 0.01152124,
      "auxiliary_loss_mlp": 0.01019254,
      "balance_loss_clip": 1.04558885,
      "balance_loss_mlp": 1.0124532,
      "epoch": 0.9103589250285576,
      "flos": 22562818179840.0,
      "grad_norm": 1.8136145324130062,
      "language_loss": 0.67824489,
      "learning_rate": 8.360774861977216e-08,
      "loss": 0.69995862,
      "num_input_tokens_seen": 163691365,
      "step": 7571,
      "time_per_iteration": 2.4458084106445312
    },
    {
      "auxiliary_loss_clip": 0.01135282,
      "auxiliary_loss_mlp": 0.01020508,
      "balance_loss_clip": 1.04011559,
      "balance_loss_mlp": 1.01348901,
      "epoch": 0.9104791679191968,
      "flos": 25373474524800.0,
      "grad_norm": 2.718233018534829,
      "language_loss": 0.74644589,
      "learning_rate": 8.338502069979281e-08,
      "loss": 0.76800382,
      "num_input_tokens_seen": 163711675,
      "step": 7572,
      "time_per_iteration": 2.529071807861328
    },
    {
      "auxiliary_loss_clip": 0.01153382,
      "auxiliary_loss_mlp": 0.01024591,
      "balance_loss_clip": 1.04505289,
      "balance_loss_mlp": 1.01723576,
      "epoch": 0.9105994108098359,
      "flos": 14426681558400.0,
      "grad_norm": 4.686721507726226,
      "language_loss": 0.80009401,
      "learning_rate": 8.316258352878214e-08,
      "loss": 0.82187378,
      "num_input_tokens_seen": 163728095,
      "step": 7573,
      "time_per_iteration": 2.409842014312744
    },
    {
      "auxiliary_loss_clip": 0.01155798,
      "auxiliary_loss_mlp": 0.0102672,
      "balance_loss_clip": 1.04461288,
      "balance_loss_mlp": 1.01912093,
      "epoch": 0.9107196537004749,
      "flos": 26718292748160.0,
      "grad_norm": 1.8767521077486224,
      "language_loss": 0.71043646,
      "learning_rate": 8.294043714048338e-08,
      "loss": 0.73226166,
      "num_input_tokens_seen": 163747175,
      "step": 7574,
      "time_per_iteration": 2.5055599212646484
    },
    {
      "auxiliary_loss_clip": 0.01041189,
      "auxiliary_loss_mlp": 0.01000876,
      "balance_loss_clip": 1.00678039,
      "balance_loss_mlp": 0.99991018,
      "epoch": 0.9108398965911141,
      "flos": 66532634703360.0,
      "grad_norm": 0.7496199689679823,
      "language_loss": 0.60471404,
      "learning_rate": 8.271858156859624e-08,
      "loss": 0.62513471,
      "num_input_tokens_seen": 163812545,
      "step": 7575,
      "time_per_iteration": 3.124560594558716
    },
    {
      "auxiliary_loss_clip": 0.01162867,
      "auxiliary_loss_mlp": 0.01018015,
      "balance_loss_clip": 1.04664505,
      "balance_loss_mlp": 1.010988,
      "epoch": 0.9109601394817531,
      "flos": 25411073086080.0,
      "grad_norm": 1.591850634327226,
      "language_loss": 0.73872173,
      "learning_rate": 8.249701684677557e-08,
      "loss": 0.76053047,
      "num_input_tokens_seen": 163833870,
      "step": 7576,
      "time_per_iteration": 2.4611191749572754
    },
    {
      "auxiliary_loss_clip": 0.01151638,
      "auxiliary_loss_mlp": 0.0102314,
      "balance_loss_clip": 1.04795575,
      "balance_loss_mlp": 1.01612711,
      "epoch": 0.9110803823723922,
      "flos": 22747794243840.0,
      "grad_norm": 1.786722172825009,
      "language_loss": 0.81441593,
      "learning_rate": 8.227574300863294e-08,
      "loss": 0.83616364,
      "num_input_tokens_seen": 163854040,
      "step": 7577,
      "time_per_iteration": 2.471400260925293
    },
    {
      "auxiliary_loss_clip": 0.01141203,
      "auxiliary_loss_mlp": 0.01021569,
      "balance_loss_clip": 1.04657054,
      "balance_loss_mlp": 1.01424098,
      "epoch": 0.9112006252630314,
      "flos": 48469924131840.0,
      "grad_norm": 1.819111450387425,
      "language_loss": 0.69685501,
      "learning_rate": 8.205476008773548e-08,
      "loss": 0.71848273,
      "num_input_tokens_seen": 163878040,
      "step": 7578,
      "time_per_iteration": 2.7352585792541504
    },
    {
      "auxiliary_loss_clip": 0.01115899,
      "auxiliary_loss_mlp": 0.01026756,
      "balance_loss_clip": 1.04291379,
      "balance_loss_mlp": 1.01962113,
      "epoch": 0.9113208681536704,
      "flos": 30009649829760.0,
      "grad_norm": 2.2898339699230763,
      "language_loss": 0.82572925,
      "learning_rate": 8.183406811760596e-08,
      "loss": 0.84715581,
      "num_input_tokens_seen": 163897770,
      "step": 7579,
      "time_per_iteration": 2.6108853816986084
    },
    {
      "auxiliary_loss_clip": 0.01111025,
      "auxiliary_loss_mlp": 0.01019578,
      "balance_loss_clip": 1.04013908,
      "balance_loss_mlp": 1.01264334,
      "epoch": 0.9114411110443095,
      "flos": 25594971742080.0,
      "grad_norm": 1.9316176250182733,
      "language_loss": 0.74161077,
      "learning_rate": 8.161366713172313e-08,
      "loss": 0.76291674,
      "num_input_tokens_seen": 163920160,
      "step": 7580,
      "time_per_iteration": 2.600269079208374
    },
    {
      "auxiliary_loss_clip": 0.01128745,
      "auxiliary_loss_mlp": 0.01025573,
      "balance_loss_clip": 1.04281259,
      "balance_loss_mlp": 1.01824141,
      "epoch": 0.9115613539349486,
      "flos": 18399729928320.0,
      "grad_norm": 2.692894319953287,
      "language_loss": 0.84322506,
      "learning_rate": 8.139355716352137e-08,
      "loss": 0.86476827,
      "num_input_tokens_seen": 163935000,
      "step": 7581,
      "time_per_iteration": 2.497633695602417
    },
    {
      "auxiliary_loss_clip": 0.01139455,
      "auxiliary_loss_mlp": 0.01026491,
      "balance_loss_clip": 1.04306817,
      "balance_loss_mlp": 1.01909161,
      "epoch": 0.9116815968255877,
      "flos": 21726171619200.0,
      "grad_norm": 2.2415526055878705,
      "language_loss": 0.70173478,
      "learning_rate": 8.117373824639196e-08,
      "loss": 0.72339422,
      "num_input_tokens_seen": 163955265,
      "step": 7582,
      "time_per_iteration": 2.486483573913574
    },
    {
      "auxiliary_loss_clip": 0.01060359,
      "auxiliary_loss_mlp": 0.01000657,
      "balance_loss_clip": 1.00708246,
      "balance_loss_mlp": 0.99976271,
      "epoch": 0.9118018397162267,
      "flos": 65363526835200.0,
      "grad_norm": 0.7236309186387634,
      "language_loss": 0.59293932,
      "learning_rate": 8.095421041368067e-08,
      "loss": 0.61354953,
      "num_input_tokens_seen": 164014680,
      "step": 7583,
      "time_per_iteration": 2.9243624210357666
    },
    {
      "auxiliary_loss_clip": 0.01134847,
      "auxiliary_loss_mlp": 0.00761967,
      "balance_loss_clip": 1.04453897,
      "balance_loss_mlp": 1.00047135,
      "epoch": 0.9119220826068659,
      "flos": 20922885815040.0,
      "grad_norm": 2.65117118910255,
      "language_loss": 0.70748264,
      "learning_rate": 8.073497369868999e-08,
      "loss": 0.72645074,
      "num_input_tokens_seen": 164033140,
      "step": 7584,
      "time_per_iteration": 2.4762933254241943
    },
    {
      "auxiliary_loss_clip": 0.0114606,
      "auxiliary_loss_mlp": 0.01023689,
      "balance_loss_clip": 1.04572368,
      "balance_loss_mlp": 1.0164355,
      "epoch": 0.912042325497505,
      "flos": 28366449327360.0,
      "grad_norm": 2.0049272295103755,
      "language_loss": 0.75278604,
      "learning_rate": 8.051602813467772e-08,
      "loss": 0.77448344,
      "num_input_tokens_seen": 164054995,
      "step": 7585,
      "time_per_iteration": 2.5583436489105225
    },
    {
      "auxiliary_loss_clip": 0.01155188,
      "auxiliary_loss_mlp": 0.01022027,
      "balance_loss_clip": 1.04712009,
      "balance_loss_mlp": 1.01530313,
      "epoch": 0.912162568388144,
      "flos": 17566782468480.0,
      "grad_norm": 5.615056663456039,
      "language_loss": 0.71047747,
      "learning_rate": 8.029737375485756e-08,
      "loss": 0.73224968,
      "num_input_tokens_seen": 164074225,
      "step": 7586,
      "time_per_iteration": 2.4545328617095947
    },
    {
      "auxiliary_loss_clip": 0.01165329,
      "auxiliary_loss_mlp": 0.01022721,
      "balance_loss_clip": 1.047382,
      "balance_loss_mlp": 1.01607215,
      "epoch": 0.9122828112787832,
      "flos": 19827897661440.0,
      "grad_norm": 1.737784491086537,
      "language_loss": 0.72438973,
      "learning_rate": 8.007901059239986e-08,
      "loss": 0.7462703,
      "num_input_tokens_seen": 164093505,
      "step": 7587,
      "time_per_iteration": 3.2481539249420166
    },
    {
      "auxiliary_loss_clip": 0.01136636,
      "auxiliary_loss_mlp": 0.01020159,
      "balance_loss_clip": 1.04129672,
      "balance_loss_mlp": 1.01310825,
      "epoch": 0.9124030541694222,
      "flos": 20813789232000.0,
      "grad_norm": 1.70199400515549,
      "language_loss": 0.80106831,
      "learning_rate": 7.986093868042964e-08,
      "loss": 0.82263625,
      "num_input_tokens_seen": 164113750,
      "step": 7588,
      "time_per_iteration": 2.4724841117858887
    },
    {
      "auxiliary_loss_clip": 0.0114988,
      "auxiliary_loss_mlp": 0.01028608,
      "balance_loss_clip": 1.04561639,
      "balance_loss_mlp": 1.02177155,
      "epoch": 0.9125232970600613,
      "flos": 25192305302400.0,
      "grad_norm": 2.0762520832926286,
      "language_loss": 0.67812824,
      "learning_rate": 7.964315805202826e-08,
      "loss": 0.69991308,
      "num_input_tokens_seen": 164134330,
      "step": 7589,
      "time_per_iteration": 3.219475269317627
    },
    {
      "auxiliary_loss_clip": 0.01136878,
      "auxiliary_loss_mlp": 0.01022026,
      "balance_loss_clip": 1.04485464,
      "balance_loss_mlp": 1.01427102,
      "epoch": 0.9126435399507005,
      "flos": 19719591177600.0,
      "grad_norm": 1.867345436542811,
      "language_loss": 0.72886568,
      "learning_rate": 7.942566874023304e-08,
      "loss": 0.75045466,
      "num_input_tokens_seen": 164153515,
      "step": 7590,
      "time_per_iteration": 3.209393262863159
    },
    {
      "auxiliary_loss_clip": 0.01132917,
      "auxiliary_loss_mlp": 0.01025876,
      "balance_loss_clip": 1.0413537,
      "balance_loss_mlp": 1.01860976,
      "epoch": 0.9127637828413395,
      "flos": 19573614305280.0,
      "grad_norm": 2.0016724650280557,
      "language_loss": 0.69446194,
      "learning_rate": 7.920847077803649e-08,
      "loss": 0.71604991,
      "num_input_tokens_seen": 164171305,
      "step": 7591,
      "time_per_iteration": 2.452043294906616
    },
    {
      "auxiliary_loss_clip": 0.01097696,
      "auxiliary_loss_mlp": 0.01024446,
      "balance_loss_clip": 1.03641629,
      "balance_loss_mlp": 1.01745701,
      "epoch": 0.9128840257319786,
      "flos": 20230635928320.0,
      "grad_norm": 2.003938756046427,
      "language_loss": 0.82037741,
      "learning_rate": 7.899156419838826e-08,
      "loss": 0.84159881,
      "num_input_tokens_seen": 164190275,
      "step": 7592,
      "time_per_iteration": 2.545992612838745
    },
    {
      "auxiliary_loss_clip": 0.01119745,
      "auxiliary_loss_mlp": 0.01023781,
      "balance_loss_clip": 1.04175746,
      "balance_loss_mlp": 1.01712632,
      "epoch": 0.9130042686226177,
      "flos": 24858658846080.0,
      "grad_norm": 1.837050280725676,
      "language_loss": 0.65535933,
      "learning_rate": 7.87749490341918e-08,
      "loss": 0.67679459,
      "num_input_tokens_seen": 164210550,
      "step": 7593,
      "time_per_iteration": 2.547178268432617
    },
    {
      "auxiliary_loss_clip": 0.01168361,
      "auxiliary_loss_mlp": 0.01024578,
      "balance_loss_clip": 1.0488019,
      "balance_loss_mlp": 1.01726437,
      "epoch": 0.9131245115132568,
      "flos": 23581747284480.0,
      "grad_norm": 2.1144910283486245,
      "language_loss": 0.8319335,
      "learning_rate": 7.855862531830836e-08,
      "loss": 0.85386288,
      "num_input_tokens_seen": 164226660,
      "step": 7594,
      "time_per_iteration": 2.405717134475708
    },
    {
      "auxiliary_loss_clip": 0.01148084,
      "auxiliary_loss_mlp": 0.01021432,
      "balance_loss_clip": 1.04354155,
      "balance_loss_mlp": 1.01450872,
      "epoch": 0.9132447544038959,
      "flos": 19931607204480.0,
      "grad_norm": 1.5790664591105037,
      "language_loss": 0.72917855,
      "learning_rate": 7.834259308355373e-08,
      "loss": 0.75087368,
      "num_input_tokens_seen": 164245425,
      "step": 7595,
      "time_per_iteration": 2.4460058212280273
    },
    {
      "auxiliary_loss_clip": 0.01078176,
      "auxiliary_loss_mlp": 0.01024884,
      "balance_loss_clip": 1.03630245,
      "balance_loss_mlp": 1.01792192,
      "epoch": 0.9133649972945349,
      "flos": 21981747864960.0,
      "grad_norm": 2.0350844150389253,
      "language_loss": 0.75049615,
      "learning_rate": 7.812685236269989e-08,
      "loss": 0.77152681,
      "num_input_tokens_seen": 164264085,
      "step": 7596,
      "time_per_iteration": 2.6115055084228516
    },
    {
      "auxiliary_loss_clip": 0.01026284,
      "auxiliary_loss_mlp": 0.01002656,
      "balance_loss_clip": 1.00898743,
      "balance_loss_mlp": 1.00182152,
      "epoch": 0.9134852401851741,
      "flos": 71240523511680.0,
      "grad_norm": 0.7881695381492786,
      "language_loss": 0.58615732,
      "learning_rate": 7.791140318847445e-08,
      "loss": 0.60644674,
      "num_input_tokens_seen": 164322220,
      "step": 7597,
      "time_per_iteration": 3.8404805660247803
    },
    {
      "auxiliary_loss_clip": 0.01134291,
      "auxiliary_loss_mlp": 0.01019556,
      "balance_loss_clip": 1.04687822,
      "balance_loss_mlp": 1.01305056,
      "epoch": 0.9136054830758131,
      "flos": 23626923615360.0,
      "grad_norm": 1.4800515167768031,
      "language_loss": 0.80263603,
      "learning_rate": 7.769624559356081e-08,
      "loss": 0.82417446,
      "num_input_tokens_seen": 164345615,
      "step": 7598,
      "time_per_iteration": 2.5635030269622803
    },
    {
      "auxiliary_loss_clip": 0.01149581,
      "auxiliary_loss_mlp": 0.01024296,
      "balance_loss_clip": 1.04385662,
      "balance_loss_mlp": 1.01632106,
      "epoch": 0.9137257259664522,
      "flos": 23438858981760.0,
      "grad_norm": 3.999870129167285,
      "language_loss": 0.75410414,
      "learning_rate": 7.748137961059842e-08,
      "loss": 0.77584291,
      "num_input_tokens_seen": 164359595,
      "step": 7599,
      "time_per_iteration": 2.4458670616149902
    },
    {
      "auxiliary_loss_clip": 0.0116198,
      "auxiliary_loss_mlp": 0.01024343,
      "balance_loss_clip": 1.04683089,
      "balance_loss_mlp": 1.01709819,
      "epoch": 0.9138459688570914,
      "flos": 19127854523520.0,
      "grad_norm": 2.5979285564238825,
      "language_loss": 0.65739489,
      "learning_rate": 7.726680527218211e-08,
      "loss": 0.67925811,
      "num_input_tokens_seen": 164376635,
      "step": 7600,
      "time_per_iteration": 2.3873424530029297
    },
    {
      "auxiliary_loss_clip": 0.01164609,
      "auxiliary_loss_mlp": 0.01022347,
      "balance_loss_clip": 1.04517794,
      "balance_loss_mlp": 1.01536131,
      "epoch": 0.9139662117477304,
      "flos": 46281240714240.0,
      "grad_norm": 1.6979897915311042,
      "language_loss": 0.75865489,
      "learning_rate": 7.70525226108627e-08,
      "loss": 0.78052443,
      "num_input_tokens_seen": 164400305,
      "step": 7601,
      "time_per_iteration": 2.619725465774536
    },
    {
      "auxiliary_loss_clip": 0.01152321,
      "auxiliary_loss_mlp": 0.01026129,
      "balance_loss_clip": 1.04807508,
      "balance_loss_mlp": 1.01919091,
      "epoch": 0.9140864546383695,
      "flos": 22273198819200.0,
      "grad_norm": 2.168382931186185,
      "language_loss": 0.79799485,
      "learning_rate": 7.683853165914666e-08,
      "loss": 0.81977934,
      "num_input_tokens_seen": 164418075,
      "step": 7602,
      "time_per_iteration": 2.4384236335754395
    },
    {
      "auxiliary_loss_clip": 0.01110692,
      "auxiliary_loss_mlp": 0.01025256,
      "balance_loss_clip": 1.04312313,
      "balance_loss_mlp": 1.01855612,
      "epoch": 0.9142066975290086,
      "flos": 17530009920000.0,
      "grad_norm": 1.9916114291657985,
      "language_loss": 0.77349365,
      "learning_rate": 7.662483244949602e-08,
      "loss": 0.79485315,
      "num_input_tokens_seen": 164435335,
      "step": 7603,
      "time_per_iteration": 2.5221846103668213
    },
    {
      "auxiliary_loss_clip": 0.01115678,
      "auxiliary_loss_mlp": 0.01019379,
      "balance_loss_clip": 1.04257107,
      "balance_loss_mlp": 1.01239026,
      "epoch": 0.9143269404196477,
      "flos": 17712148809600.0,
      "grad_norm": 2.980415360896527,
      "language_loss": 0.80810434,
      "learning_rate": 7.641142501432951e-08,
      "loss": 0.8294549,
      "num_input_tokens_seen": 164451530,
      "step": 7604,
      "time_per_iteration": 2.4473040103912354
    },
    {
      "auxiliary_loss_clip": 0.01131348,
      "auxiliary_loss_mlp": 0.01023347,
      "balance_loss_clip": 1.04267859,
      "balance_loss_mlp": 1.01647472,
      "epoch": 0.9144471833102867,
      "flos": 33323414019840.0,
      "grad_norm": 1.8313251872224725,
      "language_loss": 0.73798144,
      "learning_rate": 7.619830938602013e-08,
      "loss": 0.7595284,
      "num_input_tokens_seen": 164472755,
      "step": 7605,
      "time_per_iteration": 2.581111431121826
    },
    {
      "auxiliary_loss_clip": 0.01145574,
      "auxiliary_loss_mlp": 0.01022238,
      "balance_loss_clip": 1.04453278,
      "balance_loss_mlp": 1.01508856,
      "epoch": 0.9145674262009259,
      "flos": 21068970428160.0,
      "grad_norm": 2.413515252292556,
      "language_loss": 0.82065606,
      "learning_rate": 7.598548559689777e-08,
      "loss": 0.84233415,
      "num_input_tokens_seen": 164491155,
      "step": 7606,
      "time_per_iteration": 2.441040277481079
    },
    {
      "auxiliary_loss_clip": 0.01117324,
      "auxiliary_loss_mlp": 0.01022258,
      "balance_loss_clip": 1.04175425,
      "balance_loss_mlp": 1.01544833,
      "epoch": 0.914687669091565,
      "flos": 16800269212800.0,
      "grad_norm": 6.175178509466248,
      "language_loss": 0.80802983,
      "learning_rate": 7.577295367924751e-08,
      "loss": 0.82942563,
      "num_input_tokens_seen": 164507555,
      "step": 7607,
      "time_per_iteration": 2.474886655807495
    },
    {
      "auxiliary_loss_clip": 0.01141552,
      "auxiliary_loss_mlp": 0.0102126,
      "balance_loss_clip": 1.04582965,
      "balance_loss_mlp": 1.0139823,
      "epoch": 0.914807911982204,
      "flos": 25773627012480.0,
      "grad_norm": 1.7403130139530418,
      "language_loss": 0.82291836,
      "learning_rate": 7.556071366531002e-08,
      "loss": 0.8445465,
      "num_input_tokens_seen": 164528525,
      "step": 7608,
      "time_per_iteration": 2.517787218093872
    },
    {
      "auxiliary_loss_clip": 0.01150925,
      "auxiliary_loss_mlp": 0.01024437,
      "balance_loss_clip": 1.04643726,
      "balance_loss_mlp": 1.01704049,
      "epoch": 0.9149281548728432,
      "flos": 19208043636480.0,
      "grad_norm": 4.017934441258296,
      "language_loss": 0.79343176,
      "learning_rate": 7.53487655872822e-08,
      "loss": 0.81518531,
      "num_input_tokens_seen": 164547695,
      "step": 7609,
      "time_per_iteration": 2.4374399185180664
    },
    {
      "auxiliary_loss_clip": 0.01110623,
      "auxiliary_loss_mlp": 0.01022692,
      "balance_loss_clip": 1.03855669,
      "balance_loss_mlp": 1.01514578,
      "epoch": 0.9150483977634822,
      "flos": 26870554500480.0,
      "grad_norm": 1.8176322488121996,
      "language_loss": 0.73922813,
      "learning_rate": 7.513710947731656e-08,
      "loss": 0.76056129,
      "num_input_tokens_seen": 164568905,
      "step": 7610,
      "time_per_iteration": 2.56787109375
    },
    {
      "auxiliary_loss_clip": 0.01129543,
      "auxiliary_loss_mlp": 0.01024964,
      "balance_loss_clip": 1.04334033,
      "balance_loss_mlp": 1.01781785,
      "epoch": 0.9151686406541213,
      "flos": 21908956953600.0,
      "grad_norm": 1.950405755254037,
      "language_loss": 0.85082835,
      "learning_rate": 7.492574536752095e-08,
      "loss": 0.87237346,
      "num_input_tokens_seen": 164588895,
      "step": 7611,
      "time_per_iteration": 2.4774608612060547
    },
    {
      "auxiliary_loss_clip": 0.01136622,
      "auxiliary_loss_mlp": 0.0102584,
      "balance_loss_clip": 1.04433525,
      "balance_loss_mlp": 1.01925111,
      "epoch": 0.9152888835447605,
      "flos": 27308556944640.0,
      "grad_norm": 1.9405805394408722,
      "language_loss": 0.78302896,
      "learning_rate": 7.471467328995907e-08,
      "loss": 0.80465358,
      "num_input_tokens_seen": 164607705,
      "step": 7612,
      "time_per_iteration": 2.5159993171691895
    },
    {
      "auxiliary_loss_clip": 0.0108466,
      "auxiliary_loss_mlp": 0.0102145,
      "balance_loss_clip": 1.0383029,
      "balance_loss_mlp": 1.01409447,
      "epoch": 0.9154091264353995,
      "flos": 13370728510080.0,
      "grad_norm": 2.288525583068321,
      "language_loss": 0.6045481,
      "learning_rate": 7.450389327665018e-08,
      "loss": 0.62560916,
      "num_input_tokens_seen": 164625540,
      "step": 7613,
      "time_per_iteration": 2.595487117767334
    },
    {
      "auxiliary_loss_clip": 0.01127667,
      "auxiliary_loss_mlp": 0.01027012,
      "balance_loss_clip": 1.0492301,
      "balance_loss_mlp": 1.01963925,
      "epoch": 0.9155293693260386,
      "flos": 20193037367040.0,
      "grad_norm": 3.6124289286554423,
      "language_loss": 0.67764872,
      "learning_rate": 7.429340535957029e-08,
      "loss": 0.6991955,
      "num_input_tokens_seen": 164640735,
      "step": 7614,
      "time_per_iteration": 3.337618589401245
    },
    {
      "auxiliary_loss_clip": 0.01136766,
      "auxiliary_loss_mlp": 0.01025526,
      "balance_loss_clip": 1.04304767,
      "balance_loss_mlp": 1.01893091,
      "epoch": 0.9156496122166777,
      "flos": 19354990176000.0,
      "grad_norm": 2.188030475064479,
      "language_loss": 0.70678705,
      "learning_rate": 7.40832095706494e-08,
      "loss": 0.72841001,
      "num_input_tokens_seen": 164657430,
      "step": 7615,
      "time_per_iteration": 3.2225139141082764
    },
    {
      "auxiliary_loss_clip": 0.01129649,
      "auxiliary_loss_mlp": 0.01028783,
      "balance_loss_clip": 1.04544044,
      "balance_loss_mlp": 1.02209473,
      "epoch": 0.9157698551073168,
      "flos": 21107287261440.0,
      "grad_norm": 1.8147881568414153,
      "language_loss": 0.79950982,
      "learning_rate": 7.387330594177443e-08,
      "loss": 0.82109416,
      "num_input_tokens_seen": 164679505,
      "step": 7616,
      "time_per_iteration": 3.3735921382904053
    },
    {
      "auxiliary_loss_clip": 0.01118296,
      "auxiliary_loss_mlp": 0.01027594,
      "balance_loss_clip": 1.04283071,
      "balance_loss_mlp": 1.02089691,
      "epoch": 0.9158900979979558,
      "flos": 25193167228800.0,
      "grad_norm": 1.6284707456835912,
      "language_loss": 0.79017949,
      "learning_rate": 7.366369450478749e-08,
      "loss": 0.81163836,
      "num_input_tokens_seen": 164700615,
      "step": 7617,
      "time_per_iteration": 2.54510235786438
    },
    {
      "auxiliary_loss_clip": 0.01118042,
      "auxiliary_loss_mlp": 0.01026774,
      "balance_loss_clip": 1.04187059,
      "balance_loss_mlp": 1.01980066,
      "epoch": 0.916010340888595,
      "flos": 30146648302080.0,
      "grad_norm": 1.5856731105762434,
      "language_loss": 0.66581762,
      "learning_rate": 7.345437529148646e-08,
      "loss": 0.68726575,
      "num_input_tokens_seen": 164719625,
      "step": 7618,
      "time_per_iteration": 2.5736329555511475
    },
    {
      "auxiliary_loss_clip": 0.01121513,
      "auxiliary_loss_mlp": 0.01030444,
      "balance_loss_clip": 1.04237258,
      "balance_loss_mlp": 1.02339554,
      "epoch": 0.9161305837792341,
      "flos": 17091827907840.0,
      "grad_norm": 1.8377405237655395,
      "language_loss": 0.72814631,
      "learning_rate": 7.324534833362483e-08,
      "loss": 0.74966586,
      "num_input_tokens_seen": 164737200,
      "step": 7619,
      "time_per_iteration": 2.4841952323913574
    },
    {
      "auxiliary_loss_clip": 0.01137303,
      "auxiliary_loss_mlp": 0.01023393,
      "balance_loss_clip": 1.04526663,
      "balance_loss_mlp": 1.01624596,
      "epoch": 0.9162508266698731,
      "flos": 22893699288960.0,
      "grad_norm": 1.78582716112665,
      "language_loss": 0.68454885,
      "learning_rate": 7.303661366291192e-08,
      "loss": 0.70615578,
      "num_input_tokens_seen": 164757870,
      "step": 7620,
      "time_per_iteration": 2.5334620475769043
    },
    {
      "auxiliary_loss_clip": 0.01107017,
      "auxiliary_loss_mlp": 0.01022593,
      "balance_loss_clip": 1.04087758,
      "balance_loss_mlp": 1.0158217,
      "epoch": 0.9163710695605123,
      "flos": 19974808287360.0,
      "grad_norm": 1.7482009389320294,
      "language_loss": 0.81431764,
      "learning_rate": 7.28281713110126e-08,
      "loss": 0.83561373,
      "num_input_tokens_seen": 164775945,
      "step": 7621,
      "time_per_iteration": 2.546293258666992
    },
    {
      "auxiliary_loss_clip": 0.01134049,
      "auxiliary_loss_mlp": 0.01024958,
      "balance_loss_clip": 1.0453943,
      "balance_loss_mlp": 1.0179249,
      "epoch": 0.9164913124511513,
      "flos": 22783812606720.0,
      "grad_norm": 1.9266749360005215,
      "language_loss": 0.77157038,
      "learning_rate": 7.262002130954759e-08,
      "loss": 0.79316044,
      "num_input_tokens_seen": 164794400,
      "step": 7622,
      "time_per_iteration": 2.507571220397949
    },
    {
      "auxiliary_loss_clip": 0.01113115,
      "auxiliary_loss_mlp": 0.01027043,
      "balance_loss_clip": 1.04219127,
      "balance_loss_mlp": 1.01950264,
      "epoch": 0.9166115553417904,
      "flos": 24900854348160.0,
      "grad_norm": 1.6920187560768574,
      "language_loss": 0.78658438,
      "learning_rate": 7.241216369009296e-08,
      "loss": 0.80798596,
      "num_input_tokens_seen": 164814585,
      "step": 7623,
      "time_per_iteration": 3.3467843532562256
    },
    {
      "auxiliary_loss_clip": 0.01162854,
      "auxiliary_loss_mlp": 0.010208,
      "balance_loss_clip": 1.0452739,
      "balance_loss_mlp": 1.01393938,
      "epoch": 0.9167317982324296,
      "flos": 25702919089920.0,
      "grad_norm": 2.8370637099199048,
      "language_loss": 0.66368234,
      "learning_rate": 7.220459848418037e-08,
      "loss": 0.68551886,
      "num_input_tokens_seen": 164834660,
      "step": 7624,
      "time_per_iteration": 2.4610095024108887
    },
    {
      "auxiliary_loss_clip": 0.0116212,
      "auxiliary_loss_mlp": 0.01021668,
      "balance_loss_clip": 1.04705691,
      "balance_loss_mlp": 1.01495612,
      "epoch": 0.9168520411230686,
      "flos": 15632813370240.0,
      "grad_norm": 1.6681159869818696,
      "language_loss": 0.79513001,
      "learning_rate": 7.199732572329708e-08,
      "loss": 0.8169679,
      "num_input_tokens_seen": 164852560,
      "step": 7625,
      "time_per_iteration": 2.41252064704895
    },
    {
      "auxiliary_loss_clip": 0.01126649,
      "auxiliary_loss_mlp": 0.01027555,
      "balance_loss_clip": 1.04299486,
      "balance_loss_mlp": 1.02026534,
      "epoch": 0.9169722840137077,
      "flos": 30258151096320.0,
      "grad_norm": 2.2095812584423653,
      "language_loss": 0.75729418,
      "learning_rate": 7.179034543888684e-08,
      "loss": 0.77883625,
      "num_input_tokens_seen": 164872065,
      "step": 7626,
      "time_per_iteration": 2.5752599239349365
    },
    {
      "auxiliary_loss_clip": 0.011522,
      "auxiliary_loss_mlp": 0.01025421,
      "balance_loss_clip": 1.04423654,
      "balance_loss_mlp": 1.01856041,
      "epoch": 0.9170925269043467,
      "flos": 22491643380480.0,
      "grad_norm": 2.582222675900588,
      "language_loss": 0.77464998,
      "learning_rate": 7.158365766234808e-08,
      "loss": 0.79642618,
      "num_input_tokens_seen": 164890915,
      "step": 7627,
      "time_per_iteration": 2.4673123359680176
    },
    {
      "auxiliary_loss_clip": 0.01113003,
      "auxiliary_loss_mlp": 0.0102329,
      "balance_loss_clip": 1.03824651,
      "balance_loss_mlp": 1.0156486,
      "epoch": 0.9172127697949859,
      "flos": 22893914770560.0,
      "grad_norm": 1.9622509135071753,
      "language_loss": 0.72362083,
      "learning_rate": 7.137726242503527e-08,
      "loss": 0.74498373,
      "num_input_tokens_seen": 164909835,
      "step": 7628,
      "time_per_iteration": 2.513026714324951
    },
    {
      "auxiliary_loss_clip": 0.01149513,
      "auxiliary_loss_mlp": 0.00762049,
      "balance_loss_clip": 1.04580176,
      "balance_loss_mlp": 1.00047135,
      "epoch": 0.917333012685625,
      "flos": 17451867882240.0,
      "grad_norm": 2.3189796566703813,
      "language_loss": 0.77865136,
      "learning_rate": 7.11711597582585e-08,
      "loss": 0.79776692,
      "num_input_tokens_seen": 164927195,
      "step": 7629,
      "time_per_iteration": 2.4811604022979736
    },
    {
      "auxiliary_loss_clip": 0.01121428,
      "auxiliary_loss_mlp": 0.01020991,
      "balance_loss_clip": 1.03957272,
      "balance_loss_mlp": 1.01472926,
      "epoch": 0.917453255576264,
      "flos": 14318949692160.0,
      "grad_norm": 1.819354923664696,
      "language_loss": 0.79972935,
      "learning_rate": 7.096534969328271e-08,
      "loss": 0.82115358,
      "num_input_tokens_seen": 164944640,
      "step": 7630,
      "time_per_iteration": 2.476349115371704
    },
    {
      "auxiliary_loss_clip": 0.01140394,
      "auxiliary_loss_mlp": 0.01021659,
      "balance_loss_clip": 1.04276109,
      "balance_loss_mlp": 1.01481354,
      "epoch": 0.9175734984669032,
      "flos": 20741177888640.0,
      "grad_norm": 2.535046132391288,
      "language_loss": 0.83909428,
      "learning_rate": 7.075983226132987e-08,
      "loss": 0.86071479,
      "num_input_tokens_seen": 164963570,
      "step": 7631,
      "time_per_iteration": 2.4864015579223633
    },
    {
      "auxiliary_loss_clip": 0.01139387,
      "auxiliary_loss_mlp": 0.0076235,
      "balance_loss_clip": 1.04189718,
      "balance_loss_mlp": 1.00040793,
      "epoch": 0.9176937413575422,
      "flos": 14830497233280.0,
      "grad_norm": 2.752170995558212,
      "language_loss": 0.7917909,
      "learning_rate": 7.055460749357656e-08,
      "loss": 0.8108083,
      "num_input_tokens_seen": 164979850,
      "step": 7632,
      "time_per_iteration": 2.43985915184021
    },
    {
      "auxiliary_loss_clip": 0.01137293,
      "auxiliary_loss_mlp": 0.01026468,
      "balance_loss_clip": 1.04539037,
      "balance_loss_mlp": 1.0190413,
      "epoch": 0.9178139842481813,
      "flos": 18474603828480.0,
      "grad_norm": 1.6268610325364654,
      "language_loss": 0.70351678,
      "learning_rate": 7.034967542115521e-08,
      "loss": 0.7251544,
      "num_input_tokens_seen": 164998115,
      "step": 7633,
      "time_per_iteration": 2.4819447994232178
    },
    {
      "auxiliary_loss_clip": 0.01140377,
      "auxiliary_loss_mlp": 0.00761638,
      "balance_loss_clip": 1.04300833,
      "balance_loss_mlp": 1.00050187,
      "epoch": 0.9179342271388204,
      "flos": 20047455544320.0,
      "grad_norm": 2.1005244022234226,
      "language_loss": 0.75454772,
      "learning_rate": 7.014503607515388e-08,
      "loss": 0.77356791,
      "num_input_tokens_seen": 165017420,
      "step": 7634,
      "time_per_iteration": 2.4153895378112793
    },
    {
      "auxiliary_loss_clip": 0.01137467,
      "auxiliary_loss_mlp": 0.01028589,
      "balance_loss_clip": 1.04749775,
      "balance_loss_mlp": 1.02157021,
      "epoch": 0.9180544700294595,
      "flos": 24676232647680.0,
      "grad_norm": 2.333512025323974,
      "language_loss": 0.68532926,
      "learning_rate": 6.994068948661592e-08,
      "loss": 0.70698977,
      "num_input_tokens_seen": 165035575,
      "step": 7635,
      "time_per_iteration": 2.519930362701416
    },
    {
      "auxiliary_loss_clip": 0.01150302,
      "auxiliary_loss_mlp": 0.01024838,
      "balance_loss_clip": 1.04553819,
      "balance_loss_mlp": 1.01688635,
      "epoch": 0.9181747129200986,
      "flos": 16727478301440.0,
      "grad_norm": 2.31083351932169,
      "language_loss": 0.7637673,
      "learning_rate": 6.973663568654142e-08,
      "loss": 0.78551865,
      "num_input_tokens_seen": 165053280,
      "step": 7636,
      "time_per_iteration": 2.4259276390075684
    },
    {
      "auxiliary_loss_clip": 0.01163957,
      "auxiliary_loss_mlp": 0.0102963,
      "balance_loss_clip": 1.0475173,
      "balance_loss_mlp": 1.02247739,
      "epoch": 0.9182949558107377,
      "flos": 24271626873600.0,
      "grad_norm": 2.4403909420445866,
      "language_loss": 0.65316081,
      "learning_rate": 6.953287470588386e-08,
      "loss": 0.67509669,
      "num_input_tokens_seen": 165071235,
      "step": 7637,
      "time_per_iteration": 2.438974380493164
    },
    {
      "auxiliary_loss_clip": 0.01154247,
      "auxiliary_loss_mlp": 0.01024416,
      "balance_loss_clip": 1.04480255,
      "balance_loss_mlp": 1.01719177,
      "epoch": 0.9184151987013768,
      "flos": 22082117443200.0,
      "grad_norm": 2.2356865436449294,
      "language_loss": 0.86101222,
      "learning_rate": 6.932940657555452e-08,
      "loss": 0.88279891,
      "num_input_tokens_seen": 165087365,
      "step": 7638,
      "time_per_iteration": 2.4101572036743164
    },
    {
      "auxiliary_loss_clip": 0.01158337,
      "auxiliary_loss_mlp": 0.01021286,
      "balance_loss_clip": 1.04486382,
      "balance_loss_mlp": 1.0150187,
      "epoch": 0.9185354415920158,
      "flos": 32166732257280.0,
      "grad_norm": 1.3898957626803734,
      "language_loss": 0.76542276,
      "learning_rate": 6.912623132641938e-08,
      "loss": 0.78721899,
      "num_input_tokens_seen": 165112455,
      "step": 7639,
      "time_per_iteration": 2.5528690814971924
    },
    {
      "auxiliary_loss_clip": 0.01138677,
      "auxiliary_loss_mlp": 0.01027123,
      "balance_loss_clip": 1.04500747,
      "balance_loss_mlp": 1.01977324,
      "epoch": 0.918655684482655,
      "flos": 20997831542400.0,
      "grad_norm": 1.7388803268277686,
      "language_loss": 0.76233828,
      "learning_rate": 6.892334898929952e-08,
      "loss": 0.78399622,
      "num_input_tokens_seen": 165132700,
      "step": 7640,
      "time_per_iteration": 3.331432819366455
    },
    {
      "auxiliary_loss_clip": 0.01144526,
      "auxiliary_loss_mlp": 0.01024127,
      "balance_loss_clip": 1.04361367,
      "balance_loss_mlp": 1.01712942,
      "epoch": 0.918775927373294,
      "flos": 15560704817280.0,
      "grad_norm": 4.224198701075811,
      "language_loss": 0.84922671,
      "learning_rate": 6.872075959497236e-08,
      "loss": 0.87091327,
      "num_input_tokens_seen": 165151475,
      "step": 7641,
      "time_per_iteration": 2.4294159412384033
    },
    {
      "auxiliary_loss_clip": 0.01151841,
      "auxiliary_loss_mlp": 0.0102423,
      "balance_loss_clip": 1.04352629,
      "balance_loss_mlp": 1.01765573,
      "epoch": 0.9188961702639331,
      "flos": 29934057657600.0,
      "grad_norm": 1.834512383712307,
      "language_loss": 0.82921076,
      "learning_rate": 6.85184631741702e-08,
      "loss": 0.85097146,
      "num_input_tokens_seen": 165172040,
      "step": 7642,
      "time_per_iteration": 3.274141550064087
    },
    {
      "auxiliary_loss_clip": 0.01149807,
      "auxiliary_loss_mlp": 0.01022051,
      "balance_loss_clip": 1.0444268,
      "balance_loss_mlp": 1.01463604,
      "epoch": 0.9190164131545723,
      "flos": 20701244943360.0,
      "grad_norm": 2.053465059566003,
      "language_loss": 0.77197266,
      "learning_rate": 6.831645975758161e-08,
      "loss": 0.79369128,
      "num_input_tokens_seen": 165189980,
      "step": 7643,
      "time_per_iteration": 3.158571243286133
    },
    {
      "auxiliary_loss_clip": 0.01131273,
      "auxiliary_loss_mlp": 0.01025614,
      "balance_loss_clip": 1.04436064,
      "balance_loss_mlp": 1.0187242,
      "epoch": 0.9191366560452113,
      "flos": 25629912696960.0,
      "grad_norm": 1.8956614170926007,
      "language_loss": 0.67449749,
      "learning_rate": 6.811474937585026e-08,
      "loss": 0.69606632,
      "num_input_tokens_seen": 165209770,
      "step": 7644,
      "time_per_iteration": 2.4951059818267822
    },
    {
      "auxiliary_loss_clip": 0.0111854,
      "auxiliary_loss_mlp": 0.01022489,
      "balance_loss_clip": 1.04157495,
      "balance_loss_mlp": 1.01566386,
      "epoch": 0.9192568989358504,
      "flos": 21434325615360.0,
      "grad_norm": 1.7229902207045182,
      "language_loss": 0.79311526,
      "learning_rate": 6.79133320595755e-08,
      "loss": 0.81452554,
      "num_input_tokens_seen": 165229690,
      "step": 7645,
      "time_per_iteration": 2.5088160037994385
    },
    {
      "auxiliary_loss_clip": 0.01122449,
      "auxiliary_loss_mlp": 0.00761736,
      "balance_loss_clip": 1.04321325,
      "balance_loss_mlp": 1.00047684,
      "epoch": 0.9193771418264896,
      "flos": 23185078416000.0,
      "grad_norm": 1.6575532127398032,
      "language_loss": 0.75533366,
      "learning_rate": 6.771220783931198e-08,
      "loss": 0.77417552,
      "num_input_tokens_seen": 165249850,
      "step": 7646,
      "time_per_iteration": 2.5735199451446533
    },
    {
      "auxiliary_loss_clip": 0.01003497,
      "auxiliary_loss_mlp": 0.0100263,
      "balance_loss_clip": 1.02024174,
      "balance_loss_mlp": 1.00112808,
      "epoch": 0.9194973847171286,
      "flos": 70582963184640.0,
      "grad_norm": 0.9037482616793509,
      "language_loss": 0.64680159,
      "learning_rate": 6.751137674556994e-08,
      "loss": 0.66686285,
      "num_input_tokens_seen": 165310235,
      "step": 7647,
      "time_per_iteration": 3.3206405639648438
    },
    {
      "auxiliary_loss_clip": 0.01151934,
      "auxiliary_loss_mlp": 0.01020698,
      "balance_loss_clip": 1.04307282,
      "balance_loss_mlp": 1.01377749,
      "epoch": 0.9196176276077677,
      "flos": 14720682378240.0,
      "grad_norm": 2.176927481258253,
      "language_loss": 0.77329636,
      "learning_rate": 6.731083880881572e-08,
      "loss": 0.79502273,
      "num_input_tokens_seen": 165326455,
      "step": 7648,
      "time_per_iteration": 2.616607666015625
    },
    {
      "auxiliary_loss_clip": 0.01136327,
      "auxiliary_loss_mlp": 0.01021587,
      "balance_loss_clip": 1.04371262,
      "balance_loss_mlp": 1.01500916,
      "epoch": 0.9197378704984068,
      "flos": 23294893271040.0,
      "grad_norm": 2.090158422838104,
      "language_loss": 0.8082872,
      "learning_rate": 6.711059405947072e-08,
      "loss": 0.82986629,
      "num_input_tokens_seen": 165344645,
      "step": 7649,
      "time_per_iteration": 2.494624137878418
    },
    {
      "auxiliary_loss_clip": 0.01119597,
      "auxiliary_loss_mlp": 0.01022311,
      "balance_loss_clip": 1.0427438,
      "balance_loss_mlp": 1.01512527,
      "epoch": 0.9198581133890459,
      "flos": 20302564913280.0,
      "grad_norm": 2.082361382738563,
      "language_loss": 0.77008837,
      "learning_rate": 6.691064252791156e-08,
      "loss": 0.79150748,
      "num_input_tokens_seen": 165364120,
      "step": 7650,
      "time_per_iteration": 3.2656569480895996
    },
    {
      "auxiliary_loss_clip": 0.01102726,
      "auxiliary_loss_mlp": 0.01024218,
      "balance_loss_clip": 1.0415051,
      "balance_loss_mlp": 1.01702416,
      "epoch": 0.9199783562796849,
      "flos": 17675663569920.0,
      "grad_norm": 1.6669751251756852,
      "language_loss": 0.77895582,
      "learning_rate": 6.67109842444713e-08,
      "loss": 0.80022526,
      "num_input_tokens_seen": 165383050,
      "step": 7651,
      "time_per_iteration": 2.512842893600464
    },
    {
      "auxiliary_loss_clip": 0.01153345,
      "auxiliary_loss_mlp": 0.00762696,
      "balance_loss_clip": 1.05007184,
      "balance_loss_mlp": 1.00045204,
      "epoch": 0.9200985991703241,
      "flos": 17676022705920.0,
      "grad_norm": 1.991096165951223,
      "language_loss": 0.76562107,
      "learning_rate": 6.651161923943704e-08,
      "loss": 0.7847814,
      "num_input_tokens_seen": 165400955,
      "step": 7652,
      "time_per_iteration": 2.4329683780670166
    },
    {
      "auxiliary_loss_clip": 0.01146242,
      "auxiliary_loss_mlp": 0.01027349,
      "balance_loss_clip": 1.04374027,
      "balance_loss_mlp": 1.0198952,
      "epoch": 0.9202188420609632,
      "flos": 20996574566400.0,
      "grad_norm": 2.1668702760713856,
      "language_loss": 0.77489841,
      "learning_rate": 6.631254754305326e-08,
      "loss": 0.79663432,
      "num_input_tokens_seen": 165420415,
      "step": 7653,
      "time_per_iteration": 2.442662477493286
    },
    {
      "auxiliary_loss_clip": 0.01167091,
      "auxiliary_loss_mlp": 0.0102357,
      "balance_loss_clip": 1.04695463,
      "balance_loss_mlp": 1.01641774,
      "epoch": 0.9203390849516022,
      "flos": 13918222586880.0,
      "grad_norm": 2.497854279128939,
      "language_loss": 0.77975565,
      "learning_rate": 6.611376918551848e-08,
      "loss": 0.80166227,
      "num_input_tokens_seen": 165439200,
      "step": 7654,
      "time_per_iteration": 2.3864758014678955
    },
    {
      "auxiliary_loss_clip": 0.01121223,
      "auxiliary_loss_mlp": 0.00762089,
      "balance_loss_clip": 1.041278,
      "balance_loss_mlp": 1.00042582,
      "epoch": 0.9204593278422414,
      "flos": 21175912195200.0,
      "grad_norm": 2.0443332873869093,
      "language_loss": 0.79564846,
      "learning_rate": 6.591528419698744e-08,
      "loss": 0.81448162,
      "num_input_tokens_seen": 165458985,
      "step": 7655,
      "time_per_iteration": 2.5202560424804688
    },
    {
      "auxiliary_loss_clip": 0.01139449,
      "auxiliary_loss_mlp": 0.01025383,
      "balance_loss_clip": 1.04314375,
      "balance_loss_mlp": 1.01906824,
      "epoch": 0.9205795707328804,
      "flos": 14501375890560.0,
      "grad_norm": 2.36874502198062,
      "language_loss": 0.83454204,
      "learning_rate": 6.571709260756986e-08,
      "loss": 0.85619038,
      "num_input_tokens_seen": 165475630,
      "step": 7656,
      "time_per_iteration": 2.436110734939575
    },
    {
      "auxiliary_loss_clip": 0.01155873,
      "auxiliary_loss_mlp": 0.01028997,
      "balance_loss_clip": 1.05001628,
      "balance_loss_mlp": 1.02145672,
      "epoch": 0.9206998136235195,
      "flos": 22417559579520.0,
      "grad_norm": 2.944361512186727,
      "language_loss": 0.76342273,
      "learning_rate": 6.551919444733122e-08,
      "loss": 0.78527141,
      "num_input_tokens_seen": 165493445,
      "step": 7657,
      "time_per_iteration": 2.427248954772949
    },
    {
      "auxiliary_loss_clip": 0.01136374,
      "auxiliary_loss_mlp": 0.01025271,
      "balance_loss_clip": 1.04570186,
      "balance_loss_mlp": 1.01783276,
      "epoch": 0.9208200565141585,
      "flos": 53358407544960.0,
      "grad_norm": 3.6060789658169665,
      "language_loss": 0.65963018,
      "learning_rate": 6.53215897462931e-08,
      "loss": 0.68124658,
      "num_input_tokens_seen": 165517200,
      "step": 7658,
      "time_per_iteration": 2.7555553913116455
    },
    {
      "auxiliary_loss_clip": 0.01148493,
      "auxiliary_loss_mlp": 0.01027063,
      "balance_loss_clip": 1.04461527,
      "balance_loss_mlp": 1.01982403,
      "epoch": 0.9209402994047977,
      "flos": 30589139946240.0,
      "grad_norm": 2.2499708108735508,
      "language_loss": 0.74821383,
      "learning_rate": 6.512427853443103e-08,
      "loss": 0.76996934,
      "num_input_tokens_seen": 165539280,
      "step": 7659,
      "time_per_iteration": 2.517777919769287
    },
    {
      "auxiliary_loss_clip": 0.0115339,
      "auxiliary_loss_mlp": 0.01019071,
      "balance_loss_clip": 1.04559326,
      "balance_loss_mlp": 1.01173377,
      "epoch": 0.9210605422954368,
      "flos": 29132711187840.0,
      "grad_norm": 1.6276872320456355,
      "language_loss": 0.75560385,
      "learning_rate": 6.492726084167799e-08,
      "loss": 0.77732849,
      "num_input_tokens_seen": 165561395,
      "step": 7660,
      "time_per_iteration": 2.506317138671875
    },
    {
      "auxiliary_loss_clip": 0.01060779,
      "auxiliary_loss_mlp": 0.0100092,
      "balance_loss_clip": 1.00725389,
      "balance_loss_mlp": 1.00006723,
      "epoch": 0.9211807851860758,
      "flos": 54853838472960.0,
      "grad_norm": 0.7772790524315556,
      "language_loss": 0.57487422,
      "learning_rate": 6.473053669792072e-08,
      "loss": 0.59549117,
      "num_input_tokens_seen": 165616085,
      "step": 7661,
      "time_per_iteration": 2.8749241828918457
    },
    {
      "auxiliary_loss_clip": 0.01148598,
      "auxiliary_loss_mlp": 0.0102397,
      "balance_loss_clip": 1.04341173,
      "balance_loss_mlp": 1.01619124,
      "epoch": 0.921301028076715,
      "flos": 19201974238080.0,
      "grad_norm": 6.1458698094312405,
      "language_loss": 0.72845161,
      "learning_rate": 6.453410613300248e-08,
      "loss": 0.75017726,
      "num_input_tokens_seen": 165634015,
      "step": 7662,
      "time_per_iteration": 2.4140779972076416
    },
    {
      "auxiliary_loss_clip": 0.01094105,
      "auxiliary_loss_mlp": 0.01027295,
      "balance_loss_clip": 1.04023457,
      "balance_loss_mlp": 1.01992822,
      "epoch": 0.921421270967354,
      "flos": 27526893765120.0,
      "grad_norm": 1.671496699512302,
      "language_loss": 0.58453703,
      "learning_rate": 6.43379691767214e-08,
      "loss": 0.60575104,
      "num_input_tokens_seen": 165653220,
      "step": 7663,
      "time_per_iteration": 2.6053354740142822
    },
    {
      "auxiliary_loss_clip": 0.01022273,
      "auxiliary_loss_mlp": 0.01001594,
      "balance_loss_clip": 1.0084753,
      "balance_loss_mlp": 1.0007118,
      "epoch": 0.9215415138579931,
      "flos": 70209311955840.0,
      "grad_norm": 0.7166928675433318,
      "language_loss": 0.55127835,
      "learning_rate": 6.414212585883105e-08,
      "loss": 0.57151705,
      "num_input_tokens_seen": 165715850,
      "step": 7664,
      "time_per_iteration": 3.1619186401367188
    },
    {
      "auxiliary_loss_clip": 0.01140502,
      "auxiliary_loss_mlp": 0.01020532,
      "balance_loss_clip": 1.04497957,
      "balance_loss_mlp": 1.01301932,
      "epoch": 0.9216617567486323,
      "flos": 35553107790720.0,
      "grad_norm": 1.6033462203750817,
      "language_loss": 0.7004081,
      "learning_rate": 6.394657620904143e-08,
      "loss": 0.72201842,
      "num_input_tokens_seen": 165738960,
      "step": 7665,
      "time_per_iteration": 2.588714838027954
    },
    {
      "auxiliary_loss_clip": 0.01169377,
      "auxiliary_loss_mlp": 0.01026027,
      "balance_loss_clip": 1.04820788,
      "balance_loss_mlp": 1.01867187,
      "epoch": 0.9217819996392713,
      "flos": 29533330552320.0,
      "grad_norm": 1.7149657571500336,
      "language_loss": 0.72017533,
      "learning_rate": 6.375132025701657e-08,
      "loss": 0.74212939,
      "num_input_tokens_seen": 165761260,
      "step": 7666,
      "time_per_iteration": 2.499300956726074
    },
    {
      "auxiliary_loss_clip": 0.0116913,
      "auxiliary_loss_mlp": 0.01024412,
      "balance_loss_clip": 1.04929817,
      "balance_loss_mlp": 1.01709533,
      "epoch": 0.9219022425299104,
      "flos": 14574669592320.0,
      "grad_norm": 2.151956093164181,
      "language_loss": 0.69253838,
      "learning_rate": 6.355635803237724e-08,
      "loss": 0.71447384,
      "num_input_tokens_seen": 165776960,
      "step": 7667,
      "time_per_iteration": 3.2311604022979736
    },
    {
      "auxiliary_loss_clip": 0.01148691,
      "auxiliary_loss_mlp": 0.01024621,
      "balance_loss_clip": 1.04342127,
      "balance_loss_mlp": 1.01731324,
      "epoch": 0.9220224854205495,
      "flos": 18077503996800.0,
      "grad_norm": 1.9775197839352552,
      "language_loss": 0.79580146,
      "learning_rate": 6.336168956469867e-08,
      "loss": 0.81753457,
      "num_input_tokens_seen": 165795435,
      "step": 7668,
      "time_per_iteration": 2.414461851119995
    },
    {
      "auxiliary_loss_clip": 0.01129756,
      "auxiliary_loss_mlp": 0.0102533,
      "balance_loss_clip": 1.04336774,
      "balance_loss_mlp": 1.01898789,
      "epoch": 0.9221427283111886,
      "flos": 24790464875520.0,
      "grad_norm": 1.7744680438508844,
      "language_loss": 0.72014236,
      "learning_rate": 6.316731488351168e-08,
      "loss": 0.74169326,
      "num_input_tokens_seen": 165816625,
      "step": 7669,
      "time_per_iteration": 3.2863781452178955
    },
    {
      "auxiliary_loss_clip": 0.01150566,
      "auxiliary_loss_mlp": 0.01022543,
      "balance_loss_clip": 1.04579663,
      "balance_loss_mlp": 1.01539671,
      "epoch": 0.9222629712018277,
      "flos": 13845036625920.0,
      "grad_norm": 1.793424297577179,
      "language_loss": 0.635225,
      "learning_rate": 6.297323401830334e-08,
      "loss": 0.65695614,
      "num_input_tokens_seen": 165835410,
      "step": 7670,
      "time_per_iteration": 3.237065315246582
    },
    {
      "auxiliary_loss_clip": 0.01153341,
      "auxiliary_loss_mlp": 0.0102241,
      "balance_loss_clip": 1.04544127,
      "balance_loss_mlp": 1.01544249,
      "epoch": 0.9223832140924668,
      "flos": 21616177196160.0,
      "grad_norm": 2.8107892334190265,
      "language_loss": 0.69219255,
      "learning_rate": 6.277944699851523e-08,
      "loss": 0.71395004,
      "num_input_tokens_seen": 165854930,
      "step": 7671,
      "time_per_iteration": 2.4299607276916504
    },
    {
      "auxiliary_loss_clip": 0.01162791,
      "auxiliary_loss_mlp": 0.01025976,
      "balance_loss_clip": 1.04550004,
      "balance_loss_mlp": 1.01891541,
      "epoch": 0.9225034569831059,
      "flos": 21142084561920.0,
      "grad_norm": 2.3558633985658775,
      "language_loss": 0.73348355,
      "learning_rate": 6.25859538535447e-08,
      "loss": 0.75537121,
      "num_input_tokens_seen": 165875725,
      "step": 7672,
      "time_per_iteration": 2.3961212635040283
    },
    {
      "auxiliary_loss_clip": 0.01136135,
      "auxiliary_loss_mlp": 0.01022535,
      "balance_loss_clip": 1.04473019,
      "balance_loss_mlp": 1.01548386,
      "epoch": 0.9226236998737449,
      "flos": 12495046844160.0,
      "grad_norm": 2.4901182051713526,
      "language_loss": 0.78354412,
      "learning_rate": 6.239275461274474e-08,
      "loss": 0.80513084,
      "num_input_tokens_seen": 165892100,
      "step": 7673,
      "time_per_iteration": 2.426987886428833
    },
    {
      "auxiliary_loss_clip": 0.01152056,
      "auxiliary_loss_mlp": 0.01027172,
      "balance_loss_clip": 1.04599643,
      "balance_loss_mlp": 1.02035928,
      "epoch": 0.9227439427643841,
      "flos": 26214071581440.0,
      "grad_norm": 1.9298177639148921,
      "language_loss": 0.85972744,
      "learning_rate": 6.219984930542299e-08,
      "loss": 0.88151968,
      "num_input_tokens_seen": 165912840,
      "step": 7674,
      "time_per_iteration": 2.4786314964294434
    },
    {
      "auxiliary_loss_clip": 0.01152311,
      "auxiliary_loss_mlp": 0.01027124,
      "balance_loss_clip": 1.04498756,
      "balance_loss_mlp": 1.02031684,
      "epoch": 0.9228641856550232,
      "flos": 17967581400960.0,
      "grad_norm": 2.4581083088160924,
      "language_loss": 0.75962865,
      "learning_rate": 6.200723796084383e-08,
      "loss": 0.78142297,
      "num_input_tokens_seen": 165930935,
      "step": 7675,
      "time_per_iteration": 2.3962435722351074
    },
    {
      "auxiliary_loss_clip": 0.01035324,
      "auxiliary_loss_mlp": 0.01001526,
      "balance_loss_clip": 1.00832105,
      "balance_loss_mlp": 1.00054836,
      "epoch": 0.9229844285456622,
      "flos": 70420609710720.0,
      "grad_norm": 0.7649790507259957,
      "language_loss": 0.63101447,
      "learning_rate": 6.181492060822546e-08,
      "loss": 0.65138292,
      "num_input_tokens_seen": 165991110,
      "step": 7676,
      "time_per_iteration": 2.991464138031006
    },
    {
      "auxiliary_loss_clip": 0.01107256,
      "auxiliary_loss_mlp": 0.0102243,
      "balance_loss_clip": 1.04085457,
      "balance_loss_mlp": 1.01518166,
      "epoch": 0.9231046714363014,
      "flos": 17967832796160.0,
      "grad_norm": 2.1056106856434056,
      "language_loss": 0.81498748,
      "learning_rate": 6.162289727674274e-08,
      "loss": 0.8362844,
      "num_input_tokens_seen": 166008790,
      "step": 7677,
      "time_per_iteration": 3.2753589153289795
    },
    {
      "auxiliary_loss_clip": 0.01122257,
      "auxiliary_loss_mlp": 0.0102265,
      "balance_loss_clip": 1.04182434,
      "balance_loss_mlp": 1.01609945,
      "epoch": 0.9232249143269404,
      "flos": 17858233422720.0,
      "grad_norm": 2.154044540981909,
      "language_loss": 0.87877572,
      "learning_rate": 6.143116799552527e-08,
      "loss": 0.9002248,
      "num_input_tokens_seen": 166025035,
      "step": 7678,
      "time_per_iteration": 2.524458885192871
    },
    {
      "auxiliary_loss_clip": 0.01155175,
      "auxiliary_loss_mlp": 0.01022785,
      "balance_loss_clip": 1.04737461,
      "balance_loss_mlp": 1.01561451,
      "epoch": 0.9233451572175795,
      "flos": 23404384903680.0,
      "grad_norm": 2.3157741603411695,
      "language_loss": 0.55908358,
      "learning_rate": 6.123973279365802e-08,
      "loss": 0.58086324,
      "num_input_tokens_seen": 166044010,
      "step": 7679,
      "time_per_iteration": 2.468770742416382
    },
    {
      "auxiliary_loss_clip": 0.01155081,
      "auxiliary_loss_mlp": 0.01023463,
      "balance_loss_clip": 1.04675853,
      "balance_loss_mlp": 1.01700437,
      "epoch": 0.9234654001082186,
      "flos": 17999326045440.0,
      "grad_norm": 2.0193275203489454,
      "language_loss": 0.7778011,
      "learning_rate": 6.10485917001824e-08,
      "loss": 0.79958653,
      "num_input_tokens_seen": 166061865,
      "step": 7680,
      "time_per_iteration": 2.422494649887085
    },
    {
      "auxiliary_loss_clip": 0.0114014,
      "auxiliary_loss_mlp": 0.01021268,
      "balance_loss_clip": 1.04387641,
      "balance_loss_mlp": 1.01465154,
      "epoch": 0.9235856429988577,
      "flos": 24750747411840.0,
      "grad_norm": 1.6685948186289121,
      "language_loss": 0.81034184,
      "learning_rate": 6.085774474409322e-08,
      "loss": 0.83195591,
      "num_input_tokens_seen": 166082425,
      "step": 7681,
      "time_per_iteration": 2.503793716430664
    },
    {
      "auxiliary_loss_clip": 0.01138869,
      "auxiliary_loss_mlp": 0.01028445,
      "balance_loss_clip": 1.04790807,
      "balance_loss_mlp": 1.02164137,
      "epoch": 0.9237058858894968,
      "flos": 14099894599680.0,
      "grad_norm": 2.1480405752097687,
      "language_loss": 0.70128691,
      "learning_rate": 6.066719195434267e-08,
      "loss": 0.72296011,
      "num_input_tokens_seen": 166100225,
      "step": 7682,
      "time_per_iteration": 2.4714064598083496
    },
    {
      "auxiliary_loss_clip": 0.01153123,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.04660714,
      "balance_loss_mlp": 1.01844883,
      "epoch": 0.9238261287801359,
      "flos": 28694529175680.0,
      "grad_norm": 2.3167486633050283,
      "language_loss": 0.66617912,
      "learning_rate": 6.047693335983717e-08,
      "loss": 0.68796605,
      "num_input_tokens_seen": 166122570,
      "step": 7683,
      "time_per_iteration": 2.499300241470337
    },
    {
      "auxiliary_loss_clip": 0.01153306,
      "auxiliary_loss_mlp": 0.01023919,
      "balance_loss_clip": 1.04448271,
      "balance_loss_mlp": 1.01668024,
      "epoch": 0.923946371670775,
      "flos": 23111856541440.0,
      "grad_norm": 2.5498969938233595,
      "language_loss": 0.82253832,
      "learning_rate": 6.028696898943853e-08,
      "loss": 0.84431058,
      "num_input_tokens_seen": 166141630,
      "step": 7684,
      "time_per_iteration": 2.450551748275757
    },
    {
      "auxiliary_loss_clip": 0.01136665,
      "auxiliary_loss_mlp": 0.00762501,
      "balance_loss_clip": 1.04158926,
      "balance_loss_mlp": 1.00044954,
      "epoch": 0.924066614561414,
      "flos": 21867120587520.0,
      "grad_norm": 1.9420646668544417,
      "language_loss": 0.70732617,
      "learning_rate": 6.00972988719648e-08,
      "loss": 0.72631782,
      "num_input_tokens_seen": 166159865,
      "step": 7685,
      "time_per_iteration": 2.4805076122283936
    },
    {
      "auxiliary_loss_clip": 0.01126211,
      "auxiliary_loss_mlp": 0.00762283,
      "balance_loss_clip": 1.04354739,
      "balance_loss_mlp": 1.00046277,
      "epoch": 0.9241868574520532,
      "flos": 28511887495680.0,
      "grad_norm": 2.324000694529363,
      "language_loss": 0.70573151,
      "learning_rate": 5.990792303618807e-08,
      "loss": 0.72461641,
      "num_input_tokens_seen": 166179445,
      "step": 7686,
      "time_per_iteration": 2.560917615890503
    },
    {
      "auxiliary_loss_clip": 0.01123514,
      "auxiliary_loss_mlp": 0.01019981,
      "balance_loss_clip": 1.04523516,
      "balance_loss_mlp": 1.01284671,
      "epoch": 0.9243071003426923,
      "flos": 30518324282880.0,
      "grad_norm": 1.5605261130800718,
      "language_loss": 0.69285089,
      "learning_rate": 5.971884151083695e-08,
      "loss": 0.71428585,
      "num_input_tokens_seen": 166201855,
      "step": 7687,
      "time_per_iteration": 2.5913567543029785
    },
    {
      "auxiliary_loss_clip": 0.01138238,
      "auxiliary_loss_mlp": 0.01023886,
      "balance_loss_clip": 1.04310644,
      "balance_loss_mlp": 1.01737392,
      "epoch": 0.9244273432333313,
      "flos": 28658331244800.0,
      "grad_norm": 1.761702590115966,
      "language_loss": 0.74576336,
      "learning_rate": 5.9530054324595124e-08,
      "loss": 0.76738453,
      "num_input_tokens_seen": 166221970,
      "step": 7688,
      "time_per_iteration": 2.5443081855773926
    },
    {
      "auxiliary_loss_clip": 0.01046539,
      "auxiliary_loss_mlp": 0.0075287,
      "balance_loss_clip": 1.00763559,
      "balance_loss_mlp": 0.99982464,
      "epoch": 0.9245475861239704,
      "flos": 66230589237120.0,
      "grad_norm": 0.7171431574459651,
      "language_loss": 0.5757215,
      "learning_rate": 5.934156150610103e-08,
      "loss": 0.59371555,
      "num_input_tokens_seen": 166279335,
      "step": 7689,
      "time_per_iteration": 3.082620143890381
    },
    {
      "auxiliary_loss_clip": 0.01134944,
      "auxiliary_loss_mlp": 0.01024706,
      "balance_loss_clip": 1.04298222,
      "balance_loss_mlp": 1.0172013,
      "epoch": 0.9246678290146095,
      "flos": 24239918142720.0,
      "grad_norm": 3.624306164854581,
      "language_loss": 0.78735125,
      "learning_rate": 5.915336308394914e-08,
      "loss": 0.80894774,
      "num_input_tokens_seen": 166298170,
      "step": 7690,
      "time_per_iteration": 2.485715866088867
    },
    {
      "auxiliary_loss_clip": 0.01145328,
      "auxiliary_loss_mlp": 0.01022568,
      "balance_loss_clip": 1.0442158,
      "balance_loss_mlp": 1.01650369,
      "epoch": 0.9247880719052486,
      "flos": 18988808976000.0,
      "grad_norm": 1.5879498549657893,
      "language_loss": 0.76919603,
      "learning_rate": 5.89654590866886e-08,
      "loss": 0.79087508,
      "num_input_tokens_seen": 166317670,
      "step": 7691,
      "time_per_iteration": 2.4459073543548584
    },
    {
      "auxiliary_loss_clip": 0.01102379,
      "auxiliary_loss_mlp": 0.0102443,
      "balance_loss_clip": 1.04512763,
      "balance_loss_mlp": 1.01674128,
      "epoch": 0.9249083147958876,
      "flos": 24024095274240.0,
      "grad_norm": 1.9296176075650382,
      "language_loss": 0.88285249,
      "learning_rate": 5.877784954282483e-08,
      "loss": 0.90412056,
      "num_input_tokens_seen": 166337010,
      "step": 7692,
      "time_per_iteration": 2.593445301055908
    },
    {
      "auxiliary_loss_clip": 0.01154752,
      "auxiliary_loss_mlp": 0.01022227,
      "balance_loss_clip": 1.04630804,
      "balance_loss_mlp": 1.01444268,
      "epoch": 0.9250285576865268,
      "flos": 30773972355840.0,
      "grad_norm": 1.9143729784944223,
      "language_loss": 0.72114557,
      "learning_rate": 5.8590534480817963e-08,
      "loss": 0.74291539,
      "num_input_tokens_seen": 166358735,
      "step": 7693,
      "time_per_iteration": 3.345564126968384
    },
    {
      "auxiliary_loss_clip": 0.01166166,
      "auxiliary_loss_mlp": 0.01026565,
      "balance_loss_clip": 1.04837728,
      "balance_loss_mlp": 1.01949334,
      "epoch": 0.9251488005771659,
      "flos": 10633581348480.0,
      "grad_norm": 2.2040602087679444,
      "language_loss": 0.72239274,
      "learning_rate": 5.840351392908349e-08,
      "loss": 0.74432003,
      "num_input_tokens_seen": 166374455,
      "step": 7694,
      "time_per_iteration": 2.3825693130493164
    },
    {
      "auxiliary_loss_clip": 0.01144394,
      "auxiliary_loss_mlp": 0.00762162,
      "balance_loss_clip": 1.04419816,
      "balance_loss_mlp": 1.00046086,
      "epoch": 0.9252690434678049,
      "flos": 23586416052480.0,
      "grad_norm": 2.525139122086702,
      "language_loss": 0.70718002,
      "learning_rate": 5.821678791599205e-08,
      "loss": 0.72624552,
      "num_input_tokens_seen": 166393900,
      "step": 7695,
      "time_per_iteration": 3.231208086013794
    },
    {
      "auxiliary_loss_clip": 0.0113516,
      "auxiliary_loss_mlp": 0.01022443,
      "balance_loss_clip": 1.04476511,
      "balance_loss_mlp": 1.01569295,
      "epoch": 0.9253892863584441,
      "flos": 21469158829440.0,
      "grad_norm": 2.467006232156446,
      "language_loss": 0.80672932,
      "learning_rate": 5.803035646986965e-08,
      "loss": 0.82830536,
      "num_input_tokens_seen": 166413235,
      "step": 7696,
      "time_per_iteration": 3.293334484100342
    },
    {
      "auxiliary_loss_clip": 0.01167685,
      "auxiliary_loss_mlp": 0.0102406,
      "balance_loss_clip": 1.04764295,
      "balance_loss_mlp": 1.01640689,
      "epoch": 0.9255095292490831,
      "flos": 17456680304640.0,
      "grad_norm": 2.690685647825443,
      "language_loss": 0.67256463,
      "learning_rate": 5.7844219618998766e-08,
      "loss": 0.69448209,
      "num_input_tokens_seen": 166427560,
      "step": 7697,
      "time_per_iteration": 2.397186756134033
    },
    {
      "auxiliary_loss_clip": 0.01107582,
      "auxiliary_loss_mlp": 0.01022265,
      "balance_loss_clip": 1.03805733,
      "balance_loss_mlp": 1.01510406,
      "epoch": 0.9256297721397222,
      "flos": 24750675584640.0,
      "grad_norm": 1.7160769054601068,
      "language_loss": 0.71809733,
      "learning_rate": 5.765837739161505e-08,
      "loss": 0.7393958,
      "num_input_tokens_seen": 166446680,
      "step": 7698,
      "time_per_iteration": 2.542025566101074
    },
    {
      "auxiliary_loss_clip": 0.01123132,
      "auxiliary_loss_mlp": 0.01021095,
      "balance_loss_clip": 1.04255581,
      "balance_loss_mlp": 1.01413357,
      "epoch": 0.9257500150303614,
      "flos": 23112215677440.0,
      "grad_norm": 2.253933801140475,
      "language_loss": 0.74176347,
      "learning_rate": 5.7472829815911504e-08,
      "loss": 0.76320571,
      "num_input_tokens_seen": 166465505,
      "step": 7699,
      "time_per_iteration": 2.5108723640441895
    },
    {
      "auxiliary_loss_clip": 0.01133904,
      "auxiliary_loss_mlp": 0.01025839,
      "balance_loss_clip": 1.04401207,
      "balance_loss_mlp": 1.01847148,
      "epoch": 0.9258702579210004,
      "flos": 22564685687040.0,
      "grad_norm": 1.7266540174413394,
      "language_loss": 0.81365013,
      "learning_rate": 5.7287576920035164e-08,
      "loss": 0.83524758,
      "num_input_tokens_seen": 166484520,
      "step": 7700,
      "time_per_iteration": 2.488548517227173
    },
    {
      "auxiliary_loss_clip": 0.01120688,
      "auxiliary_loss_mlp": 0.01021268,
      "balance_loss_clip": 1.04389286,
      "balance_loss_mlp": 1.01470518,
      "epoch": 0.9259905008116395,
      "flos": 30004298703360.0,
      "grad_norm": 2.234269878052166,
      "language_loss": 0.76880634,
      "learning_rate": 5.7102618732088435e-08,
      "loss": 0.79022586,
      "num_input_tokens_seen": 166503850,
      "step": 7701,
      "time_per_iteration": 2.5677061080932617
    },
    {
      "auxiliary_loss_clip": 0.01142892,
      "auxiliary_loss_mlp": 0.01027352,
      "balance_loss_clip": 1.04523194,
      "balance_loss_mlp": 1.02100372,
      "epoch": 0.9261107437022786,
      "flos": 24572128055040.0,
      "grad_norm": 1.5401736708876554,
      "language_loss": 0.74428922,
      "learning_rate": 5.6917955280130216e-08,
      "loss": 0.76599169,
      "num_input_tokens_seen": 166525330,
      "step": 7702,
      "time_per_iteration": 2.5289223194122314
    },
    {
      "auxiliary_loss_clip": 0.01149794,
      "auxiliary_loss_mlp": 0.01023195,
      "balance_loss_clip": 1.04583275,
      "balance_loss_mlp": 1.01612616,
      "epoch": 0.9262309865929177,
      "flos": 22018448586240.0,
      "grad_norm": 2.1804399549653306,
      "language_loss": 0.72180104,
      "learning_rate": 5.6733586592172755e-08,
      "loss": 0.74353099,
      "num_input_tokens_seen": 166544825,
      "step": 7703,
      "time_per_iteration": 2.436396360397339
    },
    {
      "auxiliary_loss_clip": 0.01129411,
      "auxiliary_loss_mlp": 0.00761033,
      "balance_loss_clip": 1.04053009,
      "balance_loss_mlp": 1.00041819,
      "epoch": 0.9263512294835567,
      "flos": 20339481116160.0,
      "grad_norm": 2.078291786574832,
      "language_loss": 0.80020046,
      "learning_rate": 5.6549512696185244e-08,
      "loss": 0.81910491,
      "num_input_tokens_seen": 166563325,
      "step": 7704,
      "time_per_iteration": 3.258502960205078
    },
    {
      "auxiliary_loss_clip": 0.01162358,
      "auxiliary_loss_mlp": 0.01021805,
      "balance_loss_clip": 1.04607201,
      "balance_loss_mlp": 1.01470888,
      "epoch": 0.9264714723741959,
      "flos": 21215378263680.0,
      "grad_norm": 1.6007703747131554,
      "language_loss": 0.68425059,
      "learning_rate": 5.636573362009156e-08,
      "loss": 0.70609224,
      "num_input_tokens_seen": 166583385,
      "step": 7705,
      "time_per_iteration": 2.415381669998169
    },
    {
      "auxiliary_loss_clip": 0.01167858,
      "auxiliary_loss_mlp": 0.01026402,
      "balance_loss_clip": 1.04826832,
      "balance_loss_mlp": 1.01919281,
      "epoch": 0.926591715264835,
      "flos": 18004964480640.0,
      "grad_norm": 3.776885144725662,
      "language_loss": 0.77157074,
      "learning_rate": 5.618224939177074e-08,
      "loss": 0.79351336,
      "num_input_tokens_seen": 166601290,
      "step": 7706,
      "time_per_iteration": 2.475008249282837
    },
    {
      "auxiliary_loss_clip": 0.01125939,
      "auxiliary_loss_mlp": 0.0102428,
      "balance_loss_clip": 1.04210711,
      "balance_loss_mlp": 1.01675177,
      "epoch": 0.926711958155474,
      "flos": 36167969825280.0,
      "grad_norm": 1.7146859937973897,
      "language_loss": 0.70432246,
      "learning_rate": 5.599906003905719e-08,
      "loss": 0.72582465,
      "num_input_tokens_seen": 166623835,
      "step": 7707,
      "time_per_iteration": 2.598785877227783
    },
    {
      "auxiliary_loss_clip": 0.01149999,
      "auxiliary_loss_mlp": 0.01024928,
      "balance_loss_clip": 1.04927945,
      "balance_loss_mlp": 1.01750767,
      "epoch": 0.9268322010461132,
      "flos": 21032736583680.0,
      "grad_norm": 2.7246534353093868,
      "language_loss": 0.82195169,
      "learning_rate": 5.581616558974023e-08,
      "loss": 0.843701,
      "num_input_tokens_seen": 166642400,
      "step": 7708,
      "time_per_iteration": 2.4794108867645264
    },
    {
      "auxiliary_loss_clip": 0.01157854,
      "auxiliary_loss_mlp": 0.00762039,
      "balance_loss_clip": 1.0469873,
      "balance_loss_mlp": 1.00044227,
      "epoch": 0.9269524439367522,
      "flos": 22964838174720.0,
      "grad_norm": 2.5884560524548634,
      "language_loss": 0.79461956,
      "learning_rate": 5.5633566071565444e-08,
      "loss": 0.81381845,
      "num_input_tokens_seen": 166661640,
      "step": 7709,
      "time_per_iteration": 2.44502854347229
    },
    {
      "auxiliary_loss_clip": 0.01098035,
      "auxiliary_loss_mlp": 0.01021495,
      "balance_loss_clip": 1.04028416,
      "balance_loss_mlp": 1.01492405,
      "epoch": 0.9270726868273913,
      "flos": 41975551468800.0,
      "grad_norm": 2.2718387141297023,
      "language_loss": 0.70256007,
      "learning_rate": 5.5451261512232896e-08,
      "loss": 0.72375536,
      "num_input_tokens_seen": 166684320,
      "step": 7710,
      "time_per_iteration": 2.740999937057495
    },
    {
      "auxiliary_loss_clip": 0.01155226,
      "auxiliary_loss_mlp": 0.01024191,
      "balance_loss_clip": 1.0435636,
      "balance_loss_mlp": 1.01687133,
      "epoch": 0.9271929297180305,
      "flos": 19791771557760.0,
      "grad_norm": 2.1713689590747838,
      "language_loss": 0.62382555,
      "learning_rate": 5.5269251939397576e-08,
      "loss": 0.64561969,
      "num_input_tokens_seen": 166703835,
      "step": 7711,
      "time_per_iteration": 2.4188764095306396
    },
    {
      "auxiliary_loss_clip": 0.01121206,
      "auxiliary_loss_mlp": 0.01022366,
      "balance_loss_clip": 1.03927302,
      "balance_loss_mlp": 1.01551783,
      "epoch": 0.9273131726086695,
      "flos": 19968343839360.0,
      "grad_norm": 1.9894028545136337,
      "language_loss": 0.76486635,
      "learning_rate": 5.508753738067073e-08,
      "loss": 0.78630203,
      "num_input_tokens_seen": 166723375,
      "step": 7712,
      "time_per_iteration": 2.510390043258667
    },
    {
      "auxiliary_loss_clip": 0.01152427,
      "auxiliary_loss_mlp": 0.01023906,
      "balance_loss_clip": 1.0429548,
      "balance_loss_mlp": 1.01665831,
      "epoch": 0.9274334154993086,
      "flos": 23258587599360.0,
      "grad_norm": 2.1511972505550183,
      "language_loss": 0.79094118,
      "learning_rate": 5.4906117863617875e-08,
      "loss": 0.81270456,
      "num_input_tokens_seen": 166742760,
      "step": 7713,
      "time_per_iteration": 2.449094533920288
    },
    {
      "auxiliary_loss_clip": 0.01119578,
      "auxiliary_loss_mlp": 0.01019551,
      "balance_loss_clip": 1.04015279,
      "balance_loss_mlp": 1.01269639,
      "epoch": 0.9275536583899477,
      "flos": 31795343585280.0,
      "grad_norm": 1.9971500394771942,
      "language_loss": 0.77926266,
      "learning_rate": 5.4724993415760533e-08,
      "loss": 0.80065393,
      "num_input_tokens_seen": 166761115,
      "step": 7714,
      "time_per_iteration": 2.574300527572632
    },
    {
      "auxiliary_loss_clip": 0.01131738,
      "auxiliary_loss_mlp": 0.0076222,
      "balance_loss_clip": 1.04331052,
      "balance_loss_mlp": 1.00049901,
      "epoch": 0.9276739012805868,
      "flos": 18696998885760.0,
      "grad_norm": 2.2041691207075473,
      "language_loss": 0.74647439,
      "learning_rate": 5.454416406457496e-08,
      "loss": 0.765414,
      "num_input_tokens_seen": 166780210,
      "step": 7715,
      "time_per_iteration": 2.484652519226074
    },
    {
      "auxiliary_loss_clip": 0.01150329,
      "auxiliary_loss_mlp": 0.01024749,
      "balance_loss_clip": 1.04465127,
      "balance_loss_mlp": 1.01838326,
      "epoch": 0.9277941441712259,
      "flos": 13879079740800.0,
      "grad_norm": 3.8223549304381845,
      "language_loss": 0.74234211,
      "learning_rate": 5.436362983749299e-08,
      "loss": 0.76409286,
      "num_input_tokens_seen": 166795380,
      "step": 7716,
      "time_per_iteration": 2.382547378540039
    },
    {
      "auxiliary_loss_clip": 0.0111638,
      "auxiliary_loss_mlp": 0.01025423,
      "balance_loss_clip": 1.04309928,
      "balance_loss_mlp": 1.01898551,
      "epoch": 0.927914387061865,
      "flos": 23258659426560.0,
      "grad_norm": 1.9003039239133317,
      "language_loss": 0.64127475,
      "learning_rate": 5.418339076190137e-08,
      "loss": 0.66269279,
      "num_input_tokens_seen": 166814890,
      "step": 7717,
      "time_per_iteration": 2.507246732711792
    },
    {
      "auxiliary_loss_clip": 0.01130186,
      "auxiliary_loss_mlp": 0.01019055,
      "balance_loss_clip": 1.04356813,
      "balance_loss_mlp": 1.01203942,
      "epoch": 0.9280346299525041,
      "flos": 18073733068800.0,
      "grad_norm": 1.824236892508934,
      "language_loss": 0.88926315,
      "learning_rate": 5.400344686514202e-08,
      "loss": 0.91075552,
      "num_input_tokens_seen": 166832475,
      "step": 7718,
      "time_per_iteration": 2.4689576625823975
    },
    {
      "auxiliary_loss_clip": 0.01151405,
      "auxiliary_loss_mlp": 0.01020594,
      "balance_loss_clip": 1.0477742,
      "balance_loss_mlp": 1.01370335,
      "epoch": 0.9281548728431431,
      "flos": 22342901160960.0,
      "grad_norm": 1.928529545349362,
      "language_loss": 0.66944981,
      "learning_rate": 5.38237981745131e-08,
      "loss": 0.69116986,
      "num_input_tokens_seen": 166850590,
      "step": 7719,
      "time_per_iteration": 2.4414563179016113
    },
    {
      "auxiliary_loss_clip": 0.01154081,
      "auxiliary_loss_mlp": 0.00761852,
      "balance_loss_clip": 1.04626405,
      "balance_loss_mlp": 1.00044453,
      "epoch": 0.9282751157337822,
      "flos": 18843765857280.0,
      "grad_norm": 1.7148912425862641,
      "language_loss": 0.81268364,
      "learning_rate": 5.364444471726592e-08,
      "loss": 0.83184296,
      "num_input_tokens_seen": 166869795,
      "step": 7720,
      "time_per_iteration": 3.2681591510772705
    },
    {
      "auxiliary_loss_clip": 0.01150214,
      "auxiliary_loss_mlp": 0.01022147,
      "balance_loss_clip": 1.04457581,
      "balance_loss_mlp": 1.01527441,
      "epoch": 0.9283953586244214,
      "flos": 25556834476800.0,
      "grad_norm": 2.1036630140560133,
      "language_loss": 0.80129731,
      "learning_rate": 5.346538652060939e-08,
      "loss": 0.82302094,
      "num_input_tokens_seen": 166891150,
      "step": 7721,
      "time_per_iteration": 3.2041115760803223
    },
    {
      "auxiliary_loss_clip": 0.0113554,
      "auxiliary_loss_mlp": 0.01023081,
      "balance_loss_clip": 1.04560232,
      "balance_loss_mlp": 1.0163902,
      "epoch": 0.9285156015150604,
      "flos": 18223480869120.0,
      "grad_norm": 1.8075976976897448,
      "language_loss": 0.70257974,
      "learning_rate": 5.3286623611705994e-08,
      "loss": 0.72416592,
      "num_input_tokens_seen": 166909195,
      "step": 7722,
      "time_per_iteration": 2.441241979598999
    },
    {
      "auxiliary_loss_clip": 0.0106071,
      "auxiliary_loss_mlp": 0.01000679,
      "balance_loss_clip": 1.00722241,
      "balance_loss_mlp": 0.99983847,
      "epoch": 0.9286358444056995,
      "flos": 66400017690240.0,
      "grad_norm": 0.8209562883633076,
      "language_loss": 0.60578537,
      "learning_rate": 5.3108156017673824e-08,
      "loss": 0.62639928,
      "num_input_tokens_seen": 166970955,
      "step": 7723,
      "time_per_iteration": 3.8582022190093994
    },
    {
      "auxiliary_loss_clip": 0.01143751,
      "auxiliary_loss_mlp": 0.01024558,
      "balance_loss_clip": 1.04489994,
      "balance_loss_mlp": 1.01667273,
      "epoch": 0.9287560872963386,
      "flos": 22345630594560.0,
      "grad_norm": 1.7369532673867514,
      "language_loss": 0.71586514,
      "learning_rate": 5.2929983765586775e-08,
      "loss": 0.73754817,
      "num_input_tokens_seen": 166989735,
      "step": 7724,
      "time_per_iteration": 2.4824748039245605
    },
    {
      "auxiliary_loss_clip": 0.01168027,
      "auxiliary_loss_mlp": 0.01025992,
      "balance_loss_clip": 1.05047107,
      "balance_loss_mlp": 1.01937258,
      "epoch": 0.9288763301869777,
      "flos": 25700225569920.0,
      "grad_norm": 1.9292333302726883,
      "language_loss": 0.62602824,
      "learning_rate": 5.275210688247278e-08,
      "loss": 0.64796841,
      "num_input_tokens_seen": 167010060,
      "step": 7725,
      "time_per_iteration": 2.4439094066619873
    },
    {
      "auxiliary_loss_clip": 0.01110409,
      "auxiliary_loss_mlp": 0.01024357,
      "balance_loss_clip": 1.04322231,
      "balance_loss_mlp": 1.01732993,
      "epoch": 0.9289965730776167,
      "flos": 12312046028160.0,
      "grad_norm": 2.5179880367599026,
      "language_loss": 0.84967697,
      "learning_rate": 5.257452539531604e-08,
      "loss": 0.87102473,
      "num_input_tokens_seen": 167027130,
      "step": 7726,
      "time_per_iteration": 2.5316436290740967
    },
    {
      "auxiliary_loss_clip": 0.01151051,
      "auxiliary_loss_mlp": 0.01025356,
      "balance_loss_clip": 1.04416776,
      "balance_loss_mlp": 1.01818252,
      "epoch": 0.9291168159682559,
      "flos": 26685973486080.0,
      "grad_norm": 1.712772720054803,
      "language_loss": 0.68287933,
      "learning_rate": 5.2397239331055445e-08,
      "loss": 0.70464337,
      "num_input_tokens_seen": 167049130,
      "step": 7727,
      "time_per_iteration": 2.495070219039917
    },
    {
      "auxiliary_loss_clip": 0.01133389,
      "auxiliary_loss_mlp": 0.01022613,
      "balance_loss_clip": 1.04475236,
      "balance_loss_mlp": 1.01545489,
      "epoch": 0.929237058858895,
      "flos": 14538256179840.0,
      "grad_norm": 1.9723114897677214,
      "language_loss": 0.81500828,
      "learning_rate": 5.2220248716585036e-08,
      "loss": 0.8365683,
      "num_input_tokens_seen": 167066810,
      "step": 7728,
      "time_per_iteration": 2.4856009483337402
    },
    {
      "auxiliary_loss_clip": 0.01142898,
      "auxiliary_loss_mlp": 0.01026097,
      "balance_loss_clip": 1.04326773,
      "balance_loss_mlp": 1.01881921,
      "epoch": 0.929357301749534,
      "flos": 23835456023040.0,
      "grad_norm": 2.0918458255568253,
      "language_loss": 0.75484776,
      "learning_rate": 5.204355357875445e-08,
      "loss": 0.77653766,
      "num_input_tokens_seen": 167085155,
      "step": 7729,
      "time_per_iteration": 2.4698424339294434
    },
    {
      "auxiliary_loss_clip": 0.01134139,
      "auxiliary_loss_mlp": 0.01021903,
      "balance_loss_clip": 1.0421741,
      "balance_loss_mlp": 1.01469111,
      "epoch": 0.9294775446401732,
      "flos": 12969319046400.0,
      "grad_norm": 3.819865613437238,
      "language_loss": 0.70565271,
      "learning_rate": 5.1867153944367584e-08,
      "loss": 0.72721314,
      "num_input_tokens_seen": 167101545,
      "step": 7730,
      "time_per_iteration": 3.2055585384368896
    },
    {
      "auxiliary_loss_clip": 0.01129575,
      "auxiliary_loss_mlp": 0.0102842,
      "balance_loss_clip": 1.04517806,
      "balance_loss_mlp": 1.02137816,
      "epoch": 0.9295977875308122,
      "flos": 26211809024640.0,
      "grad_norm": 1.9733551291917715,
      "language_loss": 0.73428869,
      "learning_rate": 5.16910498401848e-08,
      "loss": 0.75586867,
      "num_input_tokens_seen": 167120995,
      "step": 7731,
      "time_per_iteration": 2.55090594291687
    },
    {
      "auxiliary_loss_clip": 0.01164265,
      "auxiliary_loss_mlp": 0.01022809,
      "balance_loss_clip": 1.04812241,
      "balance_loss_mlp": 1.01623213,
      "epoch": 0.9297180304214513,
      "flos": 16472297105280.0,
      "grad_norm": 2.761499483817806,
      "language_loss": 0.83834797,
      "learning_rate": 5.151524129292073e-08,
      "loss": 0.86021876,
      "num_input_tokens_seen": 167138890,
      "step": 7732,
      "time_per_iteration": 2.360590934753418
    },
    {
      "auxiliary_loss_clip": 0.01148188,
      "auxiliary_loss_mlp": 0.01025616,
      "balance_loss_clip": 1.04436302,
      "balance_loss_mlp": 1.01897931,
      "epoch": 0.9298382733120905,
      "flos": 24060436859520.0,
      "grad_norm": 1.8887326872437653,
      "language_loss": 0.66394758,
      "learning_rate": 5.1339728329245155e-08,
      "loss": 0.68568558,
      "num_input_tokens_seen": 167159455,
      "step": 7733,
      "time_per_iteration": 2.4520535469055176
    },
    {
      "auxiliary_loss_clip": 0.01170393,
      "auxiliary_loss_mlp": 0.01027838,
      "balance_loss_clip": 1.04811764,
      "balance_loss_mlp": 1.02027082,
      "epoch": 0.9299585162027295,
      "flos": 22127652910080.0,
      "grad_norm": 8.35413144269687,
      "language_loss": 0.78955173,
      "learning_rate": 5.116451097578367e-08,
      "loss": 0.81153399,
      "num_input_tokens_seen": 167178495,
      "step": 7734,
      "time_per_iteration": 2.397625684738159
    },
    {
      "auxiliary_loss_clip": 0.01121863,
      "auxiliary_loss_mlp": 0.01024547,
      "balance_loss_clip": 1.04243064,
      "balance_loss_mlp": 1.01761496,
      "epoch": 0.9300787590933686,
      "flos": 21471780522240.0,
      "grad_norm": 1.605495500273465,
      "language_loss": 0.74439311,
      "learning_rate": 5.0989589259115895e-08,
      "loss": 0.76585722,
      "num_input_tokens_seen": 167199380,
      "step": 7735,
      "time_per_iteration": 2.513848066329956
    },
    {
      "auxiliary_loss_clip": 0.01148072,
      "auxiliary_loss_mlp": 0.01026779,
      "balance_loss_clip": 1.04247499,
      "balance_loss_mlp": 1.01885128,
      "epoch": 0.9301990019840077,
      "flos": 17779588594560.0,
      "grad_norm": 1.8065433676061218,
      "language_loss": 0.71542984,
      "learning_rate": 5.081496320577816e-08,
      "loss": 0.73717839,
      "num_input_tokens_seen": 167216500,
      "step": 7736,
      "time_per_iteration": 2.395390272140503
    },
    {
      "auxiliary_loss_clip": 0.01045717,
      "auxiliary_loss_mlp": 0.01002243,
      "balance_loss_clip": 1.01703501,
      "balance_loss_mlp": 1.00104511,
      "epoch": 0.9303192448746468,
      "flos": 58896122307840.0,
      "grad_norm": 0.9145884418444209,
      "language_loss": 0.61160982,
      "learning_rate": 5.0640632842260835e-08,
      "loss": 0.63208944,
      "num_input_tokens_seen": 167276760,
      "step": 7737,
      "time_per_iteration": 3.1183815002441406
    },
    {
      "auxiliary_loss_clip": 0.01120504,
      "auxiliary_loss_mlp": 0.00761968,
      "balance_loss_clip": 1.04490495,
      "balance_loss_mlp": 1.00042558,
      "epoch": 0.9304394877652858,
      "flos": 57663522172800.0,
      "grad_norm": 1.5205629772131883,
      "language_loss": 0.72742152,
      "learning_rate": 5.0466598195009426e-08,
      "loss": 0.74624628,
      "num_input_tokens_seen": 167303630,
      "step": 7738,
      "time_per_iteration": 2.83030366897583
    },
    {
      "auxiliary_loss_clip": 0.01123498,
      "auxiliary_loss_mlp": 0.01022881,
      "balance_loss_clip": 1.04342103,
      "balance_loss_mlp": 1.01574373,
      "epoch": 0.930559730655925,
      "flos": 20996143603200.0,
      "grad_norm": 1.885378797414096,
      "language_loss": 0.70368576,
      "learning_rate": 5.0292859290425036e-08,
      "loss": 0.72514963,
      "num_input_tokens_seen": 167321500,
      "step": 7739,
      "time_per_iteration": 2.476635456085205
    },
    {
      "auxiliary_loss_clip": 0.01164687,
      "auxiliary_loss_mlp": 0.01022429,
      "balance_loss_clip": 1.04824924,
      "balance_loss_mlp": 1.01592886,
      "epoch": 0.9306799735465641,
      "flos": 23258264376960.0,
      "grad_norm": 1.907589577368997,
      "language_loss": 0.7770347,
      "learning_rate": 5.011941615486348e-08,
      "loss": 0.79890585,
      "num_input_tokens_seen": 167340615,
      "step": 7740,
      "time_per_iteration": 2.418637990951538
    },
    {
      "auxiliary_loss_clip": 0.0116328,
      "auxiliary_loss_mlp": 0.01020981,
      "balance_loss_clip": 1.0453999,
      "balance_loss_mlp": 1.01398671,
      "epoch": 0.9308002164372031,
      "flos": 15231547560960.0,
      "grad_norm": 2.223850965521068,
      "language_loss": 0.84600842,
      "learning_rate": 4.994626881463659e-08,
      "loss": 0.86785102,
      "num_input_tokens_seen": 167356870,
      "step": 7741,
      "time_per_iteration": 2.373375415802002
    },
    {
      "auxiliary_loss_clip": 0.0109311,
      "auxiliary_loss_mlp": 0.01023941,
      "balance_loss_clip": 1.03836513,
      "balance_loss_mlp": 1.01694906,
      "epoch": 0.9309204593278423,
      "flos": 30847481539200.0,
      "grad_norm": 2.4769621717711185,
      "language_loss": 0.71428066,
      "learning_rate": 4.9773417296009814e-08,
      "loss": 0.73545122,
      "num_input_tokens_seen": 167378390,
      "step": 7742,
      "time_per_iteration": 2.5885632038116455
    },
    {
      "auxiliary_loss_clip": 0.01156992,
      "auxiliary_loss_mlp": 0.01028758,
      "balance_loss_clip": 1.04703808,
      "balance_loss_mlp": 1.02171302,
      "epoch": 0.9310407022184813,
      "flos": 23037269950080.0,
      "grad_norm": 3.556905768030258,
      "language_loss": 0.65484577,
      "learning_rate": 4.960086162520527e-08,
      "loss": 0.67670321,
      "num_input_tokens_seen": 167398480,
      "step": 7743,
      "time_per_iteration": 2.469297409057617
    },
    {
      "auxiliary_loss_clip": 0.01117885,
      "auxiliary_loss_mlp": 0.01024157,
      "balance_loss_clip": 1.04314637,
      "balance_loss_mlp": 1.01715684,
      "epoch": 0.9311609451091204,
      "flos": 22127976132480.0,
      "grad_norm": 2.0493586048885253,
      "language_loss": 0.82337803,
      "learning_rate": 4.942860182839936e-08,
      "loss": 0.84479845,
      "num_input_tokens_seen": 167416825,
      "step": 7744,
      "time_per_iteration": 2.5263147354125977
    },
    {
      "auxiliary_loss_clip": 0.01135616,
      "auxiliary_loss_mlp": 0.010244,
      "balance_loss_clip": 1.04458964,
      "balance_loss_mlp": 1.01740253,
      "epoch": 0.9312811879997596,
      "flos": 21099206701440.0,
      "grad_norm": 1.8076055369687123,
      "language_loss": 0.79525733,
      "learning_rate": 4.925663793172341e-08,
      "loss": 0.81685758,
      "num_input_tokens_seen": 167434785,
      "step": 7745,
      "time_per_iteration": 2.477511167526245
    },
    {
      "auxiliary_loss_clip": 0.01034751,
      "auxiliary_loss_mlp": 0.00752947,
      "balance_loss_clip": 1.00645757,
      "balance_loss_mlp": 0.99990386,
      "epoch": 0.9314014308903986,
      "flos": 67148179096320.0,
      "grad_norm": 0.7989216417284355,
      "language_loss": 0.5653435,
      "learning_rate": 4.908496996126477e-08,
      "loss": 0.58322048,
      "num_input_tokens_seen": 167498245,
      "step": 7746,
      "time_per_iteration": 3.1061158180236816
    },
    {
      "auxiliary_loss_clip": 0.01153219,
      "auxiliary_loss_mlp": 0.01028678,
      "balance_loss_clip": 1.050313,
      "balance_loss_mlp": 1.02121568,
      "epoch": 0.9315216737810377,
      "flos": 22565583527040.0,
      "grad_norm": 1.4371260626113949,
      "language_loss": 0.76294762,
      "learning_rate": 4.89135979430646e-08,
      "loss": 0.78476655,
      "num_input_tokens_seen": 167518290,
      "step": 7747,
      "time_per_iteration": 3.307936191558838
    },
    {
      "auxiliary_loss_clip": 0.01166726,
      "auxiliary_loss_mlp": 0.01023054,
      "balance_loss_clip": 1.04971361,
      "balance_loss_mlp": 1.01529002,
      "epoch": 0.9316419166716768,
      "flos": 23984054588160.0,
      "grad_norm": 1.7163494940621176,
      "language_loss": 0.85623944,
      "learning_rate": 4.874252190312078e-08,
      "loss": 0.87813723,
      "num_input_tokens_seen": 167538675,
      "step": 7748,
      "time_per_iteration": 3.2668607234954834
    },
    {
      "auxiliary_loss_clip": 0.01154245,
      "auxiliary_loss_mlp": 0.0102253,
      "balance_loss_clip": 1.04525566,
      "balance_loss_mlp": 1.01544046,
      "epoch": 0.9317621595623159,
      "flos": 30230464688640.0,
      "grad_norm": 1.8446139642175259,
      "language_loss": 0.64664161,
      "learning_rate": 4.857174186738477e-08,
      "loss": 0.66840935,
      "num_input_tokens_seen": 167562025,
      "step": 7749,
      "time_per_iteration": 3.336543560028076
    },
    {
      "auxiliary_loss_clip": 0.01167269,
      "auxiliary_loss_mlp": 0.01024198,
      "balance_loss_clip": 1.04961932,
      "balance_loss_mlp": 1.0169208,
      "epoch": 0.931882402452955,
      "flos": 15742735966080.0,
      "grad_norm": 2.2914542483161298,
      "language_loss": 0.73243737,
      "learning_rate": 4.840125786176408e-08,
      "loss": 0.75435209,
      "num_input_tokens_seen": 167578230,
      "step": 7750,
      "time_per_iteration": 2.376410484313965
    },
    {
      "auxiliary_loss_clip": 0.01134074,
      "auxiliary_loss_mlp": 0.01024268,
      "balance_loss_clip": 1.04345489,
      "balance_loss_mlp": 1.01722527,
      "epoch": 0.932002645343594,
      "flos": 28366521154560.0,
      "grad_norm": 1.8739850468914647,
      "language_loss": 0.77088964,
      "learning_rate": 4.823106991212067e-08,
      "loss": 0.79247302,
      "num_input_tokens_seen": 167597470,
      "step": 7751,
      "time_per_iteration": 2.5340423583984375
    },
    {
      "auxiliary_loss_clip": 0.01151699,
      "auxiliary_loss_mlp": 0.01023895,
      "balance_loss_clip": 1.04476213,
      "balance_loss_mlp": 1.01700163,
      "epoch": 0.9321228882342332,
      "flos": 15341146934400.0,
      "grad_norm": 2.216654090081813,
      "language_loss": 0.83282322,
      "learning_rate": 4.806117804427212e-08,
      "loss": 0.85457915,
      "num_input_tokens_seen": 167615405,
      "step": 7752,
      "time_per_iteration": 2.4378795623779297
    },
    {
      "auxiliary_loss_clip": 0.01144995,
      "auxiliary_loss_mlp": 0.01027329,
      "balance_loss_clip": 1.0427556,
      "balance_loss_mlp": 1.01986051,
      "epoch": 0.9322431311248722,
      "flos": 17895365107200.0,
      "grad_norm": 3.744604978197083,
      "language_loss": 0.63993376,
      "learning_rate": 4.7891582283990926e-08,
      "loss": 0.66165698,
      "num_input_tokens_seen": 167634130,
      "step": 7753,
      "time_per_iteration": 2.406168222427368
    },
    {
      "auxiliary_loss_clip": 0.01122858,
      "auxiliary_loss_mlp": 0.01019465,
      "balance_loss_clip": 1.0418663,
      "balance_loss_mlp": 1.01253879,
      "epoch": 0.9323633740155113,
      "flos": 24169713010560.0,
      "grad_norm": 1.5018631277508483,
      "language_loss": 0.72707492,
      "learning_rate": 4.772228265700473e-08,
      "loss": 0.74849814,
      "num_input_tokens_seen": 167654990,
      "step": 7754,
      "time_per_iteration": 2.5253841876983643
    },
    {
      "auxiliary_loss_clip": 0.01155114,
      "auxiliary_loss_mlp": 0.01025631,
      "balance_loss_clip": 1.0465281,
      "balance_loss_mlp": 1.0185976,
      "epoch": 0.9324836169061504,
      "flos": 15043482927360.0,
      "grad_norm": 2.2490912748123106,
      "language_loss": 0.7558254,
      "learning_rate": 4.75532791889961e-08,
      "loss": 0.77763277,
      "num_input_tokens_seen": 167671690,
      "step": 7755,
      "time_per_iteration": 2.393376350402832
    },
    {
      "auxiliary_loss_clip": 0.01148558,
      "auxiliary_loss_mlp": 0.01024901,
      "balance_loss_clip": 1.04360056,
      "balance_loss_mlp": 1.01761401,
      "epoch": 0.9326038597967895,
      "flos": 18624890332800.0,
      "grad_norm": 1.9383887156947166,
      "language_loss": 0.65799683,
      "learning_rate": 4.738457190560252e-08,
      "loss": 0.67973143,
      "num_input_tokens_seen": 167690800,
      "step": 7756,
      "time_per_iteration": 3.1505818367004395
    },
    {
      "auxiliary_loss_clip": 0.01110736,
      "auxiliary_loss_mlp": 0.01025574,
      "balance_loss_clip": 1.04478145,
      "balance_loss_mlp": 1.0183084,
      "epoch": 0.9327241026874286,
      "flos": 18952646958720.0,
      "grad_norm": 2.0157851272316516,
      "language_loss": 0.78822607,
      "learning_rate": 4.721616083241664e-08,
      "loss": 0.80958921,
      "num_input_tokens_seen": 167709055,
      "step": 7757,
      "time_per_iteration": 2.5190184116363525
    },
    {
      "auxiliary_loss_clip": 0.01147471,
      "auxiliary_loss_mlp": 0.01023643,
      "balance_loss_clip": 1.04522109,
      "balance_loss_mlp": 1.016747,
      "epoch": 0.9328443455780677,
      "flos": 29570282668800.0,
      "grad_norm": 1.6793944576780386,
      "language_loss": 0.77681106,
      "learning_rate": 4.7048045994986684e-08,
      "loss": 0.79852223,
      "num_input_tokens_seen": 167729915,
      "step": 7758,
      "time_per_iteration": 2.4970245361328125
    },
    {
      "auxiliary_loss_clip": 0.01157731,
      "auxiliary_loss_mlp": 0.01023386,
      "balance_loss_clip": 1.04786968,
      "balance_loss_mlp": 1.01630521,
      "epoch": 0.9329645884687068,
      "flos": 30081722469120.0,
      "grad_norm": 2.9284981467544413,
      "language_loss": 0.90691423,
      "learning_rate": 4.688022741881559e-08,
      "loss": 0.92872536,
      "num_input_tokens_seen": 167750440,
      "step": 7759,
      "time_per_iteration": 2.4959499835968018
    },
    {
      "auxiliary_loss_clip": 0.01147888,
      "auxiliary_loss_mlp": 0.01022146,
      "balance_loss_clip": 1.04418719,
      "balance_loss_mlp": 1.01595926,
      "epoch": 0.9330848313593458,
      "flos": 21867982513920.0,
      "grad_norm": 1.5600949313376908,
      "language_loss": 0.75029552,
      "learning_rate": 4.671270512936076e-08,
      "loss": 0.77199578,
      "num_input_tokens_seen": 167769600,
      "step": 7760,
      "time_per_iteration": 2.4216272830963135
    },
    {
      "auxiliary_loss_clip": 0.01115762,
      "auxiliary_loss_mlp": 0.01021622,
      "balance_loss_clip": 1.04108334,
      "balance_loss_mlp": 1.01483846,
      "epoch": 0.933205074249985,
      "flos": 22127221946880.0,
      "grad_norm": 1.6774029024005355,
      "language_loss": 0.82507986,
      "learning_rate": 4.6545479152035884e-08,
      "loss": 0.84645379,
      "num_input_tokens_seen": 167788770,
      "step": 7761,
      "time_per_iteration": 2.50838565826416
    },
    {
      "auxiliary_loss_clip": 0.01150832,
      "auxiliary_loss_mlp": 0.01020741,
      "balance_loss_clip": 1.0461936,
      "balance_loss_mlp": 1.01398754,
      "epoch": 0.9333253171406241,
      "flos": 15341254675200.0,
      "grad_norm": 1.902626518266492,
      "language_loss": 0.7608161,
      "learning_rate": 4.637854951220821e-08,
      "loss": 0.78253186,
      "num_input_tokens_seen": 167805555,
      "step": 7762,
      "time_per_iteration": 2.3953194618225098
    },
    {
      "auxiliary_loss_clip": 0.01118119,
      "auxiliary_loss_mlp": 0.01021795,
      "balance_loss_clip": 1.04127622,
      "balance_loss_mlp": 1.01502419,
      "epoch": 0.9334455600312631,
      "flos": 15706142985600.0,
      "grad_norm": 2.2045194435547866,
      "language_loss": 0.74879515,
      "learning_rate": 4.621191623520171e-08,
      "loss": 0.77019429,
      "num_input_tokens_seen": 167823985,
      "step": 7763,
      "time_per_iteration": 2.4767022132873535
    },
    {
      "auxiliary_loss_clip": 0.01109057,
      "auxiliary_loss_mlp": 0.01025436,
      "balance_loss_clip": 1.04244483,
      "balance_loss_mlp": 1.01818228,
      "epoch": 0.9335658029219023,
      "flos": 22163563532160.0,
      "grad_norm": 2.4869808445278814,
      "language_loss": 0.84675682,
      "learning_rate": 4.604557934629372e-08,
      "loss": 0.86810172,
      "num_input_tokens_seen": 167843060,
      "step": 7764,
      "time_per_iteration": 2.5723752975463867
    },
    {
      "auxiliary_loss_clip": 0.01133255,
      "auxiliary_loss_mlp": 0.01021582,
      "balance_loss_clip": 1.04558575,
      "balance_loss_mlp": 1.01515388,
      "epoch": 0.9336860458125413,
      "flos": 20266833859200.0,
      "grad_norm": 1.8042832881352566,
      "language_loss": 0.80361527,
      "learning_rate": 4.587953887071805e-08,
      "loss": 0.82516366,
      "num_input_tokens_seen": 167862880,
      "step": 7765,
      "time_per_iteration": 2.459326982498169
    },
    {
      "auxiliary_loss_clip": 0.01133867,
      "auxiliary_loss_mlp": 0.01026673,
      "balance_loss_clip": 1.04234624,
      "balance_loss_mlp": 1.019485,
      "epoch": 0.9338062887031804,
      "flos": 20919689504640.0,
      "grad_norm": 15.34083718671415,
      "language_loss": 0.85835814,
      "learning_rate": 4.5713794833662554e-08,
      "loss": 0.87996364,
      "num_input_tokens_seen": 167882095,
      "step": 7766,
      "time_per_iteration": 2.469886064529419
    },
    {
      "auxiliary_loss_clip": 0.01165673,
      "auxiliary_loss_mlp": 0.01027064,
      "balance_loss_clip": 1.04746377,
      "balance_loss_mlp": 1.01961958,
      "epoch": 0.9339265315938196,
      "flos": 23221635482880.0,
      "grad_norm": 1.8208497749642827,
      "language_loss": 0.63362467,
      "learning_rate": 4.5548347260270236e-08,
      "loss": 0.65555209,
      "num_input_tokens_seen": 167901385,
      "step": 7767,
      "time_per_iteration": 2.42073392868042
    },
    {
      "auxiliary_loss_clip": 0.01119422,
      "auxiliary_loss_mlp": 0.01021985,
      "balance_loss_clip": 1.04212189,
      "balance_loss_mlp": 1.01518703,
      "epoch": 0.9340467744844586,
      "flos": 22820261932800.0,
      "grad_norm": 1.7006399955007752,
      "language_loss": 0.69158447,
      "learning_rate": 4.538319617564012e-08,
      "loss": 0.71299851,
      "num_input_tokens_seen": 167920405,
      "step": 7768,
      "time_per_iteration": 2.503188133239746
    },
    {
      "auxiliary_loss_clip": 0.01132618,
      "auxiliary_loss_mlp": 0.01022276,
      "balance_loss_clip": 1.04027438,
      "balance_loss_mlp": 1.01530838,
      "epoch": 0.9341670173750977,
      "flos": 23660428026240.0,
      "grad_norm": 1.8903753494492697,
      "language_loss": 0.7408365,
      "learning_rate": 4.521834160482485e-08,
      "loss": 0.76238549,
      "num_input_tokens_seen": 167939145,
      "step": 7769,
      "time_per_iteration": 2.5169825553894043
    },
    {
      "auxiliary_loss_clip": 0.01152735,
      "auxiliary_loss_mlp": 0.01026104,
      "balance_loss_clip": 1.0453434,
      "balance_loss_mlp": 1.0189997,
      "epoch": 0.9342872602657368,
      "flos": 24824256595200.0,
      "grad_norm": 1.5718700654669397,
      "language_loss": 0.81908619,
      "learning_rate": 4.5053783572832846e-08,
      "loss": 0.84087461,
      "num_input_tokens_seen": 167959325,
      "step": 7770,
      "time_per_iteration": 2.4633185863494873
    },
    {
      "auxiliary_loss_clip": 0.01152006,
      "auxiliary_loss_mlp": 0.01024606,
      "balance_loss_clip": 1.04681134,
      "balance_loss_mlp": 1.01768017,
      "epoch": 0.9344075031563759,
      "flos": 25771831332480.0,
      "grad_norm": 1.666095821348059,
      "language_loss": 0.76537269,
      "learning_rate": 4.488952210462771e-08,
      "loss": 0.78713882,
      "num_input_tokens_seen": 167979530,
      "step": 7771,
      "time_per_iteration": 2.487454414367676
    },
    {
      "auxiliary_loss_clip": 0.01162954,
      "auxiliary_loss_mlp": 0.01021831,
      "balance_loss_clip": 1.04672909,
      "balance_loss_mlp": 1.01490474,
      "epoch": 0.9345277460470149,
      "flos": 25551303782400.0,
      "grad_norm": 2.3990158235627224,
      "language_loss": 0.8547321,
      "learning_rate": 4.4725557225127495e-08,
      "loss": 0.87658,
      "num_input_tokens_seen": 167997870,
      "step": 7772,
      "time_per_iteration": 2.4244980812072754
    },
    {
      "auxiliary_loss_clip": 0.01152041,
      "auxiliary_loss_mlp": 0.01026955,
      "balance_loss_clip": 1.0475527,
      "balance_loss_mlp": 1.02057409,
      "epoch": 0.9346479889376541,
      "flos": 34313112432000.0,
      "grad_norm": 1.8285645013147327,
      "language_loss": 0.79216981,
      "learning_rate": 4.456188895920565e-08,
      "loss": 0.81395984,
      "num_input_tokens_seen": 168019625,
      "step": 7773,
      "time_per_iteration": 2.5352118015289307
    },
    {
      "auxiliary_loss_clip": 0.011659,
      "auxiliary_loss_mlp": 0.01024555,
      "balance_loss_clip": 1.0481379,
      "balance_loss_mlp": 1.01721811,
      "epoch": 0.9347682318282932,
      "flos": 19093739581440.0,
      "grad_norm": 1.969581702050408,
      "language_loss": 0.85555845,
      "learning_rate": 4.439851733169031e-08,
      "loss": 0.87746298,
      "num_input_tokens_seen": 168037415,
      "step": 7774,
      "time_per_iteration": 3.2043185234069824
    },
    {
      "auxiliary_loss_clip": 0.01123122,
      "auxiliary_loss_mlp": 0.01029402,
      "balance_loss_clip": 1.04230046,
      "balance_loss_mlp": 1.02273536,
      "epoch": 0.9348884747189322,
      "flos": 26249587153920.0,
      "grad_norm": 2.5685939524587074,
      "language_loss": 0.69176096,
      "learning_rate": 4.4235442367365204e-08,
      "loss": 0.71328622,
      "num_input_tokens_seen": 168057725,
      "step": 7775,
      "time_per_iteration": 3.276679039001465
    },
    {
      "auxiliary_loss_clip": 0.01131988,
      "auxiliary_loss_mlp": 0.01023976,
      "balance_loss_clip": 1.04113317,
      "balance_loss_mlp": 1.01660895,
      "epoch": 0.9350087176095714,
      "flos": 18333080242560.0,
      "grad_norm": 2.3176809293446397,
      "language_loss": 0.79683447,
      "learning_rate": 4.4072664090968545e-08,
      "loss": 0.81839418,
      "num_input_tokens_seen": 168076110,
      "step": 7776,
      "time_per_iteration": 3.250821590423584
    },
    {
      "auxiliary_loss_clip": 0.01136684,
      "auxiliary_loss_mlp": 0.01025387,
      "balance_loss_clip": 1.04198039,
      "balance_loss_mlp": 1.01855874,
      "epoch": 0.9351289605002104,
      "flos": 19318253541120.0,
      "grad_norm": 1.8559219383452683,
      "language_loss": 0.84946561,
      "learning_rate": 4.391018252719347e-08,
      "loss": 0.8710863,
      "num_input_tokens_seen": 168095905,
      "step": 7777,
      "time_per_iteration": 2.4717464447021484
    },
    {
      "auxiliary_loss_clip": 0.01137661,
      "auxiliary_loss_mlp": 0.01029292,
      "balance_loss_clip": 1.0426228,
      "balance_loss_mlp": 1.02189183,
      "epoch": 0.9352492033908495,
      "flos": 18799990156800.0,
      "grad_norm": 1.9409895288930221,
      "language_loss": 0.6928851,
      "learning_rate": 4.374799770068849e-08,
      "loss": 0.71455467,
      "num_input_tokens_seen": 168112580,
      "step": 7778,
      "time_per_iteration": 2.4392635822296143
    },
    {
      "auxiliary_loss_clip": 0.01148175,
      "auxiliary_loss_mlp": 0.01023359,
      "balance_loss_clip": 1.04526949,
      "balance_loss_mlp": 1.01598263,
      "epoch": 0.9353694462814887,
      "flos": 29530134241920.0,
      "grad_norm": 1.9265869928941481,
      "language_loss": 0.74665433,
      "learning_rate": 4.358610963605658e-08,
      "loss": 0.76836967,
      "num_input_tokens_seen": 168133030,
      "step": 7779,
      "time_per_iteration": 2.4740631580352783
    },
    {
      "auxiliary_loss_clip": 0.01168087,
      "auxiliary_loss_mlp": 0.01033546,
      "balance_loss_clip": 1.04889739,
      "balance_loss_mlp": 1.0265367,
      "epoch": 0.9354896891721277,
      "flos": 30665450390400.0,
      "grad_norm": 2.273553290300527,
      "language_loss": 0.68731302,
      "learning_rate": 4.342451835785677e-08,
      "loss": 0.70932937,
      "num_input_tokens_seen": 168153940,
      "step": 7780,
      "time_per_iteration": 2.4637398719787598
    },
    {
      "auxiliary_loss_clip": 0.01134468,
      "auxiliary_loss_mlp": 0.01022129,
      "balance_loss_clip": 1.04410577,
      "balance_loss_mlp": 1.01531363,
      "epoch": 0.9356099320627668,
      "flos": 19463907191040.0,
      "grad_norm": 1.5673099519592728,
      "language_loss": 0.74924242,
      "learning_rate": 4.3263223890601665e-08,
      "loss": 0.7708084,
      "num_input_tokens_seen": 168172650,
      "step": 7781,
      "time_per_iteration": 2.440971612930298
    },
    {
      "auxiliary_loss_clip": 0.0114648,
      "auxiliary_loss_mlp": 0.00761531,
      "balance_loss_clip": 1.04738081,
      "balance_loss_mlp": 1.00047398,
      "epoch": 0.9357301749534058,
      "flos": 19098156954240.0,
      "grad_norm": 1.935911680435023,
      "language_loss": 0.79572177,
      "learning_rate": 4.31022262587597e-08,
      "loss": 0.81480187,
      "num_input_tokens_seen": 168191325,
      "step": 7782,
      "time_per_iteration": 2.4253463745117188
    },
    {
      "auxiliary_loss_clip": 0.01150224,
      "auxiliary_loss_mlp": 0.01028048,
      "balance_loss_clip": 1.04578006,
      "balance_loss_mlp": 1.02001333,
      "epoch": 0.935850417844045,
      "flos": 23550361776000.0,
      "grad_norm": 1.580729245362894,
      "language_loss": 0.6575495,
      "learning_rate": 4.2941525486754225e-08,
      "loss": 0.67933214,
      "num_input_tokens_seen": 168211645,
      "step": 7783,
      "time_per_iteration": 3.1668505668640137
    },
    {
      "auxiliary_loss_clip": 0.01116851,
      "auxiliary_loss_mlp": 0.01022312,
      "balance_loss_clip": 1.0418303,
      "balance_loss_mlp": 1.01605356,
      "epoch": 0.935970660734684,
      "flos": 18588333265920.0,
      "grad_norm": 1.7964190534449191,
      "language_loss": 0.7951709,
      "learning_rate": 4.278112159896286e-08,
      "loss": 0.81656253,
      "num_input_tokens_seen": 168229485,
      "step": 7784,
      "time_per_iteration": 2.4852676391601562
    },
    {
      "auxiliary_loss_clip": 0.01128844,
      "auxiliary_loss_mlp": 0.01021092,
      "balance_loss_clip": 1.0398128,
      "balance_loss_mlp": 1.0146699,
      "epoch": 0.9360909036253231,
      "flos": 20631255292800.0,
      "grad_norm": 1.6579822404683744,
      "language_loss": 0.67743784,
      "learning_rate": 4.2621014619719896e-08,
      "loss": 0.69893718,
      "num_input_tokens_seen": 168247250,
      "step": 7785,
      "time_per_iteration": 2.4690158367156982
    },
    {
      "auxiliary_loss_clip": 0.01038029,
      "auxiliary_loss_mlp": 0.01001393,
      "balance_loss_clip": 1.0067811,
      "balance_loss_mlp": 1.00048733,
      "epoch": 0.9362111465159623,
      "flos": 61791421052160.0,
      "grad_norm": 0.7231168886449887,
      "language_loss": 0.58630222,
      "learning_rate": 4.246120457331215e-08,
      "loss": 0.60669643,
      "num_input_tokens_seen": 168309425,
      "step": 7786,
      "time_per_iteration": 3.085767984390259
    },
    {
      "auxiliary_loss_clip": 0.01132813,
      "auxiliary_loss_mlp": 0.01024115,
      "balance_loss_clip": 1.04543233,
      "balance_loss_mlp": 1.01677513,
      "epoch": 0.9363313894066013,
      "flos": 24170395368960.0,
      "grad_norm": 2.7735232215143557,
      "language_loss": 0.72310555,
      "learning_rate": 4.2301691483983325e-08,
      "loss": 0.74467486,
      "num_input_tokens_seen": 168329545,
      "step": 7787,
      "time_per_iteration": 2.5015017986297607
    },
    {
      "auxiliary_loss_clip": 0.01152399,
      "auxiliary_loss_mlp": 0.01022984,
      "balance_loss_clip": 1.04464877,
      "balance_loss_mlp": 1.01555109,
      "epoch": 0.9364516322972404,
      "flos": 20120354196480.0,
      "grad_norm": 2.5073143857702576,
      "language_loss": 0.75994349,
      "learning_rate": 4.214247537593163e-08,
      "loss": 0.78169727,
      "num_input_tokens_seen": 168348795,
      "step": 7788,
      "time_per_iteration": 2.435617208480835
    },
    {
      "auxiliary_loss_clip": 0.01136672,
      "auxiliary_loss_mlp": 0.01033454,
      "balance_loss_clip": 1.04271424,
      "balance_loss_mlp": 1.02665269,
      "epoch": 0.9365718751878795,
      "flos": 20703758895360.0,
      "grad_norm": 1.8076671171808698,
      "language_loss": 0.80504346,
      "learning_rate": 4.1983556273309293e-08,
      "loss": 0.82674468,
      "num_input_tokens_seen": 168367545,
      "step": 7789,
      "time_per_iteration": 2.4675133228302
    },
    {
      "auxiliary_loss_clip": 0.01167386,
      "auxiliary_loss_mlp": 0.01028656,
      "balance_loss_clip": 1.04741704,
      "balance_loss_mlp": 1.02112162,
      "epoch": 0.9366921180785186,
      "flos": 18655270260480.0,
      "grad_norm": 3.11434642827236,
      "language_loss": 0.6876986,
      "learning_rate": 4.182493420022526e-08,
      "loss": 0.70965904,
      "num_input_tokens_seen": 168383215,
      "step": 7790,
      "time_per_iteration": 2.3460826873779297
    },
    {
      "auxiliary_loss_clip": 0.01126054,
      "auxiliary_loss_mlp": 0.01024757,
      "balance_loss_clip": 1.04309714,
      "balance_loss_mlp": 1.01791477,
      "epoch": 0.9368123609691577,
      "flos": 25774955815680.0,
      "grad_norm": 1.7006037433620487,
      "language_loss": 0.78611624,
      "learning_rate": 4.166660918074139e-08,
      "loss": 0.8076244,
      "num_input_tokens_seen": 168403120,
      "step": 7791,
      "time_per_iteration": 2.5708913803100586
    },
    {
      "auxiliary_loss_clip": 0.01118965,
      "auxiliary_loss_mlp": 0.01024519,
      "balance_loss_clip": 1.04160142,
      "balance_loss_mlp": 1.01745915,
      "epoch": 0.9369326038597968,
      "flos": 25553386771200.0,
      "grad_norm": 1.4620012960436688,
      "language_loss": 0.73728013,
      "learning_rate": 4.15085812388758e-08,
      "loss": 0.75871497,
      "num_input_tokens_seen": 168425340,
      "step": 7792,
      "time_per_iteration": 2.5659966468811035
    },
    {
      "auxiliary_loss_clip": 0.0113498,
      "auxiliary_loss_mlp": 0.01025806,
      "balance_loss_clip": 1.04341435,
      "balance_loss_mlp": 1.01883793,
      "epoch": 0.9370528467504359,
      "flos": 23220019370880.0,
      "grad_norm": 1.67810545195276,
      "language_loss": 0.78653908,
      "learning_rate": 4.135085039860153e-08,
      "loss": 0.80814695,
      "num_input_tokens_seen": 168444740,
      "step": 7793,
      "time_per_iteration": 2.4897918701171875
    },
    {
      "auxiliary_loss_clip": 0.01138718,
      "auxiliary_loss_mlp": 0.0102116,
      "balance_loss_clip": 1.04795122,
      "balance_loss_mlp": 1.01391828,
      "epoch": 0.9371730896410749,
      "flos": 24967468120320.0,
      "grad_norm": 2.3897843632486113,
      "language_loss": 0.78454274,
      "learning_rate": 4.1193416683845906e-08,
      "loss": 0.8061415,
      "num_input_tokens_seen": 168463670,
      "step": 7794,
      "time_per_iteration": 2.4905483722686768
    },
    {
      "auxiliary_loss_clip": 0.01126918,
      "auxiliary_loss_mlp": 0.01025202,
      "balance_loss_clip": 1.04585826,
      "balance_loss_mlp": 1.01877379,
      "epoch": 0.9372933325317141,
      "flos": 15553091134080.0,
      "grad_norm": 2.325795915003717,
      "language_loss": 0.83366895,
      "learning_rate": 4.103628011849136e-08,
      "loss": 0.85519016,
      "num_input_tokens_seen": 168479030,
      "step": 7795,
      "time_per_iteration": 2.4702987670898438
    },
    {
      "auxiliary_loss_clip": 0.01139317,
      "auxiliary_loss_mlp": 0.01023624,
      "balance_loss_clip": 1.04500806,
      "balance_loss_mlp": 1.01660562,
      "epoch": 0.9374135754223532,
      "flos": 21871861182720.0,
      "grad_norm": 1.9432192115945677,
      "language_loss": 0.76117641,
      "learning_rate": 4.0879440726375506e-08,
      "loss": 0.7828058,
      "num_input_tokens_seen": 168496815,
      "step": 7796,
      "time_per_iteration": 2.4747061729431152
    },
    {
      "auxiliary_loss_clip": 0.01133299,
      "auxiliary_loss_mlp": 0.01020199,
      "balance_loss_clip": 1.04025006,
      "balance_loss_mlp": 1.01305223,
      "epoch": 0.9375338183129922,
      "flos": 22631048064000.0,
      "grad_norm": 2.74369260518212,
      "language_loss": 0.56226516,
      "learning_rate": 4.0722898531291074e-08,
      "loss": 0.58380014,
      "num_input_tokens_seen": 168514055,
      "step": 7797,
      "time_per_iteration": 2.475250244140625
    },
    {
      "auxiliary_loss_clip": 0.01142699,
      "auxiliary_loss_mlp": 0.01023708,
      "balance_loss_clip": 1.0441041,
      "balance_loss_mlp": 1.0164547,
      "epoch": 0.9376540612036314,
      "flos": 26104292640000.0,
      "grad_norm": 1.942390645137417,
      "language_loss": 0.766204,
      "learning_rate": 4.0566653556985295e-08,
      "loss": 0.78786808,
      "num_input_tokens_seen": 168534600,
      "step": 7798,
      "time_per_iteration": 2.516303777694702
    },
    {
      "auxiliary_loss_clip": 0.01083056,
      "auxiliary_loss_mlp": 0.01030433,
      "balance_loss_clip": 1.03850102,
      "balance_loss_mlp": 1.02238905,
      "epoch": 0.9377743040942704,
      "flos": 19717580016000.0,
      "grad_norm": 2.6680341454378937,
      "language_loss": 0.81443667,
      "learning_rate": 4.0410705827159886e-08,
      "loss": 0.83557159,
      "num_input_tokens_seen": 168551895,
      "step": 7799,
      "time_per_iteration": 2.5737733840942383
    },
    {
      "auxiliary_loss_clip": 0.01133304,
      "auxiliary_loss_mlp": 0.01024367,
      "balance_loss_clip": 1.04146504,
      "balance_loss_mlp": 1.01711297,
      "epoch": 0.9378945469849095,
      "flos": 15267530010240.0,
      "grad_norm": 1.9570826450293635,
      "language_loss": 0.7145617,
      "learning_rate": 4.0255055365472356e-08,
      "loss": 0.7361384,
      "num_input_tokens_seen": 168569990,
      "step": 7800,
      "time_per_iteration": 2.4544239044189453
    },
    {
      "auxiliary_loss_clip": 0.01095665,
      "auxiliary_loss_mlp": 0.01027194,
      "balance_loss_clip": 1.03767526,
      "balance_loss_mlp": 1.02018118,
      "epoch": 0.9380147898755486,
      "flos": 20591394174720.0,
      "grad_norm": 2.9220867427633412,
      "language_loss": 0.74910402,
      "learning_rate": 4.009970219553471e-08,
      "loss": 0.77033257,
      "num_input_tokens_seen": 168586940,
      "step": 7801,
      "time_per_iteration": 3.3849499225616455
    },
    {
      "auxiliary_loss_clip": 0.01154351,
      "auxiliary_loss_mlp": 0.01024841,
      "balance_loss_clip": 1.04489481,
      "balance_loss_mlp": 1.01719368,
      "epoch": 0.9381350327661877,
      "flos": 26281116316800.0,
      "grad_norm": 2.7744341475645493,
      "language_loss": 0.77005196,
      "learning_rate": 3.99446463409141e-08,
      "loss": 0.79184389,
      "num_input_tokens_seen": 168604795,
      "step": 7802,
      "time_per_iteration": 3.2243008613586426
    },
    {
      "auxiliary_loss_clip": 0.01154225,
      "auxiliary_loss_mlp": 0.01024454,
      "balance_loss_clip": 1.04296434,
      "balance_loss_mlp": 1.01724207,
      "epoch": 0.9382552756568268,
      "flos": 23586344225280.0,
      "grad_norm": 2.002154923716347,
      "language_loss": 0.687397,
      "learning_rate": 3.978988782513215e-08,
      "loss": 0.70918369,
      "num_input_tokens_seen": 168622290,
      "step": 7803,
      "time_per_iteration": 3.318350315093994
    },
    {
      "auxiliary_loss_clip": 0.01153139,
      "auxiliary_loss_mlp": 0.01020312,
      "balance_loss_clip": 1.04404259,
      "balance_loss_mlp": 1.01345742,
      "epoch": 0.9383755185474659,
      "flos": 28438809275520.0,
      "grad_norm": 1.7941594912290635,
      "language_loss": 0.76412523,
      "learning_rate": 3.963542667166586e-08,
      "loss": 0.7858597,
      "num_input_tokens_seen": 168642395,
      "step": 7804,
      "time_per_iteration": 2.520297050476074
    },
    {
      "auxiliary_loss_clip": 0.01128937,
      "auxiliary_loss_mlp": 0.01025573,
      "balance_loss_clip": 1.04824924,
      "balance_loss_mlp": 1.0187602,
      "epoch": 0.938495761438105,
      "flos": 20449583280000.0,
      "grad_norm": 1.6813638873015362,
      "language_loss": 0.68237638,
      "learning_rate": 3.9481262903946486e-08,
      "loss": 0.70392156,
      "num_input_tokens_seen": 168661840,
      "step": 7805,
      "time_per_iteration": 2.534358501434326
    },
    {
      "auxiliary_loss_clip": 0.01024006,
      "auxiliary_loss_mlp": 0.01001157,
      "balance_loss_clip": 1.00830948,
      "balance_loss_mlp": 1.00019741,
      "epoch": 0.938616004328744,
      "flos": 69302711658240.0,
      "grad_norm": 0.7719606455094693,
      "language_loss": 0.54471767,
      "learning_rate": 3.932739654536066e-08,
      "loss": 0.5649693,
      "num_input_tokens_seen": 168724540,
      "step": 7806,
      "time_per_iteration": 3.108773946762085
    },
    {
      "auxiliary_loss_clip": 0.01150496,
      "auxiliary_loss_mlp": 0.0102358,
      "balance_loss_clip": 1.04647648,
      "balance_loss_mlp": 1.01726782,
      "epoch": 0.9387362472193832,
      "flos": 18911636605440.0,
      "grad_norm": 2.13495686068401,
      "language_loss": 0.74022591,
      "learning_rate": 3.917382761925014e-08,
      "loss": 0.76196665,
      "num_input_tokens_seen": 168740375,
      "step": 7807,
      "time_per_iteration": 2.4531922340393066
    },
    {
      "auxiliary_loss_clip": 0.01147488,
      "auxiliary_loss_mlp": 0.01026001,
      "balance_loss_clip": 1.04545498,
      "balance_loss_mlp": 1.01929867,
      "epoch": 0.9388564901100223,
      "flos": 26501967089280.0,
      "grad_norm": 1.7085085274545841,
      "language_loss": 0.79436463,
      "learning_rate": 3.9020556148910754e-08,
      "loss": 0.81609952,
      "num_input_tokens_seen": 168759730,
      "step": 7808,
      "time_per_iteration": 2.5222485065460205
    },
    {
      "auxiliary_loss_clip": 0.01044044,
      "auxiliary_loss_mlp": 0.01000411,
      "balance_loss_clip": 1.00848341,
      "balance_loss_mlp": 0.99955845,
      "epoch": 0.9389767330006613,
      "flos": 58941083157120.0,
      "grad_norm": 0.7353404238872535,
      "language_loss": 0.5674243,
      "learning_rate": 3.8867582157593895e-08,
      "loss": 0.58786893,
      "num_input_tokens_seen": 168813935,
      "step": 7809,
      "time_per_iteration": 2.924941301345825
    },
    {
      "auxiliary_loss_clip": 0.01151369,
      "auxiliary_loss_mlp": 0.01021866,
      "balance_loss_clip": 1.04829741,
      "balance_loss_mlp": 1.0151453,
      "epoch": 0.9390969758913005,
      "flos": 31102554994560.0,
      "grad_norm": 2.5688877658802576,
      "language_loss": 0.76330429,
      "learning_rate": 3.871490566850544e-08,
      "loss": 0.78503668,
      "num_input_tokens_seen": 168838145,
      "step": 7810,
      "time_per_iteration": 3.3108325004577637
    },
    {
      "auxiliary_loss_clip": 0.01133487,
      "auxiliary_loss_mlp": 0.0102335,
      "balance_loss_clip": 1.04414129,
      "balance_loss_mlp": 1.01631927,
      "epoch": 0.9392172187819395,
      "flos": 22419391173120.0,
      "grad_norm": 1.8026362078527254,
      "language_loss": 0.70803982,
      "learning_rate": 3.856252670480642e-08,
      "loss": 0.72960818,
      "num_input_tokens_seen": 168856805,
      "step": 7811,
      "time_per_iteration": 2.4905502796173096
    },
    {
      "auxiliary_loss_clip": 0.01133843,
      "auxiliary_loss_mlp": 0.01026459,
      "balance_loss_clip": 1.0414331,
      "balance_loss_mlp": 1.01903772,
      "epoch": 0.9393374616725786,
      "flos": 19719483436800.0,
      "grad_norm": 1.8284271896453064,
      "language_loss": 0.81367469,
      "learning_rate": 3.841044528961279e-08,
      "loss": 0.83527768,
      "num_input_tokens_seen": 168874600,
      "step": 7812,
      "time_per_iteration": 2.485079050064087
    },
    {
      "auxiliary_loss_clip": 0.01164374,
      "auxiliary_loss_mlp": 0.01022322,
      "balance_loss_clip": 1.04488468,
      "balance_loss_mlp": 1.01506829,
      "epoch": 0.9394577045632178,
      "flos": 24170215800960.0,
      "grad_norm": 1.9090386087029503,
      "language_loss": 0.78516585,
      "learning_rate": 3.825866144599477e-08,
      "loss": 0.80703282,
      "num_input_tokens_seen": 168893655,
      "step": 7813,
      "time_per_iteration": 2.4292402267456055
    },
    {
      "auxiliary_loss_clip": 0.01136229,
      "auxiliary_loss_mlp": 0.01021705,
      "balance_loss_clip": 1.04269695,
      "balance_loss_mlp": 1.01446617,
      "epoch": 0.9395779474538568,
      "flos": 19023929498880.0,
      "grad_norm": 1.9382582998029014,
      "language_loss": 0.75169373,
      "learning_rate": 3.8107175196978145e-08,
      "loss": 0.77327305,
      "num_input_tokens_seen": 168909960,
      "step": 7814,
      "time_per_iteration": 2.447415590286255
    },
    {
      "auxiliary_loss_clip": 0.01121879,
      "auxiliary_loss_mlp": 0.01026275,
      "balance_loss_clip": 1.04426503,
      "balance_loss_mlp": 1.01953089,
      "epoch": 0.9396981903444959,
      "flos": 14319129260160.0,
      "grad_norm": 4.050239831428167,
      "language_loss": 0.76650649,
      "learning_rate": 3.7955986565542996e-08,
      "loss": 0.78798807,
      "num_input_tokens_seen": 168928040,
      "step": 7815,
      "time_per_iteration": 2.4909706115722656
    },
    {
      "auxiliary_loss_clip": 0.01122221,
      "auxiliary_loss_mlp": 0.01029132,
      "balance_loss_clip": 1.04187632,
      "balance_loss_mlp": 1.02242899,
      "epoch": 0.9398184332351349,
      "flos": 34787564202240.0,
      "grad_norm": 1.8837717056348675,
      "language_loss": 0.68207049,
      "learning_rate": 3.780509557462497e-08,
      "loss": 0.70358396,
      "num_input_tokens_seen": 168948240,
      "step": 7816,
      "time_per_iteration": 2.6253855228424072
    },
    {
      "auxiliary_loss_clip": 0.01132494,
      "auxiliary_loss_mlp": 0.01022974,
      "balance_loss_clip": 1.04111564,
      "balance_loss_mlp": 1.01553535,
      "epoch": 0.9399386761257741,
      "flos": 25372253462400.0,
      "grad_norm": 1.5061363765502904,
      "language_loss": 0.75301248,
      "learning_rate": 3.765450224711375e-08,
      "loss": 0.77456713,
      "num_input_tokens_seen": 168968745,
      "step": 7817,
      "time_per_iteration": 2.5172901153564453
    },
    {
      "auxiliary_loss_clip": 0.0113344,
      "auxiliary_loss_mlp": 0.01023986,
      "balance_loss_clip": 1.04583824,
      "balance_loss_mlp": 1.01706004,
      "epoch": 0.9400589190164131,
      "flos": 27304965584640.0,
      "grad_norm": 1.7555954865867462,
      "language_loss": 0.79811132,
      "learning_rate": 3.750420660585396e-08,
      "loss": 0.81968558,
      "num_input_tokens_seen": 168990685,
      "step": 7818,
      "time_per_iteration": 2.539062261581421
    },
    {
      "auxiliary_loss_clip": 0.01164381,
      "auxiliary_loss_mlp": 0.0102412,
      "balance_loss_clip": 1.04792666,
      "balance_loss_mlp": 1.01726007,
      "epoch": 0.9401791619070522,
      "flos": 23399859790080.0,
      "grad_norm": 2.0165620934241266,
      "language_loss": 0.7980516,
      "learning_rate": 3.735420867364603e-08,
      "loss": 0.81993663,
      "num_input_tokens_seen": 169011665,
      "step": 7819,
      "time_per_iteration": 2.435166835784912
    },
    {
      "auxiliary_loss_clip": 0.01085759,
      "auxiliary_loss_mlp": 0.01020414,
      "balance_loss_clip": 1.03537977,
      "balance_loss_mlp": 1.01383698,
      "epoch": 0.9402994047976914,
      "flos": 35881403120640.0,
      "grad_norm": 1.6147118527315714,
      "language_loss": 0.6174134,
      "learning_rate": 3.7204508473244186e-08,
      "loss": 0.63847518,
      "num_input_tokens_seen": 169035290,
      "step": 7820,
      "time_per_iteration": 2.6953890323638916
    },
    {
      "auxiliary_loss_clip": 0.01075353,
      "auxiliary_loss_mlp": 0.01022725,
      "balance_loss_clip": 1.0386436,
      "balance_loss_mlp": 1.01636803,
      "epoch": 0.9404196476883304,
      "flos": 22236821320320.0,
      "grad_norm": 3.2616484913639825,
      "language_loss": 0.69448584,
      "learning_rate": 3.7055106027357395e-08,
      "loss": 0.71546662,
      "num_input_tokens_seen": 169055155,
      "step": 7821,
      "time_per_iteration": 2.607365608215332
    },
    {
      "auxiliary_loss_clip": 0.011494,
      "auxiliary_loss_mlp": 0.01022732,
      "balance_loss_clip": 1.04647636,
      "balance_loss_mlp": 1.015275,
      "epoch": 0.9405398905789695,
      "flos": 18915802583040.0,
      "grad_norm": 1.9504624223105993,
      "language_loss": 0.71733874,
      "learning_rate": 3.690600135865063e-08,
      "loss": 0.73906004,
      "num_input_tokens_seen": 169072080,
      "step": 7822,
      "time_per_iteration": 2.414651870727539
    },
    {
      "auxiliary_loss_clip": 0.01020244,
      "auxiliary_loss_mlp": 0.01001222,
      "balance_loss_clip": 1.00700426,
      "balance_loss_mlp": 1.00032222,
      "epoch": 0.9406601334696086,
      "flos": 70274130048000.0,
      "grad_norm": 0.7955488122472639,
      "language_loss": 0.58150923,
      "learning_rate": 3.675719448974246e-08,
      "loss": 0.60172391,
      "num_input_tokens_seen": 169137170,
      "step": 7823,
      "time_per_iteration": 3.236582040786743
    },
    {
      "auxiliary_loss_clip": 0.01106744,
      "auxiliary_loss_mlp": 0.00761564,
      "balance_loss_clip": 1.04196453,
      "balance_loss_mlp": 1.00043607,
      "epoch": 0.9407803763602477,
      "flos": 22165071903360.0,
      "grad_norm": 1.9763820737392357,
      "language_loss": 0.60037947,
      "learning_rate": 3.6608685443207054e-08,
      "loss": 0.6190626,
      "num_input_tokens_seen": 169156320,
      "step": 7824,
      "time_per_iteration": 2.839423418045044
    },
    {
      "auxiliary_loss_clip": 0.01125414,
      "auxiliary_loss_mlp": 0.0102409,
      "balance_loss_clip": 1.04265451,
      "balance_loss_mlp": 1.01742303,
      "epoch": 0.9409006192508867,
      "flos": 18879496911360.0,
      "grad_norm": 2.4829460325040564,
      "language_loss": 0.66655159,
      "learning_rate": 3.646047424157306e-08,
      "loss": 0.68804657,
      "num_input_tokens_seen": 169173295,
      "step": 7825,
      "time_per_iteration": 2.4948480129241943
    },
    {
      "auxiliary_loss_clip": 0.01137966,
      "auxiliary_loss_mlp": 0.01026241,
      "balance_loss_clip": 1.04580545,
      "balance_loss_mlp": 1.01870084,
      "epoch": 0.9410208621415259,
      "flos": 23368258800000.0,
      "grad_norm": 2.5389447741234665,
      "language_loss": 0.68397832,
      "learning_rate": 3.631256090732382e-08,
      "loss": 0.70562041,
      "num_input_tokens_seen": 169193755,
      "step": 7826,
      "time_per_iteration": 2.4857218265533447
    },
    {
      "auxiliary_loss_clip": 0.01124,
      "auxiliary_loss_mlp": 0.01024014,
      "balance_loss_clip": 1.04449153,
      "balance_loss_mlp": 1.01751721,
      "epoch": 0.941141105032165,
      "flos": 22742227635840.0,
      "grad_norm": 2.0194203161119013,
      "language_loss": 0.82629824,
      "learning_rate": 3.6164945462897833e-08,
      "loss": 0.84777832,
      "num_input_tokens_seen": 169213045,
      "step": 7827,
      "time_per_iteration": 2.547708034515381
    },
    {
      "auxiliary_loss_clip": 0.01149818,
      "auxiliary_loss_mlp": 0.00761317,
      "balance_loss_clip": 1.04693913,
      "balance_loss_mlp": 1.00037503,
      "epoch": 0.941261347922804,
      "flos": 20704908130560.0,
      "grad_norm": 1.65488498100959,
      "language_loss": 0.75595284,
      "learning_rate": 3.6017627930687856e-08,
      "loss": 0.77506417,
      "num_input_tokens_seen": 169232870,
      "step": 7828,
      "time_per_iteration": 3.2857019901275635
    },
    {
      "auxiliary_loss_clip": 0.01105444,
      "auxiliary_loss_mlp": 0.01021398,
      "balance_loss_clip": 1.03908086,
      "balance_loss_mlp": 1.01471066,
      "epoch": 0.9413815908134432,
      "flos": 19421998997760.0,
      "grad_norm": 2.0088585946240354,
      "language_loss": 0.77042317,
      "learning_rate": 3.587060833304267e-08,
      "loss": 0.79169154,
      "num_input_tokens_seen": 169251060,
      "step": 7829,
      "time_per_iteration": 4.3125159740448
    },
    {
      "auxiliary_loss_clip": 0.01153841,
      "auxiliary_loss_mlp": 0.0102507,
      "balance_loss_clip": 1.04700339,
      "balance_loss_mlp": 1.01793277,
      "epoch": 0.9415018337040822,
      "flos": 17493452853120.0,
      "grad_norm": 1.9794684696201283,
      "language_loss": 0.64035976,
      "learning_rate": 3.5723886692264225e-08,
      "loss": 0.66214889,
      "num_input_tokens_seen": 169268600,
      "step": 7830,
      "time_per_iteration": 2.4527854919433594
    },
    {
      "auxiliary_loss_clip": 0.01131958,
      "auxiliary_loss_mlp": 0.01025191,
      "balance_loss_clip": 1.04079461,
      "balance_loss_mlp": 1.01860428,
      "epoch": 0.9416220765947213,
      "flos": 31831613343360.0,
      "grad_norm": 2.0541751213589077,
      "language_loss": 0.62432235,
      "learning_rate": 3.557746303061071e-08,
      "loss": 0.64589387,
      "num_input_tokens_seen": 169290355,
      "step": 7831,
      "time_per_iteration": 2.593571901321411
    },
    {
      "auxiliary_loss_clip": 0.01133843,
      "auxiliary_loss_mlp": 0.01021214,
      "balance_loss_clip": 1.04275656,
      "balance_loss_mlp": 1.01470542,
      "epoch": 0.9417423194853605,
      "flos": 23511973115520.0,
      "grad_norm": 1.5962669450845155,
      "language_loss": 0.72410691,
      "learning_rate": 3.543133737029391e-08,
      "loss": 0.74565744,
      "num_input_tokens_seen": 169310865,
      "step": 7832,
      "time_per_iteration": 2.5236294269561768
    },
    {
      "auxiliary_loss_clip": 0.01154959,
      "auxiliary_loss_mlp": 0.01022886,
      "balance_loss_clip": 1.04571009,
      "balance_loss_mlp": 1.01582038,
      "epoch": 0.9418625623759995,
      "flos": 23915106432000.0,
      "grad_norm": 2.718070526543077,
      "language_loss": 0.69199193,
      "learning_rate": 3.5285509733481214e-08,
      "loss": 0.71377039,
      "num_input_tokens_seen": 169330590,
      "step": 7833,
      "time_per_iteration": 2.481441020965576
    },
    {
      "auxiliary_loss_clip": 0.01147231,
      "auxiliary_loss_mlp": 0.01027193,
      "balance_loss_clip": 1.04372609,
      "balance_loss_mlp": 1.01955175,
      "epoch": 0.9419828052666386,
      "flos": 18076965292800.0,
      "grad_norm": 2.151817133250693,
      "language_loss": 0.76551831,
      "learning_rate": 3.513998014229469e-08,
      "loss": 0.78726256,
      "num_input_tokens_seen": 169349540,
      "step": 7834,
      "time_per_iteration": 2.424168825149536
    },
    {
      "auxiliary_loss_clip": 0.01137357,
      "auxiliary_loss_mlp": 0.01023545,
      "balance_loss_clip": 1.04495227,
      "balance_loss_mlp": 1.01679182,
      "epoch": 0.9421030481572777,
      "flos": 17712328377600.0,
      "grad_norm": 2.578003550508524,
      "language_loss": 0.86075974,
      "learning_rate": 3.499474861881069e-08,
      "loss": 0.88236868,
      "num_input_tokens_seen": 169366765,
      "step": 7835,
      "time_per_iteration": 2.4340837001800537
    },
    {
      "auxiliary_loss_clip": 0.01095575,
      "auxiliary_loss_mlp": 0.01019496,
      "balance_loss_clip": 1.04073191,
      "balance_loss_mlp": 1.01264429,
      "epoch": 0.9422232910479168,
      "flos": 20194114775040.0,
      "grad_norm": 1.8416254845304234,
      "language_loss": 0.67910206,
      "learning_rate": 3.4849815185061136e-08,
      "loss": 0.70025283,
      "num_input_tokens_seen": 169386655,
      "step": 7836,
      "time_per_iteration": 2.6137242317199707
    },
    {
      "auxiliary_loss_clip": 0.01147591,
      "auxiliary_loss_mlp": 0.01021292,
      "balance_loss_clip": 1.04196286,
      "balance_loss_mlp": 1.0149473,
      "epoch": 0.9423435339385559,
      "flos": 18442571875200.0,
      "grad_norm": 1.9351711929411826,
      "language_loss": 0.76200539,
      "learning_rate": 3.470517986303223e-08,
      "loss": 0.78369421,
      "num_input_tokens_seen": 169405640,
      "step": 7837,
      "time_per_iteration": 3.2082364559173584
    },
    {
      "auxiliary_loss_clip": 0.0112106,
      "auxiliary_loss_mlp": 0.01033876,
      "balance_loss_clip": 1.04501939,
      "balance_loss_mlp": 1.02684546,
      "epoch": 0.942463776829195,
      "flos": 20080636732800.0,
      "grad_norm": 1.8018896076547664,
      "language_loss": 0.79175133,
      "learning_rate": 3.4560842674664856e-08,
      "loss": 0.81330073,
      "num_input_tokens_seen": 169424155,
      "step": 7838,
      "time_per_iteration": 2.509373188018799
    },
    {
      "auxiliary_loss_clip": 0.01152592,
      "auxiliary_loss_mlp": 0.01020774,
      "balance_loss_clip": 1.04382777,
      "balance_loss_mlp": 1.01363909,
      "epoch": 0.9425840197198341,
      "flos": 22636255536000.0,
      "grad_norm": 1.7856425887286214,
      "language_loss": 0.75214523,
      "learning_rate": 3.441680364185506e-08,
      "loss": 0.77387887,
      "num_input_tokens_seen": 169444025,
      "step": 7839,
      "time_per_iteration": 2.45835280418396
    },
    {
      "auxiliary_loss_clip": 0.01141236,
      "auxiliary_loss_mlp": 0.01028442,
      "balance_loss_clip": 1.0474081,
      "balance_loss_mlp": 1.0212183,
      "epoch": 0.9427042626104731,
      "flos": 19937892084480.0,
      "grad_norm": 2.1279493500966087,
      "language_loss": 0.75157106,
      "learning_rate": 3.427306278645314e-08,
      "loss": 0.77326781,
      "num_input_tokens_seen": 169462480,
      "step": 7840,
      "time_per_iteration": 2.4605257511138916
    },
    {
      "auxiliary_loss_clip": 0.01108333,
      "auxiliary_loss_mlp": 0.01021506,
      "balance_loss_clip": 1.04143429,
      "balance_loss_mlp": 1.01475585,
      "epoch": 0.9428245055011123,
      "flos": 22856998567680.0,
      "grad_norm": 3.1156319889505286,
      "language_loss": 0.72698045,
      "learning_rate": 3.4129620130264767e-08,
      "loss": 0.74827886,
      "num_input_tokens_seen": 169480840,
      "step": 7841,
      "time_per_iteration": 2.539158821105957
    },
    {
      "auxiliary_loss_clip": 0.01142726,
      "auxiliary_loss_mlp": 0.00761415,
      "balance_loss_clip": 1.04810536,
      "balance_loss_mlp": 1.00046575,
      "epoch": 0.9429447483917514,
      "flos": 20951757371520.0,
      "grad_norm": 2.2051420720485986,
      "language_loss": 0.78080136,
      "learning_rate": 3.398647569505009e-08,
      "loss": 0.79984283,
      "num_input_tokens_seen": 169498265,
      "step": 7842,
      "time_per_iteration": 2.4850659370422363
    },
    {
      "auxiliary_loss_clip": 0.01127832,
      "auxiliary_loss_mlp": 0.0102496,
      "balance_loss_clip": 1.04265666,
      "balance_loss_mlp": 1.01764941,
      "epoch": 0.9430649912823904,
      "flos": 18843658116480.0,
      "grad_norm": 2.458175718296991,
      "language_loss": 0.74562967,
      "learning_rate": 3.384362950252373e-08,
      "loss": 0.76715755,
      "num_input_tokens_seen": 169515235,
      "step": 7843,
      "time_per_iteration": 2.502018928527832
    },
    {
      "auxiliary_loss_clip": 0.01132468,
      "auxiliary_loss_mlp": 0.01021218,
      "balance_loss_clip": 1.04088688,
      "balance_loss_mlp": 1.01443481,
      "epoch": 0.9431852341730296,
      "flos": 32556038837760.0,
      "grad_norm": 2.1438351596002847,
      "language_loss": 0.56783378,
      "learning_rate": 3.3701081574355473e-08,
      "loss": 0.58937061,
      "num_input_tokens_seen": 169537195,
      "step": 7844,
      "time_per_iteration": 2.6157724857330322
    },
    {
      "auxiliary_loss_clip": 0.01044108,
      "auxiliary_loss_mlp": 0.01000561,
      "balance_loss_clip": 1.00895619,
      "balance_loss_mlp": 0.99973881,
      "epoch": 0.9433054770636686,
      "flos": 66904490252160.0,
      "grad_norm": 0.6389719999319686,
      "language_loss": 0.51683843,
      "learning_rate": 3.3558831932169796e-08,
      "loss": 0.53728509,
      "num_input_tokens_seen": 169605865,
      "step": 7845,
      "time_per_iteration": 3.1352052688598633
    },
    {
      "auxiliary_loss_clip": 0.01147098,
      "auxiliary_loss_mlp": 0.01022537,
      "balance_loss_clip": 1.0436132,
      "balance_loss_mlp": 1.01576877,
      "epoch": 0.9434257199543077,
      "flos": 26140346916480.0,
      "grad_norm": 1.8783722097149476,
      "language_loss": 0.88667786,
      "learning_rate": 3.341688059754588e-08,
      "loss": 0.90837425,
      "num_input_tokens_seen": 169621520,
      "step": 7846,
      "time_per_iteration": 2.4822561740875244
    },
    {
      "auxiliary_loss_clip": 0.01128338,
      "auxiliary_loss_mlp": 0.00761133,
      "balance_loss_clip": 1.0414772,
      "balance_loss_mlp": 1.00044942,
      "epoch": 0.9435459628449467,
      "flos": 25003486483200.0,
      "grad_norm": 2.487402269255585,
      "language_loss": 0.7776705,
      "learning_rate": 3.327522759201762e-08,
      "loss": 0.79656523,
      "num_input_tokens_seen": 169641390,
      "step": 7847,
      "time_per_iteration": 2.557314872741699
    },
    {
      "auxiliary_loss_clip": 0.01123359,
      "auxiliary_loss_mlp": 0.01027031,
      "balance_loss_clip": 1.04450023,
      "balance_loss_mlp": 1.01948535,
      "epoch": 0.9436662057355859,
      "flos": 22163240309760.0,
      "grad_norm": 2.1591998272621566,
      "language_loss": 0.67063391,
      "learning_rate": 3.313387293707359e-08,
      "loss": 0.69213784,
      "num_input_tokens_seen": 169660095,
      "step": 7848,
      "time_per_iteration": 2.5176422595977783
    },
    {
      "auxiliary_loss_clip": 0.011214,
      "auxiliary_loss_mlp": 0.01028111,
      "balance_loss_clip": 1.04487526,
      "balance_loss_mlp": 1.02023673,
      "epoch": 0.943786448626225,
      "flos": 20118522602880.0,
      "grad_norm": 2.23324976364037,
      "language_loss": 0.68399274,
      "learning_rate": 3.29928166541571e-08,
      "loss": 0.70548785,
      "num_input_tokens_seen": 169679050,
      "step": 7849,
      "time_per_iteration": 2.5070457458496094
    },
    {
      "auxiliary_loss_clip": 0.0112696,
      "auxiliary_loss_mlp": 0.01022574,
      "balance_loss_clip": 1.04282618,
      "balance_loss_mlp": 1.01555312,
      "epoch": 0.943906691516864,
      "flos": 22090808534400.0,
      "grad_norm": 2.0133440353510057,
      "language_loss": 0.80231166,
      "learning_rate": 3.2852058764666346e-08,
      "loss": 0.823807,
      "num_input_tokens_seen": 169698150,
      "step": 7850,
      "time_per_iteration": 2.4874753952026367
    },
    {
      "auxiliary_loss_clip": 0.01111345,
      "auxiliary_loss_mlp": 0.0102702,
      "balance_loss_clip": 1.04403138,
      "balance_loss_mlp": 1.02015352,
      "epoch": 0.9440269344075032,
      "flos": 35298501212160.0,
      "grad_norm": 2.561723524048758,
      "language_loss": 0.68113899,
      "learning_rate": 3.2711599289954264e-08,
      "loss": 0.70252264,
      "num_input_tokens_seen": 169722185,
      "step": 7851,
      "time_per_iteration": 2.6302433013916016
    },
    {
      "auxiliary_loss_clip": 0.01095275,
      "auxiliary_loss_mlp": 0.01029302,
      "balance_loss_clip": 1.03962159,
      "balance_loss_mlp": 1.02248001,
      "epoch": 0.9441471772981422,
      "flos": 19238136255360.0,
      "grad_norm": 1.8332428705005133,
      "language_loss": 0.7770437,
      "learning_rate": 3.257143825132847e-08,
      "loss": 0.79828948,
      "num_input_tokens_seen": 169740355,
      "step": 7852,
      "time_per_iteration": 2.5932705402374268
    },
    {
      "auxiliary_loss_clip": 0.01136856,
      "auxiliary_loss_mlp": 0.01022315,
      "balance_loss_clip": 1.04454064,
      "balance_loss_mlp": 1.01548481,
      "epoch": 0.9442674201887813,
      "flos": 25739799379200.0,
      "grad_norm": 4.394625244606911,
      "language_loss": 0.76086712,
      "learning_rate": 3.243157567005106e-08,
      "loss": 0.78245878,
      "num_input_tokens_seen": 169758535,
      "step": 7853,
      "time_per_iteration": 2.51511287689209
    },
    {
      "auxiliary_loss_clip": 0.01170208,
      "auxiliary_loss_mlp": 0.01026276,
      "balance_loss_clip": 1.05040717,
      "balance_loss_mlp": 1.01894498,
      "epoch": 0.9443876630794205,
      "flos": 15523321737600.0,
      "grad_norm": 4.034205100257239,
      "language_loss": 0.63952571,
      "learning_rate": 3.2292011567339296e-08,
      "loss": 0.66149056,
      "num_input_tokens_seen": 169776340,
      "step": 7854,
      "time_per_iteration": 3.224992036819458
    },
    {
      "auxiliary_loss_clip": 0.01151649,
      "auxiliary_loss_mlp": 0.00761541,
      "balance_loss_clip": 1.04486895,
      "balance_loss_mlp": 1.00043523,
      "epoch": 0.9445079059700595,
      "flos": 13400821128960.0,
      "grad_norm": 2.1489651399515006,
      "language_loss": 0.55808181,
      "learning_rate": 3.21527459643649e-08,
      "loss": 0.5772137,
      "num_input_tokens_seen": 169793225,
      "step": 7855,
      "time_per_iteration": 3.1556334495544434
    },
    {
      "auxiliary_loss_clip": 0.01153853,
      "auxiliary_loss_mlp": 0.01023941,
      "balance_loss_clip": 1.04605937,
      "balance_loss_mlp": 1.01671994,
      "epoch": 0.9446281488606986,
      "flos": 23659242877440.0,
      "grad_norm": 2.310149782181084,
      "language_loss": 0.74146456,
      "learning_rate": 3.2013778882254536e-08,
      "loss": 0.76324248,
      "num_input_tokens_seen": 169812020,
      "step": 7856,
      "time_per_iteration": 3.2867541313171387
    },
    {
      "auxiliary_loss_clip": 0.01142727,
      "auxiliary_loss_mlp": 0.01029392,
      "balance_loss_clip": 1.04364061,
      "balance_loss_mlp": 1.02255809,
      "epoch": 0.9447483917513377,
      "flos": 25557337267200.0,
      "grad_norm": 1.7366303690542395,
      "language_loss": 0.75807232,
      "learning_rate": 3.1875110342088676e-08,
      "loss": 0.7797935,
      "num_input_tokens_seen": 169833470,
      "step": 7857,
      "time_per_iteration": 2.5026652812957764
    },
    {
      "auxiliary_loss_clip": 0.01133024,
      "auxiliary_loss_mlp": 0.01020985,
      "balance_loss_clip": 1.04511595,
      "balance_loss_mlp": 1.01433003,
      "epoch": 0.9448686346419768,
      "flos": 24535463247360.0,
      "grad_norm": 1.6502779326658825,
      "language_loss": 0.65774155,
      "learning_rate": 3.1736740364904035e-08,
      "loss": 0.67928159,
      "num_input_tokens_seen": 169854000,
      "step": 7858,
      "time_per_iteration": 2.4973976612091064
    },
    {
      "auxiliary_loss_clip": 0.01104734,
      "auxiliary_loss_mlp": 0.0076186,
      "balance_loss_clip": 1.04017067,
      "balance_loss_mlp": 1.00051117,
      "epoch": 0.9449888775326158,
      "flos": 14721256995840.0,
      "grad_norm": 2.04362269263793,
      "language_loss": 0.77462316,
      "learning_rate": 3.159866897169094e-08,
      "loss": 0.79328907,
      "num_input_tokens_seen": 169872200,
      "step": 7859,
      "time_per_iteration": 2.543001174926758
    },
    {
      "auxiliary_loss_clip": 0.01130372,
      "auxiliary_loss_mlp": 0.01025186,
      "balance_loss_clip": 1.04449677,
      "balance_loss_mlp": 1.0182246,
      "epoch": 0.945109120423255,
      "flos": 15447873219840.0,
      "grad_norm": 1.7353632625093676,
      "language_loss": 0.75522542,
      "learning_rate": 3.146089618339487e-08,
      "loss": 0.77678096,
      "num_input_tokens_seen": 169889055,
      "step": 7860,
      "time_per_iteration": 2.4711670875549316
    },
    {
      "auxiliary_loss_clip": 0.01124224,
      "auxiliary_loss_mlp": 0.0102003,
      "balance_loss_clip": 1.04256558,
      "balance_loss_mlp": 1.01294041,
      "epoch": 0.9452293633138941,
      "flos": 25448097029760.0,
      "grad_norm": 2.402187216531958,
      "language_loss": 0.67991292,
      "learning_rate": 3.132342202091554e-08,
      "loss": 0.70135552,
      "num_input_tokens_seen": 169909280,
      "step": 7861,
      "time_per_iteration": 2.54388427734375
    },
    {
      "auxiliary_loss_clip": 0.01165571,
      "auxiliary_loss_mlp": 0.01024379,
      "balance_loss_clip": 1.04651487,
      "balance_loss_mlp": 1.01716661,
      "epoch": 0.9453496062045331,
      "flos": 21215342350080.0,
      "grad_norm": 2.0322010002819706,
      "language_loss": 0.68738711,
      "learning_rate": 3.1186246505107595e-08,
      "loss": 0.70928663,
      "num_input_tokens_seen": 169928420,
      "step": 7862,
      "time_per_iteration": 2.4076900482177734
    },
    {
      "auxiliary_loss_clip": 0.01152482,
      "auxiliary_loss_mlp": 0.01024305,
      "balance_loss_clip": 1.04783988,
      "balance_loss_mlp": 1.01694083,
      "epoch": 0.9454698490951723,
      "flos": 20010898477440.0,
      "grad_norm": 1.751808450621398,
      "language_loss": 0.8378222,
      "learning_rate": 3.104936965678084e-08,
      "loss": 0.85959011,
      "num_input_tokens_seen": 169946750,
      "step": 7863,
      "time_per_iteration": 3.22268009185791
    },
    {
      "auxiliary_loss_clip": 0.01149984,
      "auxiliary_loss_mlp": 0.01021535,
      "balance_loss_clip": 1.04435408,
      "balance_loss_mlp": 1.01422191,
      "epoch": 0.9455900919858113,
      "flos": 21069652786560.0,
      "grad_norm": 1.9975630592408675,
      "language_loss": 0.81775296,
      "learning_rate": 3.091279149669956e-08,
      "loss": 0.83946812,
      "num_input_tokens_seen": 169965540,
      "step": 7864,
      "time_per_iteration": 2.4439079761505127
    },
    {
      "auxiliary_loss_clip": 0.01150629,
      "auxiliary_loss_mlp": 0.00761415,
      "balance_loss_clip": 1.04563642,
      "balance_loss_mlp": 1.00042129,
      "epoch": 0.9457103348764504,
      "flos": 20740854666240.0,
      "grad_norm": 1.759538006397204,
      "language_loss": 0.73269653,
      "learning_rate": 3.0776512045581624e-08,
      "loss": 0.75181699,
      "num_input_tokens_seen": 169984330,
      "step": 7865,
      "time_per_iteration": 2.4579951763153076
    },
    {
      "auxiliary_loss_clip": 0.01131592,
      "auxiliary_loss_mlp": 0.01027437,
      "balance_loss_clip": 1.04380536,
      "balance_loss_mlp": 1.02014744,
      "epoch": 0.9458305777670896,
      "flos": 21428363957760.0,
      "grad_norm": 2.1695022760122167,
      "language_loss": 0.77613556,
      "learning_rate": 3.0640531324101384e-08,
      "loss": 0.79772586,
      "num_input_tokens_seen": 170002095,
      "step": 7866,
      "time_per_iteration": 2.462996244430542
    },
    {
      "auxiliary_loss_clip": 0.01155181,
      "auxiliary_loss_mlp": 0.01024206,
      "balance_loss_clip": 1.04993081,
      "balance_loss_mlp": 1.01644576,
      "epoch": 0.9459508206577286,
      "flos": 20011185786240.0,
      "grad_norm": 1.8162068436731986,
      "language_loss": 0.76079047,
      "learning_rate": 3.0504849352886554e-08,
      "loss": 0.78258437,
      "num_input_tokens_seen": 170020240,
      "step": 7867,
      "time_per_iteration": 2.4411990642547607
    },
    {
      "auxiliary_loss_clip": 0.01151334,
      "auxiliary_loss_mlp": 0.01021391,
      "balance_loss_clip": 1.04655886,
      "balance_loss_mlp": 1.01462269,
      "epoch": 0.9460710635483677,
      "flos": 12166428291840.0,
      "grad_norm": 9.210517222510402,
      "language_loss": 0.712843,
      "learning_rate": 3.036946615252023e-08,
      "loss": 0.73457026,
      "num_input_tokens_seen": 170035770,
      "step": 7868,
      "time_per_iteration": 2.4180634021759033
    },
    {
      "auxiliary_loss_clip": 0.01140638,
      "auxiliary_loss_mlp": 0.01024854,
      "balance_loss_clip": 1.04395854,
      "balance_loss_mlp": 1.01774037,
      "epoch": 0.9461913064390068,
      "flos": 34276196229120.0,
      "grad_norm": 2.3525004973140993,
      "language_loss": 0.67115986,
      "learning_rate": 3.0234381743539984e-08,
      "loss": 0.69281471,
      "num_input_tokens_seen": 170053385,
      "step": 7869,
      "time_per_iteration": 2.5811891555786133
    },
    {
      "auxiliary_loss_clip": 0.01142579,
      "auxiliary_loss_mlp": 0.01021928,
      "balance_loss_clip": 1.04408407,
      "balance_loss_mlp": 1.0148561,
      "epoch": 0.9463115493296459,
      "flos": 19463763536640.0,
      "grad_norm": 2.0340607844145624,
      "language_loss": 0.80144501,
      "learning_rate": 3.0099596146437863e-08,
      "loss": 0.82309002,
      "num_input_tokens_seen": 170070490,
      "step": 7870,
      "time_per_iteration": 2.476649522781372
    },
    {
      "auxiliary_loss_clip": 0.01060806,
      "auxiliary_loss_mlp": 0.01000792,
      "balance_loss_clip": 1.00727773,
      "balance_loss_mlp": 0.99993342,
      "epoch": 0.946431792220285,
      "flos": 70570824387840.0,
      "grad_norm": 0.7681569221726348,
      "language_loss": 0.60079873,
      "learning_rate": 2.996510938166086e-08,
      "loss": 0.62141472,
      "num_input_tokens_seen": 170133465,
      "step": 7871,
      "time_per_iteration": 3.0822348594665527
    },
    {
      "auxiliary_loss_clip": 0.01148955,
      "auxiliary_loss_mlp": 0.0102205,
      "balance_loss_clip": 1.04755247,
      "balance_loss_mlp": 1.01548183,
      "epoch": 0.9465520351109241,
      "flos": 18947906363520.0,
      "grad_norm": 1.8692398209276266,
      "language_loss": 0.7330935,
      "learning_rate": 2.983092146960997e-08,
      "loss": 0.75480354,
      "num_input_tokens_seen": 170150810,
      "step": 7872,
      "time_per_iteration": 2.435206413269043
    },
    {
      "auxiliary_loss_clip": 0.01137302,
      "auxiliary_loss_mlp": 0.01028445,
      "balance_loss_clip": 1.04230714,
      "balance_loss_mlp": 1.02062798,
      "epoch": 0.9466722780015632,
      "flos": 19135647774720.0,
      "grad_norm": 1.9942096704682952,
      "language_loss": 0.80014908,
      "learning_rate": 2.9697032430642256e-08,
      "loss": 0.82180655,
      "num_input_tokens_seen": 170169025,
      "step": 7873,
      "time_per_iteration": 2.4758474826812744
    },
    {
      "auxiliary_loss_clip": 0.01161119,
      "auxiliary_loss_mlp": 0.0102198,
      "balance_loss_clip": 1.04604065,
      "balance_loss_mlp": 1.01557302,
      "epoch": 0.9467925208922022,
      "flos": 17237912520960.0,
      "grad_norm": 2.3568464725218714,
      "language_loss": 0.73452997,
      "learning_rate": 2.9563442285067906e-08,
      "loss": 0.75636089,
      "num_input_tokens_seen": 170186070,
      "step": 7874,
      "time_per_iteration": 2.370945692062378
    },
    {
      "auxiliary_loss_clip": 0.01153819,
      "auxiliary_loss_mlp": 0.01023831,
      "balance_loss_clip": 1.0470686,
      "balance_loss_mlp": 1.0165503,
      "epoch": 0.9469127637828414,
      "flos": 29169016859520.0,
      "grad_norm": 2.236254873262255,
      "language_loss": 0.79644465,
      "learning_rate": 2.943015105315294e-08,
      "loss": 0.81822109,
      "num_input_tokens_seen": 170206265,
      "step": 7875,
      "time_per_iteration": 2.5071771144866943
    },
    {
      "auxiliary_loss_clip": 0.01108762,
      "auxiliary_loss_mlp": 0.01023267,
      "balance_loss_clip": 1.03932202,
      "balance_loss_mlp": 1.0153513,
      "epoch": 0.9470330066734804,
      "flos": 26030460234240.0,
      "grad_norm": 3.049599271476982,
      "language_loss": 0.6647172,
      "learning_rate": 2.929715875511718e-08,
      "loss": 0.68603754,
      "num_input_tokens_seen": 170225300,
      "step": 7876,
      "time_per_iteration": 2.5589890480041504
    },
    {
      "auxiliary_loss_clip": 0.0115107,
      "auxiliary_loss_mlp": 0.01022888,
      "balance_loss_clip": 1.04243565,
      "balance_loss_mlp": 1.01573563,
      "epoch": 0.9471532495641195,
      "flos": 23440906056960.0,
      "grad_norm": 1.815602305408838,
      "language_loss": 0.69892883,
      "learning_rate": 2.9164465411135375e-08,
      "loss": 0.72066844,
      "num_input_tokens_seen": 170245070,
      "step": 7877,
      "time_per_iteration": 2.4514646530151367
    },
    {
      "auxiliary_loss_clip": 0.011531,
      "auxiliary_loss_mlp": 0.01021538,
      "balance_loss_clip": 1.04796863,
      "balance_loss_mlp": 1.01454651,
      "epoch": 0.9472734924547586,
      "flos": 15815850099840.0,
      "grad_norm": 1.7370293766497769,
      "language_loss": 0.80699301,
      "learning_rate": 2.9032071041337426e-08,
      "loss": 0.82873929,
      "num_input_tokens_seen": 170263305,
      "step": 7878,
      "time_per_iteration": 2.4093966484069824
    },
    {
      "auxiliary_loss_clip": 0.01130405,
      "auxiliary_loss_mlp": 0.01027202,
      "balance_loss_clip": 1.04324424,
      "balance_loss_mlp": 1.0203445,
      "epoch": 0.9473937353453977,
      "flos": 11181793697280.0,
      "grad_norm": 1.8808750636263978,
      "language_loss": 0.73073232,
      "learning_rate": 2.889997566580704e-08,
      "loss": 0.75230843,
      "num_input_tokens_seen": 170281460,
      "step": 7879,
      "time_per_iteration": 2.4499850273132324
    },
    {
      "auxiliary_loss_clip": 0.01165363,
      "auxiliary_loss_mlp": 0.0102494,
      "balance_loss_clip": 1.04604936,
      "balance_loss_mlp": 1.01741755,
      "epoch": 0.9475139782360368,
      "flos": 25775530433280.0,
      "grad_norm": 1.5926766350699328,
      "language_loss": 0.7053048,
      "learning_rate": 2.8768179304583086e-08,
      "loss": 0.72720784,
      "num_input_tokens_seen": 170303515,
      "step": 7880,
      "time_per_iteration": 2.4577770233154297
    },
    {
      "auxiliary_loss_clip": 0.01124808,
      "auxiliary_loss_mlp": 0.01026758,
      "balance_loss_clip": 1.04575443,
      "balance_loss_mlp": 1.01992762,
      "epoch": 0.9476342211266758,
      "flos": 22820046451200.0,
      "grad_norm": 1.833557670925971,
      "language_loss": 0.73398244,
      "learning_rate": 2.8636681977659117e-08,
      "loss": 0.75549817,
      "num_input_tokens_seen": 170323165,
      "step": 7881,
      "time_per_iteration": 3.3875198364257812
    },
    {
      "auxiliary_loss_clip": 0.01107102,
      "auxiliary_loss_mlp": 0.01026972,
      "balance_loss_clip": 1.04396105,
      "balance_loss_mlp": 1.01984334,
      "epoch": 0.947754464017315,
      "flos": 20193611984640.0,
      "grad_norm": 2.6093908905250554,
      "language_loss": 0.78018767,
      "learning_rate": 2.850548370498318e-08,
      "loss": 0.80152845,
      "num_input_tokens_seen": 170341005,
      "step": 7882,
      "time_per_iteration": 3.291841983795166
    },
    {
      "auxiliary_loss_clip": 0.01147573,
      "auxiliary_loss_mlp": 0.01020507,
      "balance_loss_clip": 1.04203844,
      "balance_loss_mlp": 1.01399851,
      "epoch": 0.9478747069079541,
      "flos": 24717925359360.0,
      "grad_norm": 4.0760837896489095,
      "language_loss": 0.71137774,
      "learning_rate": 2.8374584506457798e-08,
      "loss": 0.73305857,
      "num_input_tokens_seen": 170362280,
      "step": 7883,
      "time_per_iteration": 3.295448064804077
    },
    {
      "auxiliary_loss_clip": 0.01135331,
      "auxiliary_loss_mlp": 0.01020674,
      "balance_loss_clip": 1.04536188,
      "balance_loss_mlp": 1.01327085,
      "epoch": 0.9479949497985931,
      "flos": 21361355136000.0,
      "grad_norm": 2.475995640123661,
      "language_loss": 0.67193711,
      "learning_rate": 2.824398440193998e-08,
      "loss": 0.69349718,
      "num_input_tokens_seen": 170381080,
      "step": 7884,
      "time_per_iteration": 2.469330072402954
    },
    {
      "auxiliary_loss_clip": 0.01106307,
      "auxiliary_loss_mlp": 0.01025589,
      "balance_loss_clip": 1.04240775,
      "balance_loss_mlp": 1.01794779,
      "epoch": 0.9481151926892323,
      "flos": 18148606968960.0,
      "grad_norm": 3.7627254362232887,
      "language_loss": 0.71908408,
      "learning_rate": 2.811368341124232e-08,
      "loss": 0.74040306,
      "num_input_tokens_seen": 170400150,
      "step": 7885,
      "time_per_iteration": 2.54787015914917
    },
    {
      "auxiliary_loss_clip": 0.0113517,
      "auxiliary_loss_mlp": 0.01029065,
      "balance_loss_clip": 1.04399848,
      "balance_loss_mlp": 1.02203727,
      "epoch": 0.9482354355798713,
      "flos": 22128012046080.0,
      "grad_norm": 2.1640753261137786,
      "language_loss": 0.68272167,
      "learning_rate": 2.7983681554131222e-08,
      "loss": 0.70436406,
      "num_input_tokens_seen": 170420410,
      "step": 7886,
      "time_per_iteration": 2.4963207244873047
    },
    {
      "auxiliary_loss_clip": 0.01135129,
      "auxiliary_loss_mlp": 0.01024581,
      "balance_loss_clip": 1.04341221,
      "balance_loss_mlp": 1.01740503,
      "epoch": 0.9483556784705104,
      "flos": 19063072344960.0,
      "grad_norm": 3.9878115921925037,
      "language_loss": 0.708947,
      "learning_rate": 2.7853978850327365e-08,
      "loss": 0.73054409,
      "num_input_tokens_seen": 170439580,
      "step": 7887,
      "time_per_iteration": 2.463390827178955
    },
    {
      "auxiliary_loss_clip": 0.0112247,
      "auxiliary_loss_mlp": 0.01022628,
      "balance_loss_clip": 1.04718244,
      "balance_loss_mlp": 1.01565993,
      "epoch": 0.9484759213611496,
      "flos": 25777110631680.0,
      "grad_norm": 1.8949550063773881,
      "language_loss": 0.87284803,
      "learning_rate": 2.7724575319507225e-08,
      "loss": 0.89429903,
      "num_input_tokens_seen": 170459290,
      "step": 7888,
      "time_per_iteration": 2.5543770790100098
    },
    {
      "auxiliary_loss_clip": 0.01149039,
      "auxiliary_loss_mlp": 0.01022585,
      "balance_loss_clip": 1.04298997,
      "balance_loss_mlp": 1.01584625,
      "epoch": 0.9485961642517886,
      "flos": 20667740532480.0,
      "grad_norm": 1.7886601044194215,
      "language_loss": 0.77062273,
      "learning_rate": 2.759547098130044e-08,
      "loss": 0.79233897,
      "num_input_tokens_seen": 170478020,
      "step": 7889,
      "time_per_iteration": 2.4523630142211914
    },
    {
      "auxiliary_loss_clip": 0.01161788,
      "auxiliary_loss_mlp": 0.01023784,
      "balance_loss_clip": 1.04602659,
      "balance_loss_mlp": 1.01664329,
      "epoch": 0.9487164071424277,
      "flos": 22674069578880.0,
      "grad_norm": 2.415688712473607,
      "language_loss": 0.76696026,
      "learning_rate": 2.746666585529267e-08,
      "loss": 0.78881598,
      "num_input_tokens_seen": 170498295,
      "step": 7890,
      "time_per_iteration": 3.210886240005493
    },
    {
      "auxiliary_loss_clip": 0.01142561,
      "auxiliary_loss_mlp": 0.01025073,
      "balance_loss_clip": 1.04321933,
      "balance_loss_mlp": 1.0180279,
      "epoch": 0.9488366500330668,
      "flos": 38726461716480.0,
      "grad_norm": 2.0665936921282975,
      "language_loss": 0.74251044,
      "learning_rate": 2.73381599610234e-08,
      "loss": 0.76418686,
      "num_input_tokens_seen": 170518695,
      "step": 7891,
      "time_per_iteration": 2.6543033123016357
    },
    {
      "auxiliary_loss_clip": 0.01144506,
      "auxiliary_loss_mlp": 0.0102551,
      "balance_loss_clip": 1.04194331,
      "balance_loss_mlp": 1.01799703,
      "epoch": 0.9489568929237059,
      "flos": 27890920149120.0,
      "grad_norm": 1.6947393843977296,
      "language_loss": 0.71408939,
      "learning_rate": 2.7209953317987033e-08,
      "loss": 0.73578954,
      "num_input_tokens_seen": 170539735,
      "step": 7892,
      "time_per_iteration": 2.508558511734009
    },
    {
      "auxiliary_loss_clip": 0.01151589,
      "auxiliary_loss_mlp": 0.01021677,
      "balance_loss_clip": 1.04643822,
      "balance_loss_mlp": 1.01466143,
      "epoch": 0.9490771358143449,
      "flos": 33580642291200.0,
      "grad_norm": 3.0216657475812747,
      "language_loss": 0.78105861,
      "learning_rate": 2.7082045945631793e-08,
      "loss": 0.80279124,
      "num_input_tokens_seen": 170561950,
      "step": 7893,
      "time_per_iteration": 2.5604779720306396
    },
    {
      "auxiliary_loss_clip": 0.01114849,
      "auxiliary_loss_mlp": 0.01022422,
      "balance_loss_clip": 1.04189968,
      "balance_loss_mlp": 1.01517975,
      "epoch": 0.9491973787049841,
      "flos": 14793796512000.0,
      "grad_norm": 2.18425932258128,
      "language_loss": 0.69260025,
      "learning_rate": 2.6954437863361712e-08,
      "loss": 0.71397299,
      "num_input_tokens_seen": 170579865,
      "step": 7894,
      "time_per_iteration": 2.504115104675293
    },
    {
      "auxiliary_loss_clip": 0.01095179,
      "auxiliary_loss_mlp": 0.01021642,
      "balance_loss_clip": 1.03972673,
      "balance_loss_mlp": 1.01520801,
      "epoch": 0.9493176215956232,
      "flos": 25332535998720.0,
      "grad_norm": 2.1189384699708183,
      "language_loss": 0.70841491,
      "learning_rate": 2.6827129090534862e-08,
      "loss": 0.72958314,
      "num_input_tokens_seen": 170600165,
      "step": 7895,
      "time_per_iteration": 2.644534111022949
    },
    {
      "auxiliary_loss_clip": 0.01135989,
      "auxiliary_loss_mlp": 0.01026754,
      "balance_loss_clip": 1.04632664,
      "balance_loss_mlp": 1.019279,
      "epoch": 0.9494378644862622,
      "flos": 21029971236480.0,
      "grad_norm": 2.0963763626508687,
      "language_loss": 0.77955866,
      "learning_rate": 2.670011964646335e-08,
      "loss": 0.80118608,
      "num_input_tokens_seen": 170618845,
      "step": 7896,
      "time_per_iteration": 2.4762024879455566
    },
    {
      "auxiliary_loss_clip": 0.01083493,
      "auxiliary_loss_mlp": 0.01023583,
      "balance_loss_clip": 1.03318751,
      "balance_loss_mlp": 1.01596892,
      "epoch": 0.9495581073769014,
      "flos": 15195134148480.0,
      "grad_norm": 3.246791396021892,
      "language_loss": 0.68356919,
      "learning_rate": 2.657340955041487e-08,
      "loss": 0.70463997,
      "num_input_tokens_seen": 170637620,
      "step": 7897,
      "time_per_iteration": 2.6020145416259766
    },
    {
      "auxiliary_loss_clip": 0.01138134,
      "auxiliary_loss_mlp": 0.01026416,
      "balance_loss_clip": 1.04704404,
      "balance_loss_mlp": 1.01871181,
      "epoch": 0.9496783502675404,
      "flos": 28616566705920.0,
      "grad_norm": 1.8595902935398658,
      "language_loss": 0.7189703,
      "learning_rate": 2.6446998821611167e-08,
      "loss": 0.74061584,
      "num_input_tokens_seen": 170657815,
      "step": 7898,
      "time_per_iteration": 2.5292317867279053
    },
    {
      "auxiliary_loss_clip": 0.01109144,
      "auxiliary_loss_mlp": 0.0102707,
      "balance_loss_clip": 1.0407455,
      "balance_loss_mlp": 1.01994395,
      "epoch": 0.9497985931581795,
      "flos": 14866874732160.0,
      "grad_norm": 2.51988940484905,
      "language_loss": 0.71779197,
      "learning_rate": 2.6320887479228228e-08,
      "loss": 0.7391541,
      "num_input_tokens_seen": 170674415,
      "step": 7899,
      "time_per_iteration": 2.5008280277252197
    },
    {
      "auxiliary_loss_clip": 0.0113983,
      "auxiliary_loss_mlp": 0.01028155,
      "balance_loss_clip": 1.04374695,
      "balance_loss_mlp": 1.02104425,
      "epoch": 0.9499188360488187,
      "flos": 27193319136000.0,
      "grad_norm": 2.3891901605304007,
      "language_loss": 0.726197,
      "learning_rate": 2.619507554239786e-08,
      "loss": 0.74787682,
      "num_input_tokens_seen": 170692975,
      "step": 7900,
      "time_per_iteration": 2.5016028881073
    },
    {
      "auxiliary_loss_clip": 0.01136778,
      "auxiliary_loss_mlp": 0.01027396,
      "balance_loss_clip": 1.04434419,
      "balance_loss_mlp": 1.02016592,
      "epoch": 0.9500390789394577,
      "flos": 24316479982080.0,
      "grad_norm": 1.7727170261963179,
      "language_loss": 0.6947031,
      "learning_rate": 2.606956303020502e-08,
      "loss": 0.71634483,
      "num_input_tokens_seen": 170713780,
      "step": 7901,
      "time_per_iteration": 2.5260939598083496
    },
    {
      "auxiliary_loss_clip": 0.0115266,
      "auxiliary_loss_mlp": 0.01022778,
      "balance_loss_clip": 1.04780269,
      "balance_loss_mlp": 1.01538396,
      "epoch": 0.9501593218300968,
      "flos": 14354752573440.0,
      "grad_norm": 1.7317561122193883,
      "language_loss": 0.8402015,
      "learning_rate": 2.5944349961690036e-08,
      "loss": 0.86195588,
      "num_input_tokens_seen": 170730800,
      "step": 7902,
      "time_per_iteration": 2.4129440784454346
    },
    {
      "auxiliary_loss_clip": 0.0112145,
      "auxiliary_loss_mlp": 0.01021667,
      "balance_loss_clip": 1.04260767,
      "balance_loss_mlp": 1.01452649,
      "epoch": 0.9502795647207359,
      "flos": 38728113742080.0,
      "grad_norm": 1.5990768469515926,
      "language_loss": 0.73021787,
      "learning_rate": 2.581943635584749e-08,
      "loss": 0.75164902,
      "num_input_tokens_seen": 170753630,
      "step": 7903,
      "time_per_iteration": 2.6594784259796143
    },
    {
      "auxiliary_loss_clip": 0.01128696,
      "auxiliary_loss_mlp": 0.01019697,
      "balance_loss_clip": 1.04357338,
      "balance_loss_mlp": 1.01336074,
      "epoch": 0.950399807611375,
      "flos": 40808023799040.0,
      "grad_norm": 1.510223638928864,
      "language_loss": 0.65172625,
      "learning_rate": 2.569482223162689e-08,
      "loss": 0.67321008,
      "num_input_tokens_seen": 170777605,
      "step": 7904,
      "time_per_iteration": 2.660789966583252
    },
    {
      "auxiliary_loss_clip": 0.01150816,
      "auxiliary_loss_mlp": 0.01021235,
      "balance_loss_clip": 1.04420209,
      "balance_loss_mlp": 1.01404631,
      "epoch": 0.950520050502014,
      "flos": 23440403266560.0,
      "grad_norm": 1.7361663881273115,
      "language_loss": 0.72268748,
      "learning_rate": 2.5570507607932e-08,
      "loss": 0.74440795,
      "num_input_tokens_seen": 170797520,
      "step": 7905,
      "time_per_iteration": 2.4628560543060303
    },
    {
      "auxiliary_loss_clip": 0.01155819,
      "auxiliary_loss_mlp": 0.01025197,
      "balance_loss_clip": 1.04578066,
      "balance_loss_mlp": 1.01793456,
      "epoch": 0.9506402933926532,
      "flos": 17783718658560.0,
      "grad_norm": 3.220973105537502,
      "language_loss": 0.63576317,
      "learning_rate": 2.54464925036213e-08,
      "loss": 0.65757334,
      "num_input_tokens_seen": 170814810,
      "step": 7906,
      "time_per_iteration": 2.4090638160705566
    },
    {
      "auxiliary_loss_clip": 0.01149802,
      "auxiliary_loss_mlp": 0.0102232,
      "balance_loss_clip": 1.04518294,
      "balance_loss_mlp": 1.01466048,
      "epoch": 0.9507605362832923,
      "flos": 32561928668160.0,
      "grad_norm": 1.8310937891401635,
      "language_loss": 0.60720944,
      "learning_rate": 2.532277693750773e-08,
      "loss": 0.62893069,
      "num_input_tokens_seen": 170835735,
      "step": 7907,
      "time_per_iteration": 3.38250470161438
    },
    {
      "auxiliary_loss_clip": 0.01109587,
      "auxiliary_loss_mlp": 0.0102353,
      "balance_loss_clip": 1.04533446,
      "balance_loss_mlp": 1.01631212,
      "epoch": 0.9508807791739313,
      "flos": 19602054898560.0,
      "grad_norm": 2.0829247431788755,
      "language_loss": 0.76211607,
      "learning_rate": 2.5199360928358948e-08,
      "loss": 0.78344727,
      "num_input_tokens_seen": 170852970,
      "step": 7908,
      "time_per_iteration": 3.339094400405884
    },
    {
      "auxiliary_loss_clip": 0.01139686,
      "auxiliary_loss_mlp": 0.00761169,
      "balance_loss_clip": 1.0419178,
      "balance_loss_mlp": 1.00043821,
      "epoch": 0.9510010220645704,
      "flos": 21471852349440.0,
      "grad_norm": 1.778389177432278,
      "language_loss": 0.86937213,
      "learning_rate": 2.507624449489665e-08,
      "loss": 0.88838065,
      "num_input_tokens_seen": 170871600,
      "step": 7909,
      "time_per_iteration": 3.2919118404388428
    },
    {
      "auxiliary_loss_clip": 0.01137541,
      "auxiliary_loss_mlp": 0.01026501,
      "balance_loss_clip": 1.04638052,
      "balance_loss_mlp": 1.01918125,
      "epoch": 0.9511212649552095,
      "flos": 18879999701760.0,
      "grad_norm": 1.8383305851584208,
      "language_loss": 0.65043491,
      "learning_rate": 2.495342765579811e-08,
      "loss": 0.67207533,
      "num_input_tokens_seen": 170890260,
      "step": 7910,
      "time_per_iteration": 2.4687838554382324
    },
    {
      "auxiliary_loss_clip": 0.01108182,
      "auxiliary_loss_mlp": 0.01020726,
      "balance_loss_clip": 1.04540312,
      "balance_loss_mlp": 1.01413095,
      "epoch": 0.9512415078458486,
      "flos": 20810521094400.0,
      "grad_norm": 1.7067830872009233,
      "language_loss": 0.70883018,
      "learning_rate": 2.4830910429693984e-08,
      "loss": 0.73011923,
      "num_input_tokens_seen": 170910220,
      "step": 7911,
      "time_per_iteration": 2.5718703269958496
    },
    {
      "auxiliary_loss_clip": 0.01163859,
      "auxiliary_loss_mlp": 0.01028249,
      "balance_loss_clip": 1.04575384,
      "balance_loss_mlp": 1.02097416,
      "epoch": 0.9513617507364877,
      "flos": 18369565482240.0,
      "grad_norm": 2.249737197616849,
      "language_loss": 0.79921162,
      "learning_rate": 2.470869283517052e-08,
      "loss": 0.82113266,
      "num_input_tokens_seen": 170928255,
      "step": 7912,
      "time_per_iteration": 2.3896517753601074
    },
    {
      "auxiliary_loss_clip": 0.01143383,
      "auxiliary_loss_mlp": 0.0102455,
      "balance_loss_clip": 1.04266667,
      "balance_loss_mlp": 1.0173403,
      "epoch": 0.9514819936271268,
      "flos": 25010166412800.0,
      "grad_norm": 1.5636190551080271,
      "language_loss": 0.76957893,
      "learning_rate": 2.458677489076777e-08,
      "loss": 0.79125828,
      "num_input_tokens_seen": 170949265,
      "step": 7913,
      "time_per_iteration": 2.507042407989502
    },
    {
      "auxiliary_loss_clip": 0.01140404,
      "auxiliary_loss_mlp": 0.01027256,
      "balance_loss_clip": 1.04278743,
      "balance_loss_mlp": 1.02054405,
      "epoch": 0.9516022365177659,
      "flos": 18662129758080.0,
      "grad_norm": 1.8812950454134774,
      "language_loss": 0.82991415,
      "learning_rate": 2.446515661498072e-08,
      "loss": 0.85159069,
      "num_input_tokens_seen": 170968595,
      "step": 7914,
      "time_per_iteration": 2.4373035430908203
    },
    {
      "auxiliary_loss_clip": 0.01093446,
      "auxiliary_loss_mlp": 0.01026178,
      "balance_loss_clip": 1.04129684,
      "balance_loss_mlp": 1.01921654,
      "epoch": 0.9517224794084049,
      "flos": 25372109808000.0,
      "grad_norm": 3.3668795184327522,
      "language_loss": 0.74233401,
      "learning_rate": 2.434383802625861e-08,
      "loss": 0.76353025,
      "num_input_tokens_seen": 170987550,
      "step": 7915,
      "time_per_iteration": 2.6068801879882812
    },
    {
      "auxiliary_loss_clip": 0.0112181,
      "auxiliary_loss_mlp": 0.01024353,
      "balance_loss_clip": 1.03996992,
      "balance_loss_mlp": 1.01738262,
      "epoch": 0.9518427222990441,
      "flos": 21470918595840.0,
      "grad_norm": 1.7502016370595734,
      "language_loss": 0.73787653,
      "learning_rate": 2.4222819143005168e-08,
      "loss": 0.75933814,
      "num_input_tokens_seen": 171007145,
      "step": 7916,
      "time_per_iteration": 3.2857820987701416
    },
    {
      "auxiliary_loss_clip": 0.01162166,
      "auxiliary_loss_mlp": 0.01023244,
      "balance_loss_clip": 1.04657066,
      "balance_loss_mlp": 1.01639533,
      "epoch": 0.9519629651896832,
      "flos": 21033634423680.0,
      "grad_norm": 4.0626144362388175,
      "language_loss": 0.81006205,
      "learning_rate": 2.4102099983579706e-08,
      "loss": 0.83191621,
      "num_input_tokens_seen": 171026295,
      "step": 7917,
      "time_per_iteration": 2.430516481399536
    },
    {
      "auxiliary_loss_clip": 0.01149456,
      "auxiliary_loss_mlp": 0.01026124,
      "balance_loss_clip": 1.04395151,
      "balance_loss_mlp": 1.01835704,
      "epoch": 0.9520832080803222,
      "flos": 21689219502720.0,
      "grad_norm": 1.8973443977570357,
      "language_loss": 0.77114999,
      "learning_rate": 2.3981680566294236e-08,
      "loss": 0.79290581,
      "num_input_tokens_seen": 171045895,
      "step": 7918,
      "time_per_iteration": 2.4527502059936523
    },
    {
      "auxiliary_loss_clip": 0.01163303,
      "auxiliary_loss_mlp": 0.01025004,
      "balance_loss_clip": 1.04807699,
      "balance_loss_mlp": 1.01848936,
      "epoch": 0.9522034509709614,
      "flos": 23145289125120.0,
      "grad_norm": 1.812190183045559,
      "language_loss": 0.73185623,
      "learning_rate": 2.3861560909416822e-08,
      "loss": 0.75373924,
      "num_input_tokens_seen": 171065445,
      "step": 7919,
      "time_per_iteration": 2.4244539737701416
    },
    {
      "auxiliary_loss_clip": 0.01111961,
      "auxiliary_loss_mlp": 0.01026699,
      "balance_loss_clip": 1.04474926,
      "balance_loss_mlp": 1.01994562,
      "epoch": 0.9523236938616004,
      "flos": 24679428958080.0,
      "grad_norm": 1.6848474862599423,
      "language_loss": 0.82475185,
      "learning_rate": 2.3741741031169325e-08,
      "loss": 0.84613848,
      "num_input_tokens_seen": 171085015,
      "step": 7920,
      "time_per_iteration": 2.5710854530334473
    },
    {
      "auxiliary_loss_clip": 0.01103872,
      "auxiliary_loss_mlp": 0.01027915,
      "balance_loss_clip": 1.03927994,
      "balance_loss_mlp": 1.02136779,
      "epoch": 0.9524439367522395,
      "flos": 22672309812480.0,
      "grad_norm": 1.8287716841394348,
      "language_loss": 0.71649098,
      "learning_rate": 2.3622220949728544e-08,
      "loss": 0.73780882,
      "num_input_tokens_seen": 171103900,
      "step": 7921,
      "time_per_iteration": 2.547158718109131
    },
    {
      "auxiliary_loss_clip": 0.0114276,
      "auxiliary_loss_mlp": 0.01027537,
      "balance_loss_clip": 1.04337382,
      "balance_loss_mlp": 1.01993132,
      "epoch": 0.9525641796428787,
      "flos": 34055525024640.0,
      "grad_norm": 2.8791901054580578,
      "language_loss": 0.61022198,
      "learning_rate": 2.3503000683225526e-08,
      "loss": 0.63192499,
      "num_input_tokens_seen": 171121615,
      "step": 7922,
      "time_per_iteration": 2.545583963394165
    },
    {
      "auxiliary_loss_clip": 0.01165562,
      "auxiliary_loss_mlp": 0.01024035,
      "balance_loss_clip": 1.04637206,
      "balance_loss_mlp": 1.01675129,
      "epoch": 0.9526844225335177,
      "flos": 16727083251840.0,
      "grad_norm": 2.132282022322332,
      "language_loss": 0.84494841,
      "learning_rate": 2.3384080249745585e-08,
      "loss": 0.86684436,
      "num_input_tokens_seen": 171139505,
      "step": 7923,
      "time_per_iteration": 2.4152626991271973
    },
    {
      "auxiliary_loss_clip": 0.01111939,
      "auxiliary_loss_mlp": 0.01024284,
      "balance_loss_clip": 1.04118335,
      "balance_loss_mlp": 1.01775384,
      "epoch": 0.9528046654241568,
      "flos": 36939367330560.0,
      "grad_norm": 2.3650445255666943,
      "language_loss": 0.82840222,
      "learning_rate": 2.3265459667329178e-08,
      "loss": 0.84976447,
      "num_input_tokens_seen": 171158995,
      "step": 7924,
      "time_per_iteration": 2.6985020637512207
    },
    {
      "auxiliary_loss_clip": 0.01137995,
      "auxiliary_loss_mlp": 0.01019972,
      "balance_loss_clip": 1.04471958,
      "balance_loss_mlp": 1.0129447,
      "epoch": 0.9529249083147959,
      "flos": 18255010032000.0,
      "grad_norm": 2.0093128489894374,
      "language_loss": 0.85960305,
      "learning_rate": 2.31471389539708e-08,
      "loss": 0.88118267,
      "num_input_tokens_seen": 171176120,
      "step": 7925,
      "time_per_iteration": 2.4927144050598145
    },
    {
      "auxiliary_loss_clip": 0.01151615,
      "auxiliary_loss_mlp": 0.00760929,
      "balance_loss_clip": 1.0464139,
      "balance_loss_mlp": 1.0004046,
      "epoch": 0.953045151205435,
      "flos": 28658438985600.0,
      "grad_norm": 2.239207463529204,
      "language_loss": 0.7294575,
      "learning_rate": 2.3029118127619872e-08,
      "loss": 0.74858296,
      "num_input_tokens_seen": 171195835,
      "step": 7926,
      "time_per_iteration": 2.5613765716552734
    },
    {
      "auxiliary_loss_clip": 0.01128845,
      "auxiliary_loss_mlp": 0.01022349,
      "balance_loss_clip": 1.04283845,
      "balance_loss_mlp": 1.01494622,
      "epoch": 0.953165394096074,
      "flos": 21835232288640.0,
      "grad_norm": 2.4410223684445755,
      "language_loss": 0.87269485,
      "learning_rate": 2.2911397206179628e-08,
      "loss": 0.89420676,
      "num_input_tokens_seen": 171212585,
      "step": 7927,
      "time_per_iteration": 2.4757931232452393
    },
    {
      "auxiliary_loss_clip": 0.01161736,
      "auxiliary_loss_mlp": 0.01024627,
      "balance_loss_clip": 1.04604971,
      "balance_loss_mlp": 1.01794589,
      "epoch": 0.9532856369867132,
      "flos": 19975059682560.0,
      "grad_norm": 6.581115793409175,
      "language_loss": 0.62622267,
      "learning_rate": 2.279397620750845e-08,
      "loss": 0.64808631,
      "num_input_tokens_seen": 171231630,
      "step": 7928,
      "time_per_iteration": 2.4068655967712402
    },
    {
      "auxiliary_loss_clip": 0.01133473,
      "auxiliary_loss_mlp": 0.01024837,
      "balance_loss_clip": 1.04270196,
      "balance_loss_mlp": 1.01843214,
      "epoch": 0.9534058798773523,
      "flos": 15049588239360.0,
      "grad_norm": 2.1314840585865342,
      "language_loss": 0.78899407,
      "learning_rate": 2.2676855149419195e-08,
      "loss": 0.81057721,
      "num_input_tokens_seen": 171248800,
      "step": 7929,
      "time_per_iteration": 2.44342303276062
    },
    {
      "auxiliary_loss_clip": 0.0113582,
      "auxiliary_loss_mlp": 0.0102509,
      "balance_loss_clip": 1.04875267,
      "balance_loss_mlp": 1.01795495,
      "epoch": 0.9535261227679913,
      "flos": 17602800831360.0,
      "grad_norm": 2.6539089806052143,
      "language_loss": 0.75727117,
      "learning_rate": 2.2560034049678988e-08,
      "loss": 0.77888036,
      "num_input_tokens_seen": 171263150,
      "step": 7930,
      "time_per_iteration": 2.443641424179077
    },
    {
      "auxiliary_loss_clip": 0.01169942,
      "auxiliary_loss_mlp": 0.01024312,
      "balance_loss_clip": 1.0495379,
      "balance_loss_mlp": 1.0171833,
      "epoch": 0.9536463656586305,
      "flos": 23142954741120.0,
      "grad_norm": 1.780541748263086,
      "language_loss": 0.75321794,
      "learning_rate": 2.2443512926008988e-08,
      "loss": 0.77516055,
      "num_input_tokens_seen": 171282480,
      "step": 7931,
      "time_per_iteration": 2.4090731143951416
    },
    {
      "auxiliary_loss_clip": 0.01123183,
      "auxiliary_loss_mlp": 0.01026154,
      "balance_loss_clip": 1.04190397,
      "balance_loss_mlp": 1.01912355,
      "epoch": 0.9537666085492695,
      "flos": 18625033987200.0,
      "grad_norm": 2.6809850364366263,
      "language_loss": 0.69789481,
      "learning_rate": 2.2327291796085946e-08,
      "loss": 0.71938813,
      "num_input_tokens_seen": 171300840,
      "step": 7932,
      "time_per_iteration": 2.4717390537261963
    },
    {
      "auxiliary_loss_clip": 0.01164471,
      "auxiliary_loss_mlp": 0.01025609,
      "balance_loss_clip": 1.0455513,
      "balance_loss_mlp": 1.01853418,
      "epoch": 0.9538868514399086,
      "flos": 18989347680000.0,
      "grad_norm": 3.124842473277786,
      "language_loss": 0.77398485,
      "learning_rate": 2.2211370677540197e-08,
      "loss": 0.79588568,
      "num_input_tokens_seen": 171317365,
      "step": 7933,
      "time_per_iteration": 3.242696762084961
    },
    {
      "auxiliary_loss_clip": 0.0116536,
      "auxiliary_loss_mlp": 0.01026135,
      "balance_loss_clip": 1.0467689,
      "balance_loss_mlp": 1.01891732,
      "epoch": 0.9540070943305478,
      "flos": 16800556521600.0,
      "grad_norm": 3.348007318913691,
      "language_loss": 0.7835173,
      "learning_rate": 2.2095749587957012e-08,
      "loss": 0.8054322,
      "num_input_tokens_seen": 171335270,
      "step": 7934,
      "time_per_iteration": 2.389634609222412
    },
    {
      "auxiliary_loss_clip": 0.01131952,
      "auxiliary_loss_mlp": 0.01026264,
      "balance_loss_clip": 1.04043746,
      "balance_loss_mlp": 1.01871836,
      "epoch": 0.9541273372211868,
      "flos": 20156911263360.0,
      "grad_norm": 2.830768016858018,
      "language_loss": 0.69362223,
      "learning_rate": 2.1980428544876138e-08,
      "loss": 0.71520436,
      "num_input_tokens_seen": 171353910,
      "step": 7935,
      "time_per_iteration": 3.239562511444092
    },
    {
      "auxiliary_loss_clip": 0.01100154,
      "auxiliary_loss_mlp": 0.01024546,
      "balance_loss_clip": 1.03607368,
      "balance_loss_mlp": 1.01690483,
      "epoch": 0.9542475801118259,
      "flos": 26725511381760.0,
      "grad_norm": 1.5325552328243468,
      "language_loss": 0.73867846,
      "learning_rate": 2.1865407565791584e-08,
      "loss": 0.75992548,
      "num_input_tokens_seen": 171375480,
      "step": 7936,
      "time_per_iteration": 3.4552974700927734
    },
    {
      "auxiliary_loss_clip": 0.01135908,
      "auxiliary_loss_mlp": 0.01022589,
      "balance_loss_clip": 1.0413959,
      "balance_loss_mlp": 1.01513267,
      "epoch": 0.954367823002465,
      "flos": 23330911633920.0,
      "grad_norm": 6.233755643949018,
      "language_loss": 0.77534211,
      "learning_rate": 2.175068666815183e-08,
      "loss": 0.79692709,
      "num_input_tokens_seen": 171396320,
      "step": 7937,
      "time_per_iteration": 2.5032925605773926
    },
    {
      "auxiliary_loss_clip": 0.01124115,
      "auxiliary_loss_mlp": 0.01030638,
      "balance_loss_clip": 1.04264879,
      "balance_loss_mlp": 1.02337813,
      "epoch": 0.9544880658931041,
      "flos": 14902713527040.0,
      "grad_norm": 2.1005712454665537,
      "language_loss": 0.78787315,
      "learning_rate": 2.163626586935985e-08,
      "loss": 0.80942065,
      "num_input_tokens_seen": 171412860,
      "step": 7938,
      "time_per_iteration": 2.511671781539917
    },
    {
      "auxiliary_loss_clip": 0.01146452,
      "auxiliary_loss_mlp": 0.0102855,
      "balance_loss_clip": 1.04323411,
      "balance_loss_mlp": 1.02118242,
      "epoch": 0.9546083087837431,
      "flos": 29095902725760.0,
      "grad_norm": 2.280249562238388,
      "language_loss": 0.63033187,
      "learning_rate": 2.1522145186773755e-08,
      "loss": 0.65208197,
      "num_input_tokens_seen": 171431780,
      "step": 7939,
      "time_per_iteration": 2.5599637031555176
    },
    {
      "auxiliary_loss_clip": 0.0113347,
      "auxiliary_loss_mlp": 0.01025053,
      "balance_loss_clip": 1.04318857,
      "balance_loss_mlp": 1.01823378,
      "epoch": 0.9547285516743822,
      "flos": 21142335957120.0,
      "grad_norm": 2.3914949473883462,
      "language_loss": 0.85148418,
      "learning_rate": 2.140832463770481e-08,
      "loss": 0.87306941,
      "num_input_tokens_seen": 171450975,
      "step": 7940,
      "time_per_iteration": 2.4612669944763184
    },
    {
      "auxiliary_loss_clip": 0.0113996,
      "auxiliary_loss_mlp": 0.01021349,
      "balance_loss_clip": 1.04350042,
      "balance_loss_mlp": 1.01423788,
      "epoch": 0.9548487945650214,
      "flos": 27490157130240.0,
      "grad_norm": 2.594106354653213,
      "language_loss": 0.75912237,
      "learning_rate": 2.129480423941987e-08,
      "loss": 0.78073543,
      "num_input_tokens_seen": 171467645,
      "step": 7941,
      "time_per_iteration": 2.5190694332122803
    },
    {
      "auxiliary_loss_clip": 0.01138088,
      "auxiliary_loss_mlp": 0.01022245,
      "balance_loss_clip": 1.04352808,
      "balance_loss_mlp": 1.01566744,
      "epoch": 0.9549690374556604,
      "flos": 22273198819200.0,
      "grad_norm": 1.628924507148828,
      "language_loss": 0.80196249,
      "learning_rate": 2.1181584009140052e-08,
      "loss": 0.82356584,
      "num_input_tokens_seen": 171487185,
      "step": 7942,
      "time_per_iteration": 3.177783250808716
    },
    {
      "auxiliary_loss_clip": 0.01131414,
      "auxiliary_loss_mlp": 0.01025015,
      "balance_loss_clip": 1.04435813,
      "balance_loss_mlp": 1.01856613,
      "epoch": 0.9550892803462995,
      "flos": 17595294888960.0,
      "grad_norm": 2.2168011145549005,
      "language_loss": 0.83744979,
      "learning_rate": 2.10686639640405e-08,
      "loss": 0.85901415,
      "num_input_tokens_seen": 171501275,
      "step": 7943,
      "time_per_iteration": 2.4626402854919434
    },
    {
      "auxiliary_loss_clip": 0.01153053,
      "auxiliary_loss_mlp": 0.01024652,
      "balance_loss_clip": 1.04559457,
      "balance_loss_mlp": 1.01724911,
      "epoch": 0.9552095232369386,
      "flos": 24353144789760.0,
      "grad_norm": 1.7419552880218059,
      "language_loss": 0.81155145,
      "learning_rate": 2.0956044121251294e-08,
      "loss": 0.83332855,
      "num_input_tokens_seen": 171520060,
      "step": 7944,
      "time_per_iteration": 2.489194393157959
    },
    {
      "auxiliary_loss_clip": 0.01123822,
      "auxiliary_loss_mlp": 0.01024473,
      "balance_loss_clip": 1.04558134,
      "balance_loss_mlp": 1.01712394,
      "epoch": 0.9553297661275777,
      "flos": 22746860490240.0,
      "grad_norm": 1.7479338731445038,
      "language_loss": 0.81214035,
      "learning_rate": 2.084372449785654e-08,
      "loss": 0.83362329,
      "num_input_tokens_seen": 171539895,
      "step": 7945,
      "time_per_iteration": 2.5234341621398926
    },
    {
      "auxiliary_loss_clip": 0.01131955,
      "auxiliary_loss_mlp": 0.01025876,
      "balance_loss_clip": 1.04187047,
      "balance_loss_mlp": 1.01883042,
      "epoch": 0.9554500090182168,
      "flos": 15413866018560.0,
      "grad_norm": 1.7917656836217764,
      "language_loss": 0.68747103,
      "learning_rate": 2.0731705110895282e-08,
      "loss": 0.70904934,
      "num_input_tokens_seen": 171557385,
      "step": 7946,
      "time_per_iteration": 2.4476161003112793
    },
    {
      "auxiliary_loss_clip": 0.01155151,
      "auxiliary_loss_mlp": 0.01024817,
      "balance_loss_clip": 1.0483501,
      "balance_loss_mlp": 1.01714623,
      "epoch": 0.9555702519088559,
      "flos": 23513517400320.0,
      "grad_norm": 2.0195917611910446,
      "language_loss": 0.86830354,
      "learning_rate": 2.0619985977360587e-08,
      "loss": 0.89010322,
      "num_input_tokens_seen": 171575705,
      "step": 7947,
      "time_per_iteration": 2.4610509872436523
    },
    {
      "auxiliary_loss_clip": 0.01107197,
      "auxiliary_loss_mlp": 0.0102604,
      "balance_loss_clip": 1.03770041,
      "balance_loss_mlp": 1.01914978,
      "epoch": 0.955690494799495,
      "flos": 22962072827520.0,
      "grad_norm": 1.853082052047393,
      "language_loss": 0.76820534,
      "learning_rate": 2.0508567114200237e-08,
      "loss": 0.78953773,
      "num_input_tokens_seen": 171595620,
      "step": 7948,
      "time_per_iteration": 2.5383851528167725
    },
    {
      "auxiliary_loss_clip": 0.01139389,
      "auxiliary_loss_mlp": 0.01022111,
      "balance_loss_clip": 1.04455066,
      "balance_loss_mlp": 1.0154798,
      "epoch": 0.955810737690134,
      "flos": 26031250333440.0,
      "grad_norm": 3.7066386509107794,
      "language_loss": 0.78812921,
      "learning_rate": 2.0397448538316485e-08,
      "loss": 0.80974424,
      "num_input_tokens_seen": 171616660,
      "step": 7949,
      "time_per_iteration": 2.549020290374756
    },
    {
      "auxiliary_loss_clip": 0.01117263,
      "auxiliary_loss_mlp": 0.01023825,
      "balance_loss_clip": 1.04193544,
      "balance_loss_mlp": 1.01698875,
      "epoch": 0.9559309805807732,
      "flos": 20849951249280.0,
      "grad_norm": 2.79569559823281,
      "language_loss": 0.66580039,
      "learning_rate": 2.028663026656563e-08,
      "loss": 0.68721128,
      "num_input_tokens_seen": 171635515,
      "step": 7950,
      "time_per_iteration": 2.505826473236084
    },
    {
      "auxiliary_loss_clip": 0.01162852,
      "auxiliary_loss_mlp": 0.00762192,
      "balance_loss_clip": 1.04665995,
      "balance_loss_mlp": 1.00042915,
      "epoch": 0.9560512234714122,
      "flos": 21578219498880.0,
      "grad_norm": 1.9237452955086372,
      "language_loss": 0.71839231,
      "learning_rate": 2.0176112315758885e-08,
      "loss": 0.73764277,
      "num_input_tokens_seen": 171653305,
      "step": 7951,
      "time_per_iteration": 2.4165947437286377
    },
    {
      "auxiliary_loss_clip": 0.01116675,
      "auxiliary_loss_mlp": 0.01028879,
      "balance_loss_clip": 1.04335201,
      "balance_loss_mlp": 1.02158952,
      "epoch": 0.9561714663620513,
      "flos": 17450144029440.0,
      "grad_norm": 3.264957688021787,
      "language_loss": 0.69134521,
      "learning_rate": 2.0065894702661957e-08,
      "loss": 0.71280074,
      "num_input_tokens_seen": 171669980,
      "step": 7952,
      "time_per_iteration": 2.504631519317627
    },
    {
      "auxiliary_loss_clip": 0.01115792,
      "auxiliary_loss_mlp": 0.00762011,
      "balance_loss_clip": 1.04038048,
      "balance_loss_mlp": 1.00043249,
      "epoch": 0.9562917092526905,
      "flos": 26098510550400.0,
      "grad_norm": 1.9200275155092934,
      "language_loss": 0.7782023,
      "learning_rate": 1.9955977443994577e-08,
      "loss": 0.79698032,
      "num_input_tokens_seen": 171689970,
      "step": 7953,
      "time_per_iteration": 2.5528924465179443
    },
    {
      "auxiliary_loss_clip": 0.01139017,
      "auxiliary_loss_mlp": 0.0102832,
      "balance_loss_clip": 1.04606366,
      "balance_loss_mlp": 1.02061868,
      "epoch": 0.9564119521433295,
      "flos": 24096742531200.0,
      "grad_norm": 1.9844726399692598,
      "language_loss": 0.61818546,
      "learning_rate": 1.9846360556430965e-08,
      "loss": 0.63985884,
      "num_input_tokens_seen": 171708270,
      "step": 7954,
      "time_per_iteration": 2.486854076385498
    },
    {
      "auxiliary_loss_clip": 0.01161706,
      "auxiliary_loss_mlp": 0.01020885,
      "balance_loss_clip": 1.0447855,
      "balance_loss_mlp": 1.01412845,
      "epoch": 0.9565321950339686,
      "flos": 32008903896960.0,
      "grad_norm": 2.1461605526826943,
      "language_loss": 0.61596388,
      "learning_rate": 1.973704405660004e-08,
      "loss": 0.63778973,
      "num_input_tokens_seen": 171729385,
      "step": 7955,
      "time_per_iteration": 2.4807586669921875
    },
    {
      "auxiliary_loss_clip": 0.0109332,
      "auxiliary_loss_mlp": 0.01024786,
      "balance_loss_clip": 1.04128313,
      "balance_loss_mlp": 1.01813412,
      "epoch": 0.9566524379246077,
      "flos": 23588642695680.0,
      "grad_norm": 1.4773026871911497,
      "language_loss": 0.78230816,
      "learning_rate": 1.9628027961085203e-08,
      "loss": 0.80348927,
      "num_input_tokens_seen": 171752615,
      "step": 7956,
      "time_per_iteration": 2.6084582805633545
    },
    {
      "auxiliary_loss_clip": 0.01109845,
      "auxiliary_loss_mlp": 0.01021198,
      "balance_loss_clip": 1.03786778,
      "balance_loss_mlp": 1.01438785,
      "epoch": 0.9567726808152468,
      "flos": 38067716240640.0,
      "grad_norm": 1.7272869933097705,
      "language_loss": 0.83818829,
      "learning_rate": 1.9519312286423894e-08,
      "loss": 0.85949874,
      "num_input_tokens_seen": 171775810,
      "step": 7957,
      "time_per_iteration": 2.6442482471466064
    },
    {
      "auxiliary_loss_clip": 0.01147918,
      "auxiliary_loss_mlp": 0.01020035,
      "balance_loss_clip": 1.04619336,
      "balance_loss_mlp": 1.0128262,
      "epoch": 0.9568929237058859,
      "flos": 22744059229440.0,
      "grad_norm": 2.0668731747494506,
      "language_loss": 0.77823257,
      "learning_rate": 1.9410897049108255e-08,
      "loss": 0.79991215,
      "num_input_tokens_seen": 171795090,
      "step": 7958,
      "time_per_iteration": 2.4371297359466553
    },
    {
      "auxiliary_loss_clip": 0.01172366,
      "auxiliary_loss_mlp": 0.01024694,
      "balance_loss_clip": 1.05144978,
      "balance_loss_mlp": 1.01723456,
      "epoch": 0.957013166596525,
      "flos": 23841633162240.0,
      "grad_norm": 2.1492120647910298,
      "language_loss": 0.91172862,
      "learning_rate": 1.9302782265584905e-08,
      "loss": 0.93369925,
      "num_input_tokens_seen": 171815755,
      "step": 7959,
      "time_per_iteration": 3.2843074798583984
    },
    {
      "auxiliary_loss_clip": 0.01098423,
      "auxiliary_loss_mlp": 0.01026196,
      "balance_loss_clip": 1.04239273,
      "balance_loss_mlp": 1.01895714,
      "epoch": 0.9571334094871641,
      "flos": 17639286071040.0,
      "grad_norm": 2.1963974297954882,
      "language_loss": 0.87164724,
      "learning_rate": 1.9194967952254282e-08,
      "loss": 0.89289337,
      "num_input_tokens_seen": 171834330,
      "step": 7960,
      "time_per_iteration": 2.5030741691589355
    },
    {
      "auxiliary_loss_clip": 0.0115075,
      "auxiliary_loss_mlp": 0.0102257,
      "balance_loss_clip": 1.04653263,
      "balance_loss_mlp": 1.01517904,
      "epoch": 0.9572536523778031,
      "flos": 15369623441280.0,
      "grad_norm": 2.1855534926130233,
      "language_loss": 0.81008863,
      "learning_rate": 1.9087454125472635e-08,
      "loss": 0.83182186,
      "num_input_tokens_seen": 171848805,
      "step": 7961,
      "time_per_iteration": 2.4061107635498047
    },
    {
      "auxiliary_loss_clip": 0.01164972,
      "auxiliary_loss_mlp": 0.01021859,
      "balance_loss_clip": 1.04690921,
      "balance_loss_mlp": 1.01451898,
      "epoch": 0.9573738952684423,
      "flos": 24969838417920.0,
      "grad_norm": 2.024653760470153,
      "language_loss": 0.78619277,
      "learning_rate": 1.8980240801548696e-08,
      "loss": 0.80806106,
      "num_input_tokens_seen": 171867995,
      "step": 7962,
      "time_per_iteration": 3.1779263019561768
    },
    {
      "auxiliary_loss_clip": 0.01135594,
      "auxiliary_loss_mlp": 0.01022001,
      "balance_loss_clip": 1.04747093,
      "balance_loss_mlp": 1.01507747,
      "epoch": 0.9574941381590814,
      "flos": 25769461034880.0,
      "grad_norm": 1.6299165619662728,
      "language_loss": 0.74275267,
      "learning_rate": 1.8873327996747458e-08,
      "loss": 0.7643286,
      "num_input_tokens_seen": 171886495,
      "step": 7963,
      "time_per_iteration": 3.326136589050293
    },
    {
      "auxiliary_loss_clip": 0.01152431,
      "auxiliary_loss_mlp": 0.01024561,
      "balance_loss_clip": 1.0443151,
      "balance_loss_mlp": 1.01774478,
      "epoch": 0.9576143810497204,
      "flos": 32307178435200.0,
      "grad_norm": 1.9668565514878975,
      "language_loss": 0.65918481,
      "learning_rate": 1.8766715727287053e-08,
      "loss": 0.68095475,
      "num_input_tokens_seen": 171908200,
      "step": 7964,
      "time_per_iteration": 2.525831699371338
    },
    {
      "auxiliary_loss_clip": 0.01153872,
      "auxiliary_loss_mlp": 0.00761995,
      "balance_loss_clip": 1.04485583,
      "balance_loss_mlp": 1.00050509,
      "epoch": 0.9577346239403596,
      "flos": 27745733376000.0,
      "grad_norm": 1.5841434800486671,
      "language_loss": 0.79479587,
      "learning_rate": 1.8660404009340546e-08,
      "loss": 0.81395447,
      "num_input_tokens_seen": 171928650,
      "step": 7965,
      "time_per_iteration": 2.4738192558288574
    },
    {
      "auxiliary_loss_clip": 0.01052641,
      "auxiliary_loss_mlp": 0.01001325,
      "balance_loss_clip": 1.00757861,
      "balance_loss_mlp": 1.00046718,
      "epoch": 0.9578548668309986,
      "flos": 57468313710720.0,
      "grad_norm": 0.8629714234026357,
      "language_loss": 0.59536701,
      "learning_rate": 1.8554392859035485e-08,
      "loss": 0.61590672,
      "num_input_tokens_seen": 171986400,
      "step": 7966,
      "time_per_iteration": 3.0143535137176514
    },
    {
      "auxiliary_loss_clip": 0.01082606,
      "auxiliary_loss_mlp": 0.01023758,
      "balance_loss_clip": 1.03738642,
      "balance_loss_mlp": 1.01682019,
      "epoch": 0.9579751097216377,
      "flos": 19756040503680.0,
      "grad_norm": 1.7255162415298613,
      "language_loss": 0.78941423,
      "learning_rate": 1.8448682292453444e-08,
      "loss": 0.81047785,
      "num_input_tokens_seen": 172005475,
      "step": 7967,
      "time_per_iteration": 2.575777530670166
    },
    {
      "auxiliary_loss_clip": 0.011644,
      "auxiliary_loss_mlp": 0.00761294,
      "balance_loss_clip": 1.04726803,
      "balance_loss_mlp": 1.00043714,
      "epoch": 0.9580953526122769,
      "flos": 18041270152320.0,
      "grad_norm": 1.9599523951132716,
      "language_loss": 0.65948665,
      "learning_rate": 1.8343272325631154e-08,
      "loss": 0.6787436,
      "num_input_tokens_seen": 172024420,
      "step": 7968,
      "time_per_iteration": 2.3758771419525146
    },
    {
      "auxiliary_loss_clip": 0.01085485,
      "auxiliary_loss_mlp": 0.01028224,
      "balance_loss_clip": 1.03934121,
      "balance_loss_mlp": 1.02048755,
      "epoch": 0.9582155955029159,
      "flos": 24270154416000.0,
      "grad_norm": 2.2049259503429677,
      "language_loss": 0.77987635,
      "learning_rate": 1.8238162974558492e-08,
      "loss": 0.80101335,
      "num_input_tokens_seen": 172038350,
      "step": 7969,
      "time_per_iteration": 3.3561575412750244
    },
    {
      "auxiliary_loss_clip": 0.01134892,
      "auxiliary_loss_mlp": 0.0102909,
      "balance_loss_clip": 1.04508424,
      "balance_loss_mlp": 1.02177644,
      "epoch": 0.958335838393555,
      "flos": 22783309816320.0,
      "grad_norm": 2.2488678942268163,
      "language_loss": 0.74685216,
      "learning_rate": 1.8133354255181144e-08,
      "loss": 0.76849198,
      "num_input_tokens_seen": 172058665,
      "step": 7970,
      "time_per_iteration": 2.4849183559417725
    },
    {
      "auxiliary_loss_clip": 0.01143761,
      "auxiliary_loss_mlp": 0.01025063,
      "balance_loss_clip": 1.04262257,
      "balance_loss_mlp": 1.01827741,
      "epoch": 0.958456081284194,
      "flos": 16911484698240.0,
      "grad_norm": 2.037698247234785,
      "language_loss": 0.74285889,
      "learning_rate": 1.802884618339795e-08,
      "loss": 0.76454711,
      "num_input_tokens_seen": 172077470,
      "step": 7971,
      "time_per_iteration": 2.4473037719726562
    },
    {
      "auxiliary_loss_clip": 0.01152308,
      "auxiliary_loss_mlp": 0.01022579,
      "balance_loss_clip": 1.04821038,
      "balance_loss_mlp": 1.01518214,
      "epoch": 0.9585763241748332,
      "flos": 19974951941760.0,
      "grad_norm": 2.000170344588468,
      "language_loss": 0.81331193,
      "learning_rate": 1.7924638775062894e-08,
      "loss": 0.83506083,
      "num_input_tokens_seen": 172096590,
      "step": 7972,
      "time_per_iteration": 2.4401140213012695
    },
    {
      "auxiliary_loss_clip": 0.01118317,
      "auxiliary_loss_mlp": 0.01025462,
      "balance_loss_clip": 1.04457879,
      "balance_loss_mlp": 1.01850927,
      "epoch": 0.9586965670654722,
      "flos": 21395649646080.0,
      "grad_norm": 1.967503584092124,
      "language_loss": 0.81185436,
      "learning_rate": 1.7820732045984444e-08,
      "loss": 0.83329213,
      "num_input_tokens_seen": 172116735,
      "step": 7973,
      "time_per_iteration": 2.5631072521209717
    },
    {
      "auxiliary_loss_clip": 0.01148353,
      "auxiliary_loss_mlp": 0.01024575,
      "balance_loss_clip": 1.04366922,
      "balance_loss_mlp": 1.01722538,
      "epoch": 0.9588168099561113,
      "flos": 21435115714560.0,
      "grad_norm": 1.712006301599308,
      "language_loss": 0.74138802,
      "learning_rate": 1.7717126011924655e-08,
      "loss": 0.76311725,
      "num_input_tokens_seen": 172138320,
      "step": 7974,
      "time_per_iteration": 2.505289316177368
    },
    {
      "auxiliary_loss_clip": 0.01101998,
      "auxiliary_loss_mlp": 0.01024133,
      "balance_loss_clip": 1.0366025,
      "balance_loss_mlp": 1.01701307,
      "epoch": 0.9589370528467505,
      "flos": 11763761852160.0,
      "grad_norm": 2.674116235276764,
      "language_loss": 0.76672012,
      "learning_rate": 1.7613820688600957e-08,
      "loss": 0.78798145,
      "num_input_tokens_seen": 172154225,
      "step": 7975,
      "time_per_iteration": 2.5380561351776123
    },
    {
      "auxiliary_loss_clip": 0.01142381,
      "auxiliary_loss_mlp": 0.01024792,
      "balance_loss_clip": 1.0422163,
      "balance_loss_mlp": 1.01819932,
      "epoch": 0.9590572957373895,
      "flos": 23441516588160.0,
      "grad_norm": 1.8174654296468669,
      "language_loss": 0.7863487,
      "learning_rate": 1.7510816091684588e-08,
      "loss": 0.80802035,
      "num_input_tokens_seen": 172174150,
      "step": 7976,
      "time_per_iteration": 2.4980061054229736
    },
    {
      "auxiliary_loss_clip": 0.01138007,
      "auxiliary_loss_mlp": 0.01026886,
      "balance_loss_clip": 1.04491901,
      "balance_loss_mlp": 1.01980507,
      "epoch": 0.9591775386280286,
      "flos": 22528272274560.0,
      "grad_norm": 2.996606761601389,
      "language_loss": 0.78698194,
      "learning_rate": 1.740811223680083e-08,
      "loss": 0.80863088,
      "num_input_tokens_seen": 172191005,
      "step": 7977,
      "time_per_iteration": 2.467432737350464
    },
    {
      "auxiliary_loss_clip": 0.01164301,
      "auxiliary_loss_mlp": 0.0102541,
      "balance_loss_clip": 1.04647398,
      "balance_loss_mlp": 1.018013,
      "epoch": 0.9592977815186677,
      "flos": 18186959715840.0,
      "grad_norm": 2.3649816352065645,
      "language_loss": 0.7385416,
      "learning_rate": 1.7305709139530334e-08,
      "loss": 0.7604388,
      "num_input_tokens_seen": 172209785,
      "step": 7978,
      "time_per_iteration": 2.3876655101776123
    },
    {
      "auxiliary_loss_clip": 0.01143611,
      "auxiliary_loss_mlp": 0.0102323,
      "balance_loss_clip": 1.04281366,
      "balance_loss_mlp": 1.01604724,
      "epoch": 0.9594180244093068,
      "flos": 16537797555840.0,
      "grad_norm": 2.297310040866127,
      "language_loss": 0.74602473,
      "learning_rate": 1.7203606815407334e-08,
      "loss": 0.76769316,
      "num_input_tokens_seen": 172224380,
      "step": 7979,
      "time_per_iteration": 2.3791959285736084
    },
    {
      "auxiliary_loss_clip": 0.01143854,
      "auxiliary_loss_mlp": 0.0102526,
      "balance_loss_clip": 1.04835653,
      "balance_loss_mlp": 1.0182147,
      "epoch": 0.9595382672999458,
      "flos": 20554334317440.0,
      "grad_norm": 1.6140123118052505,
      "language_loss": 0.79366875,
      "learning_rate": 1.7101805279920557e-08,
      "loss": 0.81535983,
      "num_input_tokens_seen": 172242540,
      "step": 7980,
      "time_per_iteration": 2.459242105484009
    },
    {
      "auxiliary_loss_clip": 0.01164608,
      "auxiliary_loss_mlp": 0.01021678,
      "balance_loss_clip": 1.04711461,
      "balance_loss_mlp": 1.01416826,
      "epoch": 0.959658510190585,
      "flos": 22638266697600.0,
      "grad_norm": 2.1376761115108267,
      "language_loss": 0.80899394,
      "learning_rate": 1.7000304548513643e-08,
      "loss": 0.8308568,
      "num_input_tokens_seen": 172262645,
      "step": 7981,
      "time_per_iteration": 2.4031434059143066
    },
    {
      "auxiliary_loss_clip": 0.0111909,
      "auxiliary_loss_mlp": 0.01025055,
      "balance_loss_clip": 1.04051685,
      "balance_loss_mlp": 1.0180366,
      "epoch": 0.9597787530812241,
      "flos": 19135252725120.0,
      "grad_norm": 2.354070304963403,
      "language_loss": 0.82593018,
      "learning_rate": 1.6899104636583394e-08,
      "loss": 0.84737158,
      "num_input_tokens_seen": 172280695,
      "step": 7982,
      "time_per_iteration": 2.5189952850341797
    },
    {
      "auxiliary_loss_clip": 0.01052758,
      "auxiliary_loss_mlp": 0.01000838,
      "balance_loss_clip": 1.00722671,
      "balance_loss_mlp": 1.0000155,
      "epoch": 0.9598989959718631,
      "flos": 60098124055680.0,
      "grad_norm": 0.7247936510611176,
      "language_loss": 0.61941171,
      "learning_rate": 1.6798205559482638e-08,
      "loss": 0.63994765,
      "num_input_tokens_seen": 172343075,
      "step": 7983,
      "time_per_iteration": 3.1807339191436768
    },
    {
      "auxiliary_loss_clip": 0.01126898,
      "auxiliary_loss_mlp": 0.01027383,
      "balance_loss_clip": 1.04594326,
      "balance_loss_mlp": 1.0198071,
      "epoch": 0.9600192388625023,
      "flos": 20886795624960.0,
      "grad_norm": 2.5727226969359105,
      "language_loss": 0.76545107,
      "learning_rate": 1.669760733251713e-08,
      "loss": 0.78699386,
      "num_input_tokens_seen": 172361950,
      "step": 7984,
      "time_per_iteration": 2.5094847679138184
    },
    {
      "auxiliary_loss_clip": 0.01104054,
      "auxiliary_loss_mlp": 0.01023181,
      "balance_loss_clip": 1.04141057,
      "balance_loss_mlp": 1.0167048,
      "epoch": 0.9601394817531413,
      "flos": 20445740524800.0,
      "grad_norm": 1.6676778808878976,
      "language_loss": 0.8231622,
      "learning_rate": 1.659730997094755e-08,
      "loss": 0.84443456,
      "num_input_tokens_seen": 172380440,
      "step": 7985,
      "time_per_iteration": 2.5675392150878906
    },
    {
      "auxiliary_loss_clip": 0.01143451,
      "auxiliary_loss_mlp": 0.01023737,
      "balance_loss_clip": 1.04309511,
      "balance_loss_mlp": 1.01680207,
      "epoch": 0.9602597246437804,
      "flos": 21507152440320.0,
      "grad_norm": 2.189371102896242,
      "language_loss": 0.62065828,
      "learning_rate": 1.6497313489989283e-08,
      "loss": 0.64233017,
      "num_input_tokens_seen": 172400265,
      "step": 7986,
      "time_per_iteration": 3.309988021850586
    },
    {
      "auxiliary_loss_clip": 0.0110739,
      "auxiliary_loss_mlp": 0.01024528,
      "balance_loss_clip": 1.0347755,
      "balance_loss_mlp": 1.01708341,
      "epoch": 0.9603799675344196,
      "flos": 29935099152000.0,
      "grad_norm": 2.176662426371312,
      "language_loss": 0.70074391,
      "learning_rate": 1.639761790481131e-08,
      "loss": 0.72206306,
      "num_input_tokens_seen": 172421145,
      "step": 7987,
      "time_per_iteration": 2.610603094100952
    },
    {
      "auxiliary_loss_clip": 0.01153397,
      "auxiliary_loss_mlp": 0.01022966,
      "balance_loss_clip": 1.04622841,
      "balance_loss_mlp": 1.01599848,
      "epoch": 0.9605002104250586,
      "flos": 28001525103360.0,
      "grad_norm": 1.8062192392037126,
      "language_loss": 0.79249227,
      "learning_rate": 1.6298223230537754e-08,
      "loss": 0.81425589,
      "num_input_tokens_seen": 172438945,
      "step": 7988,
      "time_per_iteration": 2.484236717224121
    },
    {
      "auxiliary_loss_clip": 0.01133696,
      "auxiliary_loss_mlp": 0.00761984,
      "balance_loss_clip": 1.04295492,
      "balance_loss_mlp": 1.00039208,
      "epoch": 0.9606204533156977,
      "flos": 35590490870400.0,
      "grad_norm": 2.976650739555826,
      "language_loss": 0.69689834,
      "learning_rate": 1.619912948224611e-08,
      "loss": 0.71585512,
      "num_input_tokens_seen": 172460150,
      "step": 7989,
      "time_per_iteration": 4.188363552093506
    },
    {
      "auxiliary_loss_clip": 0.01118726,
      "auxiliary_loss_mlp": 0.01029061,
      "balance_loss_clip": 1.04359126,
      "balance_loss_mlp": 1.0212115,
      "epoch": 0.9607406962063368,
      "flos": 26574614346240.0,
      "grad_norm": 2.5082978190502794,
      "language_loss": 0.6079818,
      "learning_rate": 1.6100336674969682e-08,
      "loss": 0.62945962,
      "num_input_tokens_seen": 172478990,
      "step": 7990,
      "time_per_iteration": 2.54133939743042
    },
    {
      "auxiliary_loss_clip": 0.01112353,
      "auxiliary_loss_mlp": 0.01030608,
      "balance_loss_clip": 1.04011869,
      "balance_loss_mlp": 1.02298164,
      "epoch": 0.9608609390969759,
      "flos": 25331781813120.0,
      "grad_norm": 1.9375775869444478,
      "language_loss": 0.76695067,
      "learning_rate": 1.600184482369449e-08,
      "loss": 0.78838027,
      "num_input_tokens_seen": 172498905,
      "step": 7991,
      "time_per_iteration": 2.5657451152801514
    },
    {
      "auxiliary_loss_clip": 0.01126856,
      "auxiliary_loss_mlp": 0.01022371,
      "balance_loss_clip": 1.04248345,
      "balance_loss_mlp": 1.01446128,
      "epoch": 0.960981181987615,
      "flos": 21069114082560.0,
      "grad_norm": 2.6783739872364594,
      "language_loss": 0.8941471,
      "learning_rate": 1.5903653943362126e-08,
      "loss": 0.9156394,
      "num_input_tokens_seen": 172517900,
      "step": 7992,
      "time_per_iteration": 2.4946417808532715
    },
    {
      "auxiliary_loss_clip": 0.01137278,
      "auxiliary_loss_mlp": 0.01022231,
      "balance_loss_clip": 1.04416943,
      "balance_loss_mlp": 1.01561487,
      "epoch": 0.9611014248782541,
      "flos": 17823256554240.0,
      "grad_norm": 1.9655420976840847,
      "language_loss": 0.76993406,
      "learning_rate": 1.580576404886802e-08,
      "loss": 0.79152912,
      "num_input_tokens_seen": 172536430,
      "step": 7993,
      "time_per_iteration": 2.4922561645507812
    },
    {
      "auxiliary_loss_clip": 0.01149915,
      "auxiliary_loss_mlp": 0.01025593,
      "balance_loss_clip": 1.04450226,
      "balance_loss_mlp": 1.01886058,
      "epoch": 0.9612216677688932,
      "flos": 19354631040000.0,
      "grad_norm": 2.9965339583526482,
      "language_loss": 0.79640782,
      "learning_rate": 1.570817515506162e-08,
      "loss": 0.81816292,
      "num_input_tokens_seen": 172555120,
      "step": 7994,
      "time_per_iteration": 2.429800033569336
    },
    {
      "auxiliary_loss_clip": 0.01162984,
      "auxiliary_loss_mlp": 0.01023004,
      "balance_loss_clip": 1.04747438,
      "balance_loss_mlp": 1.01651645,
      "epoch": 0.9613419106595322,
      "flos": 15808739207040.0,
      "grad_norm": 2.1654653277555913,
      "language_loss": 0.81549096,
      "learning_rate": 1.561088727674753e-08,
      "loss": 0.83735085,
      "num_input_tokens_seen": 172569330,
      "step": 7995,
      "time_per_iteration": 3.1712207794189453
    },
    {
      "auxiliary_loss_clip": 0.01124153,
      "auxiliary_loss_mlp": 0.01031186,
      "balance_loss_clip": 1.04307365,
      "balance_loss_mlp": 1.02313077,
      "epoch": 0.9614621535501714,
      "flos": 25702488126720.0,
      "grad_norm": 4.480018090819754,
      "language_loss": 0.71564418,
      "learning_rate": 1.551390042868417e-08,
      "loss": 0.73719752,
      "num_input_tokens_seen": 172591100,
      "step": 7996,
      "time_per_iteration": 2.6296262741088867
    },
    {
      "auxiliary_loss_clip": 0.0115296,
      "auxiliary_loss_mlp": 0.01022448,
      "balance_loss_clip": 1.04740715,
      "balance_loss_mlp": 1.01515555,
      "epoch": 0.9615823964408104,
      "flos": 17819054663040.0,
      "grad_norm": 2.4070389326038573,
      "language_loss": 0.70827353,
      "learning_rate": 1.5417214625584207e-08,
      "loss": 0.73002762,
      "num_input_tokens_seen": 172608755,
      "step": 7997,
      "time_per_iteration": 2.4281885623931885
    },
    {
      "auxiliary_loss_clip": 0.01145111,
      "auxiliary_loss_mlp": 0.01021469,
      "balance_loss_clip": 1.04231751,
      "balance_loss_mlp": 1.01415002,
      "epoch": 0.9617026393314495,
      "flos": 20190020624640.0,
      "grad_norm": 1.7035342154509898,
      "language_loss": 0.8509208,
      "learning_rate": 1.5320829882114806e-08,
      "loss": 0.87258661,
      "num_input_tokens_seen": 172626830,
      "step": 7998,
      "time_per_iteration": 2.4530656337738037
    },
    {
      "auxiliary_loss_clip": 0.01162587,
      "auxiliary_loss_mlp": 0.01024893,
      "balance_loss_clip": 1.04414797,
      "balance_loss_mlp": 1.01784742,
      "epoch": 0.9618228822220887,
      "flos": 20267013427200.0,
      "grad_norm": 2.103792161629396,
      "language_loss": 0.7945298,
      "learning_rate": 1.5224746212897378e-08,
      "loss": 0.81640458,
      "num_input_tokens_seen": 172646125,
      "step": 7999,
      "time_per_iteration": 2.397569179534912
    },
    {
      "auxiliary_loss_clip": 0.01161861,
      "auxiliary_loss_mlp": 0.01022478,
      "balance_loss_clip": 1.04578233,
      "balance_loss_mlp": 1.01566219,
      "epoch": 0.9619431251127277,
      "flos": 21031300039680.0,
      "grad_norm": 1.7070868867199418,
      "language_loss": 0.77420163,
      "learning_rate": 1.512896363250804e-08,
      "loss": 0.79604501,
      "num_input_tokens_seen": 172666235,
      "step": 8000,
      "time_per_iteration": 2.4012913703918457
    },
    {
      "auxiliary_loss_clip": 0.01151927,
      "auxiliary_loss_mlp": 0.01024658,
      "balance_loss_clip": 1.04453051,
      "balance_loss_mlp": 1.01761556,
      "epoch": 0.9620633680033668,
      "flos": 22382654538240.0,
      "grad_norm": 2.0752732430047525,
      "language_loss": 0.75432396,
      "learning_rate": 1.503348215547673e-08,
      "loss": 0.77608979,
      "num_input_tokens_seen": 172687325,
      "step": 8001,
      "time_per_iteration": 2.463216781616211
    },
    {
      "auxiliary_loss_clip": 0.01133512,
      "auxiliary_loss_mlp": 0.01024969,
      "balance_loss_clip": 1.04337692,
      "balance_loss_mlp": 1.01809025,
      "epoch": 0.962183610894006,
      "flos": 18471730740480.0,
      "grad_norm": 1.7598745730702738,
      "language_loss": 0.80655944,
      "learning_rate": 1.4938301796288078e-08,
      "loss": 0.82814425,
      "num_input_tokens_seen": 172703895,
      "step": 8002,
      "time_per_iteration": 2.439070701599121
    },
    {
      "auxiliary_loss_clip": 0.01163782,
      "auxiliary_loss_mlp": 0.01022461,
      "balance_loss_clip": 1.04655981,
      "balance_loss_mlp": 1.01488554,
      "epoch": 0.962303853784645,
      "flos": 18435245500800.0,
      "grad_norm": 3.8070569528311458,
      "language_loss": 0.81698322,
      "learning_rate": 1.4843422569380537e-08,
      "loss": 0.83884573,
      "num_input_tokens_seen": 172720650,
      "step": 8003,
      "time_per_iteration": 2.3881473541259766
    },
    {
      "auxiliary_loss_clip": 0.01105355,
      "auxiliary_loss_mlp": 0.01022258,
      "balance_loss_clip": 1.03931236,
      "balance_loss_mlp": 1.01543045,
      "epoch": 0.9624240966752841,
      "flos": 26391074826240.0,
      "grad_norm": 2.2307647353505273,
      "language_loss": 0.82862902,
      "learning_rate": 1.4748844489147483e-08,
      "loss": 0.84990513,
      "num_input_tokens_seen": 172737640,
      "step": 8004,
      "time_per_iteration": 2.5483381748199463
    },
    {
      "auxiliary_loss_clip": 0.01135586,
      "auxiliary_loss_mlp": 0.0102131,
      "balance_loss_clip": 1.04193461,
      "balance_loss_mlp": 1.01512051,
      "epoch": 0.9625443395659231,
      "flos": 14647675985280.0,
      "grad_norm": 1.819601697079657,
      "language_loss": 0.71162355,
      "learning_rate": 1.4654567569936326e-08,
      "loss": 0.7331925,
      "num_input_tokens_seen": 172755215,
      "step": 8005,
      "time_per_iteration": 2.44250750541687
    },
    {
      "auxiliary_loss_clip": 0.01103186,
      "auxiliary_loss_mlp": 0.01028933,
      "balance_loss_clip": 1.03991902,
      "balance_loss_mlp": 1.02194762,
      "epoch": 0.9626645824565623,
      "flos": 18367626147840.0,
      "grad_norm": 1.866163557316775,
      "language_loss": 0.83024216,
      "learning_rate": 1.456059182604874e-08,
      "loss": 0.85156333,
      "num_input_tokens_seen": 172774020,
      "step": 8006,
      "time_per_iteration": 2.5249311923980713
    },
    {
      "auxiliary_loss_clip": 0.01165618,
      "auxiliary_loss_mlp": 0.01027199,
      "balance_loss_clip": 1.04809225,
      "balance_loss_mlp": 1.01969528,
      "epoch": 0.9627848253472013,
      "flos": 16580424021120.0,
      "grad_norm": 2.0115270960916143,
      "language_loss": 0.76493579,
      "learning_rate": 1.4466917271740653e-08,
      "loss": 0.78686398,
      "num_input_tokens_seen": 172792220,
      "step": 8007,
      "time_per_iteration": 2.3745837211608887
    },
    {
      "auxiliary_loss_clip": 0.01132993,
      "auxiliary_loss_mlp": 0.01025764,
      "balance_loss_clip": 1.0425992,
      "balance_loss_mlp": 1.01844716,
      "epoch": 0.9629050682378404,
      "flos": 20886867452160.0,
      "grad_norm": 1.8869264242336798,
      "language_loss": 0.68009061,
      "learning_rate": 1.4373543921222697e-08,
      "loss": 0.70167816,
      "num_input_tokens_seen": 172811805,
      "step": 8008,
      "time_per_iteration": 2.478329658508301
    },
    {
      "auxiliary_loss_clip": 0.01134216,
      "auxiliary_loss_mlp": 0.01025379,
      "balance_loss_clip": 1.04455376,
      "balance_loss_mlp": 1.01825011,
      "epoch": 0.9630253111284796,
      "flos": 17019252478080.0,
      "grad_norm": 1.819200845425801,
      "language_loss": 0.78233594,
      "learning_rate": 1.428047178865932e-08,
      "loss": 0.80393189,
      "num_input_tokens_seen": 172828595,
      "step": 8009,
      "time_per_iteration": 2.444878578186035
    },
    {
      "auxiliary_loss_clip": 0.01134872,
      "auxiliary_loss_mlp": 0.01025375,
      "balance_loss_clip": 1.04145074,
      "balance_loss_mlp": 1.01797211,
      "epoch": 0.9631455540191186,
      "flos": 20338942412160.0,
      "grad_norm": 1.6557753227576513,
      "language_loss": 0.74411094,
      "learning_rate": 1.4187700888169451e-08,
      "loss": 0.76571333,
      "num_input_tokens_seen": 172847770,
      "step": 8010,
      "time_per_iteration": 2.4842169284820557
    },
    {
      "auxiliary_loss_clip": 0.01050607,
      "auxiliary_loss_mlp": 0.01000932,
      "balance_loss_clip": 1.00816846,
      "balance_loss_mlp": 1.00010931,
      "epoch": 0.9632657969097577,
      "flos": 65956700033280.0,
      "grad_norm": 0.7521184666515475,
      "language_loss": 0.57041264,
      "learning_rate": 1.40952312338265e-08,
      "loss": 0.59092808,
      "num_input_tokens_seen": 172912415,
      "step": 8011,
      "time_per_iteration": 3.1016392707824707
    },
    {
      "auxiliary_loss_clip": 0.01123823,
      "auxiliary_loss_mlp": 0.01024639,
      "balance_loss_clip": 1.04143739,
      "balance_loss_mlp": 1.01772463,
      "epoch": 0.9633860398003968,
      "flos": 44419523823360.0,
      "grad_norm": 1.8663555050622491,
      "language_loss": 0.68743664,
      "learning_rate": 1.4003062839657909e-08,
      "loss": 0.70892125,
      "num_input_tokens_seen": 172934895,
      "step": 8012,
      "time_per_iteration": 3.6033921241760254
    },
    {
      "auxiliary_loss_clip": 0.01124424,
      "auxiliary_loss_mlp": 0.0101895,
      "balance_loss_clip": 1.04278994,
      "balance_loss_mlp": 1.01215243,
      "epoch": 0.9635062826910359,
      "flos": 24827704300800.0,
      "grad_norm": 1.6342624660680392,
      "language_loss": 0.79802448,
      "learning_rate": 1.391119571964583e-08,
      "loss": 0.81945819,
      "num_input_tokens_seen": 172955835,
      "step": 8013,
      "time_per_iteration": 2.540250778198242
    },
    {
      "auxiliary_loss_clip": 0.01150097,
      "auxiliary_loss_mlp": 0.01026722,
      "balance_loss_clip": 1.04688573,
      "balance_loss_mlp": 1.01936686,
      "epoch": 0.9636265255816749,
      "flos": 15961360095360.0,
      "grad_norm": 1.7553171824376277,
      "language_loss": 0.72552949,
      "learning_rate": 1.3819629887726225e-08,
      "loss": 0.74729764,
      "num_input_tokens_seen": 172973925,
      "step": 8014,
      "time_per_iteration": 2.42315936088562
    },
    {
      "auxiliary_loss_clip": 0.01142499,
      "auxiliary_loss_mlp": 0.01023382,
      "balance_loss_clip": 1.04757643,
      "balance_loss_mlp": 1.01628637,
      "epoch": 0.9637467684723141,
      "flos": 22601781457920.0,
      "grad_norm": 2.148582884711216,
      "language_loss": 0.76414245,
      "learning_rate": 1.3728365357789317e-08,
      "loss": 0.78580129,
      "num_input_tokens_seen": 172993290,
      "step": 8015,
      "time_per_iteration": 3.309800386428833
    },
    {
      "auxiliary_loss_clip": 0.01086686,
      "auxiliary_loss_mlp": 0.01021758,
      "balance_loss_clip": 1.0387907,
      "balance_loss_mlp": 1.01396489,
      "epoch": 0.9638670113629532,
      "flos": 17565812801280.0,
      "grad_norm": 2.9348131178571517,
      "language_loss": 0.76806539,
      "learning_rate": 1.3637402143680254e-08,
      "loss": 0.78914988,
      "num_input_tokens_seen": 173008190,
      "step": 8016,
      "time_per_iteration": 3.4087624549865723
    },
    {
      "auxiliary_loss_clip": 0.01027878,
      "auxiliary_loss_mlp": 0.01005326,
      "balance_loss_clip": 1.01040268,
      "balance_loss_mlp": 1.00456893,
      "epoch": 0.9639872542535922,
      "flos": 55072139379840.0,
      "grad_norm": 0.7240401755084293,
      "language_loss": 0.55069923,
      "learning_rate": 1.3546740259197998e-08,
      "loss": 0.57103133,
      "num_input_tokens_seen": 173061000,
      "step": 8017,
      "time_per_iteration": 3.0416760444641113
    },
    {
      "auxiliary_loss_clip": 0.01136952,
      "auxiliary_loss_mlp": 0.01027191,
      "balance_loss_clip": 1.04455602,
      "balance_loss_mlp": 1.01946592,
      "epoch": 0.9641074971442314,
      "flos": 24134484746880.0,
      "grad_norm": 2.3182985443941777,
      "language_loss": 0.70437837,
      "learning_rate": 1.3456379718095989e-08,
      "loss": 0.7260198,
      "num_input_tokens_seen": 173081415,
      "step": 8018,
      "time_per_iteration": 2.493027448654175
    },
    {
      "auxiliary_loss_clip": 0.01038254,
      "auxiliary_loss_mlp": 0.010025,
      "balance_loss_clip": 1.0074898,
      "balance_loss_mlp": 1.00148702,
      "epoch": 0.9642277400348704,
      "flos": 66747416077440.0,
      "grad_norm": 0.8401229538323925,
      "language_loss": 0.6205194,
      "learning_rate": 1.3366320534081487e-08,
      "loss": 0.64092696,
      "num_input_tokens_seen": 173144095,
      "step": 8019,
      "time_per_iteration": 3.0780186653137207
    },
    {
      "auxiliary_loss_clip": 0.01150467,
      "auxiliary_loss_mlp": 0.01022583,
      "balance_loss_clip": 1.04524851,
      "balance_loss_mlp": 1.01557374,
      "epoch": 0.9643479829255095,
      "flos": 30920272450560.0,
      "grad_norm": 2.3020841670504706,
      "language_loss": 0.762299,
      "learning_rate": 1.3276562720816675e-08,
      "loss": 0.78402954,
      "num_input_tokens_seen": 173165605,
      "step": 8020,
      "time_per_iteration": 2.5196034908294678
    },
    {
      "auxiliary_loss_clip": 0.01164528,
      "auxiliary_loss_mlp": 0.01024522,
      "balance_loss_clip": 1.04593754,
      "balance_loss_mlp": 1.0171932,
      "epoch": 0.9644682258161487,
      "flos": 20048245643520.0,
      "grad_norm": 2.2740153508614322,
      "language_loss": 0.82784462,
      "learning_rate": 1.3187106291917549e-08,
      "loss": 0.84973514,
      "num_input_tokens_seen": 173182595,
      "step": 8021,
      "time_per_iteration": 2.4315030574798584
    },
    {
      "auxiliary_loss_clip": 0.0114628,
      "auxiliary_loss_mlp": 0.01020621,
      "balance_loss_clip": 1.04464972,
      "balance_loss_mlp": 1.01437151,
      "epoch": 0.9645884687067877,
      "flos": 21178713456000.0,
      "grad_norm": 1.6772075495342211,
      "language_loss": 0.70642328,
      "learning_rate": 1.309795126095503e-08,
      "loss": 0.72809225,
      "num_input_tokens_seen": 173200895,
      "step": 8022,
      "time_per_iteration": 3.219061851501465
    },
    {
      "auxiliary_loss_clip": 0.01077201,
      "auxiliary_loss_mlp": 0.01023354,
      "balance_loss_clip": 1.0384481,
      "balance_loss_mlp": 1.01615071,
      "epoch": 0.9647087115974268,
      "flos": 18945967029120.0,
      "grad_norm": 2.0479811731810837,
      "language_loss": 0.80459267,
      "learning_rate": 1.3009097641453192e-08,
      "loss": 0.82559824,
      "num_input_tokens_seen": 173218745,
      "step": 8023,
      "time_per_iteration": 2.6150705814361572
    },
    {
      "auxiliary_loss_clip": 0.01138224,
      "auxiliary_loss_mlp": 0.01020576,
      "balance_loss_clip": 1.0463717,
      "balance_loss_mlp": 1.01352477,
      "epoch": 0.9648289544880659,
      "flos": 16545088016640.0,
      "grad_norm": 1.6444242924151706,
      "language_loss": 0.75677907,
      "learning_rate": 1.2920545446891474e-08,
      "loss": 0.77836704,
      "num_input_tokens_seen": 173235465,
      "step": 8024,
      "time_per_iteration": 2.4677605628967285
    },
    {
      "auxiliary_loss_clip": 0.01138745,
      "auxiliary_loss_mlp": 0.01032666,
      "balance_loss_clip": 1.04642177,
      "balance_loss_mlp": 1.02527535,
      "epoch": 0.964949197378705,
      "flos": 24057527857920.0,
      "grad_norm": 1.731194863134548,
      "language_loss": 0.70507318,
      "learning_rate": 1.2832294690703127e-08,
      "loss": 0.72678733,
      "num_input_tokens_seen": 173254440,
      "step": 8025,
      "time_per_iteration": 2.5037741661071777
    },
    {
      "auxiliary_loss_clip": 0.01151583,
      "auxiliary_loss_mlp": 0.01024383,
      "balance_loss_clip": 1.04662836,
      "balance_loss_mlp": 1.01701283,
      "epoch": 0.965069440269344,
      "flos": 23365565280000.0,
      "grad_norm": 2.622304409622533,
      "language_loss": 0.77384508,
      "learning_rate": 1.2744345386275668e-08,
      "loss": 0.79560471,
      "num_input_tokens_seen": 173273980,
      "step": 8026,
      "time_per_iteration": 2.452924966812134
    },
    {
      "auxiliary_loss_clip": 0.01146549,
      "auxiliary_loss_mlp": 0.01023452,
      "balance_loss_clip": 1.04961061,
      "balance_loss_mlp": 1.01627266,
      "epoch": 0.9651896831599832,
      "flos": 25374875155200.0,
      "grad_norm": 2.094194159590921,
      "language_loss": 0.78486365,
      "learning_rate": 1.265669754695109e-08,
      "loss": 0.80656362,
      "num_input_tokens_seen": 173293550,
      "step": 8027,
      "time_per_iteration": 2.525113344192505
    },
    {
      "auxiliary_loss_clip": 0.01097247,
      "auxiliary_loss_mlp": 0.01026168,
      "balance_loss_clip": 1.0392282,
      "balance_loss_mlp": 1.01872373,
      "epoch": 0.9653099260506223,
      "flos": 22272875596800.0,
      "grad_norm": 1.9654110756514116,
      "language_loss": 0.8197763,
      "learning_rate": 1.2569351186025201e-08,
      "loss": 0.84101051,
      "num_input_tokens_seen": 173312005,
      "step": 8028,
      "time_per_iteration": 2.5802760124206543
    },
    {
      "auxiliary_loss_clip": 0.01109185,
      "auxiliary_loss_mlp": 0.01022687,
      "balance_loss_clip": 1.03895581,
      "balance_loss_mlp": 1.01583862,
      "epoch": 0.9654301689412613,
      "flos": 26760847386240.0,
      "grad_norm": 1.4557781139830313,
      "language_loss": 0.75236416,
      "learning_rate": 1.2482306316748737e-08,
      "loss": 0.77368289,
      "num_input_tokens_seen": 173332450,
      "step": 8029,
      "time_per_iteration": 2.558187961578369
    },
    {
      "auxiliary_loss_clip": 0.01155324,
      "auxiliary_loss_mlp": 0.01021752,
      "balance_loss_clip": 1.04522073,
      "balance_loss_mlp": 1.01528478,
      "epoch": 0.9655504118319005,
      "flos": 17412689122560.0,
      "grad_norm": 2.8016751496578234,
      "language_loss": 0.78271544,
      "learning_rate": 1.2395562952326021e-08,
      "loss": 0.80448627,
      "num_input_tokens_seen": 173349610,
      "step": 8030,
      "time_per_iteration": 2.42197585105896
    },
    {
      "auxiliary_loss_clip": 0.01146571,
      "auxiliary_loss_mlp": 0.01031628,
      "balance_loss_clip": 1.04655778,
      "balance_loss_mlp": 1.02334261,
      "epoch": 0.9656706547225395,
      "flos": 22126970551680.0,
      "grad_norm": 2.689096297879985,
      "language_loss": 0.81186926,
      "learning_rate": 1.2309121105916309e-08,
      "loss": 0.83365124,
      "num_input_tokens_seen": 173367900,
      "step": 8031,
      "time_per_iteration": 2.510634422302246
    },
    {
      "auxiliary_loss_clip": 0.01153724,
      "auxiliary_loss_mlp": 0.01022617,
      "balance_loss_clip": 1.0461762,
      "balance_loss_mlp": 1.01580405,
      "epoch": 0.9657908976131786,
      "flos": 37049289926400.0,
      "grad_norm": 1.942218903475128,
      "language_loss": 0.69271475,
      "learning_rate": 1.222298079063222e-08,
      "loss": 0.71447814,
      "num_input_tokens_seen": 173389040,
      "step": 8032,
      "time_per_iteration": 2.573167085647583
    },
    {
      "auxiliary_loss_clip": 0.01148338,
      "auxiliary_loss_mlp": 0.0102345,
      "balance_loss_clip": 1.04495907,
      "balance_loss_mlp": 1.01681042,
      "epoch": 0.9659111405038178,
      "flos": 24389809597440.0,
      "grad_norm": 2.367894053191486,
      "language_loss": 0.7275871,
      "learning_rate": 1.2137142019541524e-08,
      "loss": 0.74930501,
      "num_input_tokens_seen": 173407595,
      "step": 8033,
      "time_per_iteration": 2.483229637145996
    },
    {
      "auxiliary_loss_clip": 0.0114312,
      "auxiliary_loss_mlp": 0.01025484,
      "balance_loss_clip": 1.04368556,
      "balance_loss_mlp": 1.01894808,
      "epoch": 0.9660313833944568,
      "flos": 25009412227200.0,
      "grad_norm": 2.0519206501182703,
      "language_loss": 0.73214149,
      "learning_rate": 1.2051604805666027e-08,
      "loss": 0.75382757,
      "num_input_tokens_seen": 173424720,
      "step": 8034,
      "time_per_iteration": 2.508850574493408
    },
    {
      "auxiliary_loss_clip": 0.01164758,
      "auxiliary_loss_mlp": 0.00761615,
      "balance_loss_clip": 1.04720926,
      "balance_loss_mlp": 1.0004679,
      "epoch": 0.9661516262850959,
      "flos": 11801575895040.0,
      "grad_norm": 2.2018230336157836,
      "language_loss": 0.78638655,
      "learning_rate": 1.196636916198135e-08,
      "loss": 0.80565029,
      "num_input_tokens_seen": 173442260,
      "step": 8035,
      "time_per_iteration": 2.434105634689331
    },
    {
      "auxiliary_loss_clip": 0.01166037,
      "auxiliary_loss_mlp": 0.01021356,
      "balance_loss_clip": 1.04735994,
      "balance_loss_mlp": 1.01456451,
      "epoch": 0.9662718691757349,
      "flos": 20047778766720.0,
      "grad_norm": 2.3415810863217192,
      "language_loss": 0.76836479,
      "learning_rate": 1.1881435101418036e-08,
      "loss": 0.79023874,
      "num_input_tokens_seen": 173461675,
      "step": 8036,
      "time_per_iteration": 2.4042298793792725
    },
    {
      "auxiliary_loss_clip": 0.01040999,
      "auxiliary_loss_mlp": 0.01000853,
      "balance_loss_clip": 1.0084275,
      "balance_loss_mlp": 1.00004792,
      "epoch": 0.9663921120663741,
      "flos": 68027703517440.0,
      "grad_norm": 0.7314223204400929,
      "language_loss": 0.65539575,
      "learning_rate": 1.1796802636860003e-08,
      "loss": 0.67581427,
      "num_input_tokens_seen": 173530205,
      "step": 8037,
      "time_per_iteration": 3.1086549758911133
    },
    {
      "auxiliary_loss_clip": 0.01164308,
      "auxiliary_loss_mlp": 0.01025713,
      "balance_loss_clip": 1.04581904,
      "balance_loss_mlp": 1.01864982,
      "epoch": 0.9665123549570132,
      "flos": 26322916769280.0,
      "grad_norm": 2.6093817005600024,
      "language_loss": 0.73843199,
      "learning_rate": 1.1712471781146316e-08,
      "loss": 0.76033217,
      "num_input_tokens_seen": 173549540,
      "step": 8038,
      "time_per_iteration": 2.4813179969787598
    },
    {
      "auxiliary_loss_clip": 0.01162232,
      "auxiliary_loss_mlp": 0.01024857,
      "balance_loss_clip": 1.04458714,
      "balance_loss_mlp": 1.01766288,
      "epoch": 0.9666325978476522,
      "flos": 43941121557120.0,
      "grad_norm": 2.4006004273797945,
      "language_loss": 0.66401631,
      "learning_rate": 1.1628442547069628e-08,
      "loss": 0.68588722,
      "num_input_tokens_seen": 173571740,
      "step": 8039,
      "time_per_iteration": 3.4404666423797607
    },
    {
      "auxiliary_loss_clip": 0.01154698,
      "auxiliary_loss_mlp": 0.00762184,
      "balance_loss_clip": 1.04512525,
      "balance_loss_mlp": 1.00049806,
      "epoch": 0.9667528407382914,
      "flos": 21543422198400.0,
      "grad_norm": 1.869458742041927,
      "language_loss": 0.77185112,
      "learning_rate": 1.1544714947377521e-08,
      "loss": 0.79101992,
      "num_input_tokens_seen": 173589425,
      "step": 8040,
      "time_per_iteration": 2.4575209617614746
    },
    {
      "auxiliary_loss_clip": 0.011663,
      "auxiliary_loss_mlp": 0.0102747,
      "balance_loss_clip": 1.04730177,
      "balance_loss_mlp": 1.01940536,
      "epoch": 0.9668730836289304,
      "flos": 23878585278720.0,
      "grad_norm": 2.0660478233787587,
      "language_loss": 0.70098913,
      "learning_rate": 1.1461288994770945e-08,
      "loss": 0.72292686,
      "num_input_tokens_seen": 173608500,
      "step": 8041,
      "time_per_iteration": 2.4305317401885986
    },
    {
      "auxiliary_loss_clip": 0.01165788,
      "auxiliary_loss_mlp": 0.0102665,
      "balance_loss_clip": 1.04506731,
      "balance_loss_mlp": 1.01915205,
      "epoch": 0.9669933265195695,
      "flos": 28293011971200.0,
      "grad_norm": 1.8540090468184476,
      "language_loss": 0.77420712,
      "learning_rate": 1.1378164701906002e-08,
      "loss": 0.79613149,
      "num_input_tokens_seen": 173630265,
      "step": 8042,
      "time_per_iteration": 3.2934224605560303
    },
    {
      "auxiliary_loss_clip": 0.0116739,
      "auxiliary_loss_mlp": 0.0102494,
      "balance_loss_clip": 1.04746318,
      "balance_loss_mlp": 1.01774251,
      "epoch": 0.9671135694102087,
      "flos": 22454763091200.0,
      "grad_norm": 1.7582208785185078,
      "language_loss": 0.67034984,
      "learning_rate": 1.1295342081392156e-08,
      "loss": 0.69227314,
      "num_input_tokens_seen": 173649625,
      "step": 8043,
      "time_per_iteration": 3.2975029945373535
    },
    {
      "auxiliary_loss_clip": 0.01138482,
      "auxiliary_loss_mlp": 0.01021146,
      "balance_loss_clip": 1.04324889,
      "balance_loss_mlp": 1.01383877,
      "epoch": 0.9672338123008477,
      "flos": 20155941596160.0,
      "grad_norm": 1.6501541485930686,
      "language_loss": 0.69209087,
      "learning_rate": 1.1212821145793804e-08,
      "loss": 0.71368718,
      "num_input_tokens_seen": 173669240,
      "step": 8044,
      "time_per_iteration": 2.472506284713745
    },
    {
      "auxiliary_loss_clip": 0.01137483,
      "auxiliary_loss_mlp": 0.01025928,
      "balance_loss_clip": 1.04343355,
      "balance_loss_mlp": 1.01865637,
      "epoch": 0.9673540551914868,
      "flos": 16977487939200.0,
      "grad_norm": 1.965621934423793,
      "language_loss": 0.78626651,
      "learning_rate": 1.1130601907629156e-08,
      "loss": 0.80790061,
      "num_input_tokens_seen": 173686970,
      "step": 8045,
      "time_per_iteration": 2.465703010559082
    },
    {
      "auxiliary_loss_clip": 0.01052286,
      "auxiliary_loss_mlp": 0.01001207,
      "balance_loss_clip": 1.00705481,
      "balance_loss_mlp": 1.00024748,
      "epoch": 0.9674742980821259,
      "flos": 61892903952000.0,
      "grad_norm": 0.8310061385023739,
      "language_loss": 0.64818013,
      "learning_rate": 1.1048684379370899e-08,
      "loss": 0.668715,
      "num_input_tokens_seen": 173747655,
      "step": 8046,
      "time_per_iteration": 3.0365893840789795
    },
    {
      "auxiliary_loss_clip": 0.01127874,
      "auxiliary_loss_mlp": 0.01023229,
      "balance_loss_clip": 1.04432428,
      "balance_loss_mlp": 1.0167743,
      "epoch": 0.967594540972765,
      "flos": 18697824898560.0,
      "grad_norm": 1.975165150900015,
      "language_loss": 0.74594229,
      "learning_rate": 1.0967068573445759e-08,
      "loss": 0.76745331,
      "num_input_tokens_seen": 173765140,
      "step": 8047,
      "time_per_iteration": 2.4526827335357666
    },
    {
      "auxiliary_loss_clip": 0.01131824,
      "auxiliary_loss_mlp": 0.01024695,
      "balance_loss_clip": 1.04140198,
      "balance_loss_mlp": 1.01767612,
      "epoch": 0.967714783863404,
      "flos": 20777411733120.0,
      "grad_norm": 2.2523430456295594,
      "language_loss": 0.65201139,
      "learning_rate": 1.0885754502234945e-08,
      "loss": 0.67357659,
      "num_input_tokens_seen": 173784800,
      "step": 8048,
      "time_per_iteration": 2.486752510070801
    },
    {
      "auxiliary_loss_clip": 0.01120488,
      "auxiliary_loss_mlp": 0.01022743,
      "balance_loss_clip": 1.04309964,
      "balance_loss_mlp": 1.01585615,
      "epoch": 0.9678350267540432,
      "flos": 23185473465600.0,
      "grad_norm": 2.930101495036063,
      "language_loss": 0.77860063,
      "learning_rate": 1.08047421780737e-08,
      "loss": 0.80003291,
      "num_input_tokens_seen": 173803990,
      "step": 8049,
      "time_per_iteration": 3.294926166534424
    },
    {
      "auxiliary_loss_clip": 0.01143411,
      "auxiliary_loss_mlp": 0.00761413,
      "balance_loss_clip": 1.04477239,
      "balance_loss_mlp": 1.00044656,
      "epoch": 0.9679552696446823,
      "flos": 21726063878400.0,
      "grad_norm": 2.39012814248842,
      "language_loss": 0.73707503,
      "learning_rate": 1.0724031613251305e-08,
      "loss": 0.75612324,
      "num_input_tokens_seen": 173821890,
      "step": 8050,
      "time_per_iteration": 2.4826231002807617
    },
    {
      "auxiliary_loss_clip": 0.01158448,
      "auxiliary_loss_mlp": 0.010281,
      "balance_loss_clip": 1.04752302,
      "balance_loss_mlp": 1.02058959,
      "epoch": 0.9680755125353213,
      "flos": 26869046129280.0,
      "grad_norm": 2.1964798174937377,
      "language_loss": 0.66212511,
      "learning_rate": 1.0643622820011744e-08,
      "loss": 0.6839906,
      "num_input_tokens_seen": 173842945,
      "step": 8051,
      "time_per_iteration": 2.506913423538208
    },
    {
      "auxiliary_loss_clip": 0.01167649,
      "auxiliary_loss_mlp": 0.01027458,
      "balance_loss_clip": 1.04657817,
      "balance_loss_mlp": 1.01967049,
      "epoch": 0.9681957554259605,
      "flos": 28325008010880.0,
      "grad_norm": 3.250580618000372,
      "language_loss": 0.67942607,
      "learning_rate": 1.0563515810552814e-08,
      "loss": 0.70137715,
      "num_input_tokens_seen": 173859915,
      "step": 8052,
      "time_per_iteration": 2.4511044025421143
    },
    {
      "auxiliary_loss_clip": 0.01167943,
      "auxiliary_loss_mlp": 0.01024931,
      "balance_loss_clip": 1.05024862,
      "balance_loss_mlp": 1.01860392,
      "epoch": 0.9683159983165995,
      "flos": 20557674282240.0,
      "grad_norm": 1.545115350126205,
      "language_loss": 0.73504066,
      "learning_rate": 1.0483710597026795e-08,
      "loss": 0.75696933,
      "num_input_tokens_seen": 173879775,
      "step": 8053,
      "time_per_iteration": 2.428269386291504
    },
    {
      "auxiliary_loss_clip": 0.01121768,
      "auxiliary_loss_mlp": 0.01027398,
      "balance_loss_clip": 1.0420028,
      "balance_loss_mlp": 1.02030206,
      "epoch": 0.9684362412072386,
      "flos": 24207958016640.0,
      "grad_norm": 1.9586683093633246,
      "language_loss": 0.74320185,
      "learning_rate": 1.0404207191540227e-08,
      "loss": 0.76469362,
      "num_input_tokens_seen": 173900230,
      "step": 8054,
      "time_per_iteration": 2.552485704421997
    },
    {
      "auxiliary_loss_clip": 0.01163779,
      "auxiliary_loss_mlp": 0.01023646,
      "balance_loss_clip": 1.04628396,
      "balance_loss_mlp": 1.016675,
      "epoch": 0.9685564840978778,
      "flos": 22346241125760.0,
      "grad_norm": 1.9054769218514387,
      "language_loss": 0.74835038,
      "learning_rate": 1.0325005606153236e-08,
      "loss": 0.77022457,
      "num_input_tokens_seen": 173919690,
      "step": 8055,
      "time_per_iteration": 2.4212632179260254
    },
    {
      "auxiliary_loss_clip": 0.01112163,
      "auxiliary_loss_mlp": 0.01024409,
      "balance_loss_clip": 1.04104924,
      "balance_loss_mlp": 1.01762271,
      "epoch": 0.9686767269885168,
      "flos": 14386389477120.0,
      "grad_norm": 2.4766435932876942,
      "language_loss": 0.78819972,
      "learning_rate": 1.0246105852881104e-08,
      "loss": 0.80956542,
      "num_input_tokens_seen": 173934790,
      "step": 8056,
      "time_per_iteration": 2.5299460887908936
    },
    {
      "auxiliary_loss_clip": 0.01165878,
      "auxiliary_loss_mlp": 0.0102007,
      "balance_loss_clip": 1.0464952,
      "balance_loss_mlp": 1.01280379,
      "epoch": 0.9687969698791559,
      "flos": 21287630471040.0,
      "grad_norm": 1.9109274534879837,
      "language_loss": 0.78610182,
      "learning_rate": 1.0167507943692476e-08,
      "loss": 0.80796129,
      "num_input_tokens_seen": 173953875,
      "step": 8057,
      "time_per_iteration": 2.458423137664795
    },
    {
      "auxiliary_loss_clip": 0.01148511,
      "auxiliary_loss_mlp": 0.01029773,
      "balance_loss_clip": 1.04701614,
      "balance_loss_mlp": 1.02248335,
      "epoch": 0.968917212769795,
      "flos": 19828328624640.0,
      "grad_norm": 2.2689689022155544,
      "language_loss": 0.71478963,
      "learning_rate": 1.008921189051093e-08,
      "loss": 0.73657244,
      "num_input_tokens_seen": 173971220,
      "step": 8058,
      "time_per_iteration": 2.424872875213623
    },
    {
      "auxiliary_loss_clip": 0.01166278,
      "auxiliary_loss_mlp": 0.01023702,
      "balance_loss_clip": 1.04768538,
      "balance_loss_mlp": 1.01653421,
      "epoch": 0.9690374556604341,
      "flos": 21681749473920.0,
      "grad_norm": 1.9392450467574212,
      "language_loss": 0.77505368,
      "learning_rate": 1.0011217705213848e-08,
      "loss": 0.79695344,
      "num_input_tokens_seen": 173989095,
      "step": 8059,
      "time_per_iteration": 2.43770432472229
    },
    {
      "auxiliary_loss_clip": 0.01148331,
      "auxiliary_loss_mlp": 0.01022937,
      "balance_loss_clip": 1.04623878,
      "balance_loss_mlp": 1.01673484,
      "epoch": 0.9691576985510731,
      "flos": 32635437851520.0,
      "grad_norm": 2.123343916845261,
      "language_loss": 0.74618512,
      "learning_rate": 9.933525399632658e-09,
      "loss": 0.76789784,
      "num_input_tokens_seen": 174007330,
      "step": 8060,
      "time_per_iteration": 2.5370001792907715
    },
    {
      "auxiliary_loss_clip": 0.01135209,
      "auxiliary_loss_mlp": 0.01024343,
      "balance_loss_clip": 1.0440166,
      "balance_loss_mlp": 1.01672876,
      "epoch": 0.9692779414417123,
      "flos": 35663174040960.0,
      "grad_norm": 1.7216145954454074,
      "language_loss": 0.65012652,
      "learning_rate": 9.856134985553488e-09,
      "loss": 0.67172205,
      "num_input_tokens_seen": 174027055,
      "step": 8061,
      "time_per_iteration": 2.6319658756256104
    },
    {
      "auxiliary_loss_clip": 0.01165094,
      "auxiliary_loss_mlp": 0.01023794,
      "balance_loss_clip": 1.04694629,
      "balance_loss_mlp": 1.01640892,
      "epoch": 0.9693981843323514,
      "flos": 28366952117760.0,
      "grad_norm": 1.8314184513888254,
      "language_loss": 0.73636782,
      "learning_rate": 9.77904647471628e-09,
      "loss": 0.75825667,
      "num_input_tokens_seen": 174050235,
      "step": 8062,
      "time_per_iteration": 2.462043523788452
    },
    {
      "auxiliary_loss_clip": 0.01100896,
      "auxiliary_loss_mlp": 0.01023935,
      "balance_loss_clip": 1.03848267,
      "balance_loss_mlp": 1.01683605,
      "epoch": 0.9695184272229904,
      "flos": 23622865378560.0,
      "grad_norm": 1.627487402718025,
      "language_loss": 0.73876345,
      "learning_rate": 9.702259878815454e-09,
      "loss": 0.76001179,
      "num_input_tokens_seen": 174070560,
      "step": 8063,
      "time_per_iteration": 2.5861976146698
    },
    {
      "auxiliary_loss_clip": 0.01154809,
      "auxiliary_loss_mlp": 0.01025747,
      "balance_loss_clip": 1.04831481,
      "balance_loss_mlp": 1.01801002,
      "epoch": 0.9696386701136296,
      "flos": 23294677789440.0,
      "grad_norm": 2.2755134982597247,
      "language_loss": 0.74469686,
      "learning_rate": 9.625775209499254e-09,
      "loss": 0.76650244,
      "num_input_tokens_seen": 174090565,
      "step": 8064,
      "time_per_iteration": 2.446993827819824
    },
    {
      "auxiliary_loss_clip": 0.01117169,
      "auxiliary_loss_mlp": 0.01026692,
      "balance_loss_clip": 1.04014075,
      "balance_loss_mlp": 1.01973653,
      "epoch": 0.9697589130042686,
      "flos": 15121876360320.0,
      "grad_norm": 2.1073407652064695,
      "language_loss": 0.74444348,
      "learning_rate": 9.549592478370172e-09,
      "loss": 0.76588207,
      "num_input_tokens_seen": 174108745,
      "step": 8065,
      "time_per_iteration": 2.489091396331787
    },
    {
      "auxiliary_loss_clip": 0.01151399,
      "auxiliary_loss_mlp": 0.01022232,
      "balance_loss_clip": 1.04410863,
      "balance_loss_mlp": 1.0153296,
      "epoch": 0.9698791558949077,
      "flos": 18879532824960.0,
      "grad_norm": 1.8332367474606057,
      "language_loss": 0.79838645,
      "learning_rate": 9.473711696985632e-09,
      "loss": 0.82012278,
      "num_input_tokens_seen": 174128075,
      "step": 8066,
      "time_per_iteration": 3.286348581314087
    },
    {
      "auxiliary_loss_clip": 0.01134805,
      "auxiliary_loss_mlp": 0.01026192,
      "balance_loss_clip": 1.04295421,
      "balance_loss_mlp": 1.01897371,
      "epoch": 0.9699993987855468,
      "flos": 17931455297280.0,
      "grad_norm": 2.5280725916669637,
      "language_loss": 0.75695115,
      "learning_rate": 9.398132876856201e-09,
      "loss": 0.77856117,
      "num_input_tokens_seen": 174147040,
      "step": 8067,
      "time_per_iteration": 2.44856858253479
    },
    {
      "auxiliary_loss_clip": 0.01019933,
      "auxiliary_loss_mlp": 0.01001354,
      "balance_loss_clip": 1.00769925,
      "balance_loss_mlp": 1.00036418,
      "epoch": 0.9701196416761859,
      "flos": 67182186297600.0,
      "grad_norm": 0.7725819647919476,
      "language_loss": 0.60848546,
      "learning_rate": 9.322856029447379e-09,
      "loss": 0.62869823,
      "num_input_tokens_seen": 174208225,
      "step": 8068,
      "time_per_iteration": 3.8124983310699463
    },
    {
      "auxiliary_loss_clip": 0.01163216,
      "auxiliary_loss_mlp": 0.01027093,
      "balance_loss_clip": 1.04721701,
      "balance_loss_mlp": 1.02021766,
      "epoch": 0.970239884566825,
      "flos": 24277804012800.0,
      "grad_norm": 3.0484449822882316,
      "language_loss": 0.80546355,
      "learning_rate": 9.247881166178695e-09,
      "loss": 0.82736659,
      "num_input_tokens_seen": 174226935,
      "step": 8069,
      "time_per_iteration": 2.439831018447876
    },
    {
      "auxiliary_loss_clip": 0.0113341,
      "auxiliary_loss_mlp": 0.01025749,
      "balance_loss_clip": 1.04554296,
      "balance_loss_mlp": 1.01879644,
      "epoch": 0.970360127457464,
      "flos": 25301689194240.0,
      "grad_norm": 2.2817928502613936,
      "language_loss": 0.76568025,
      "learning_rate": 9.173208298423274e-09,
      "loss": 0.78727186,
      "num_input_tokens_seen": 174248140,
      "step": 8070,
      "time_per_iteration": 3.4088246822357178
    },
    {
      "auxiliary_loss_clip": 0.01107803,
      "auxiliary_loss_mlp": 0.00762062,
      "balance_loss_clip": 1.04301524,
      "balance_loss_mlp": 1.00040603,
      "epoch": 0.9704803703481032,
      "flos": 29572473398400.0,
      "grad_norm": 1.603832583363965,
      "language_loss": 0.76314199,
      "learning_rate": 9.09883743750961e-09,
      "loss": 0.78184056,
      "num_input_tokens_seen": 174271030,
      "step": 8071,
      "time_per_iteration": 2.653918504714966
    },
    {
      "auxiliary_loss_clip": 0.0113509,
      "auxiliary_loss_mlp": 0.01023769,
      "balance_loss_clip": 1.04407382,
      "balance_loss_mlp": 1.0167892,
      "epoch": 0.9706006132387422,
      "flos": 17380046638080.0,
      "grad_norm": 1.5501400718443472,
      "language_loss": 0.83708215,
      "learning_rate": 9.024768594719124e-09,
      "loss": 0.85867071,
      "num_input_tokens_seen": 174289410,
      "step": 8072,
      "time_per_iteration": 2.5123648643493652
    },
    {
      "auxiliary_loss_clip": 0.01127061,
      "auxiliary_loss_mlp": 0.0101999,
      "balance_loss_clip": 1.04702282,
      "balance_loss_mlp": 1.01304591,
      "epoch": 0.9707208561293813,
      "flos": 18186421011840.0,
      "grad_norm": 2.3703793545808134,
      "language_loss": 0.72509736,
      "learning_rate": 8.95100178128816e-09,
      "loss": 0.74656785,
      "num_input_tokens_seen": 174308550,
      "step": 8073,
      "time_per_iteration": 2.4899332523345947
    },
    {
      "auxiliary_loss_clip": 0.01137115,
      "auxiliary_loss_mlp": 0.01024758,
      "balance_loss_clip": 1.0438652,
      "balance_loss_mlp": 1.01708102,
      "epoch": 0.9708410990200205,
      "flos": 31248388212480.0,
      "grad_norm": 2.188654683360268,
      "language_loss": 0.69983995,
      "learning_rate": 8.877537008407321e-09,
      "loss": 0.72145867,
      "num_input_tokens_seen": 174328600,
      "step": 8074,
      "time_per_iteration": 2.5641233921051025
    },
    {
      "auxiliary_loss_clip": 0.01140478,
      "auxiliary_loss_mlp": 0.01022969,
      "balance_loss_clip": 1.04446626,
      "balance_loss_mlp": 1.01608467,
      "epoch": 0.9709613419106595,
      "flos": 30554450386560.0,
      "grad_norm": 1.7964615062313898,
      "language_loss": 0.68543047,
      "learning_rate": 8.804374287221028e-09,
      "loss": 0.70706493,
      "num_input_tokens_seen": 174349835,
      "step": 8075,
      "time_per_iteration": 2.560720920562744
    },
    {
      "auxiliary_loss_clip": 0.01114021,
      "auxiliary_loss_mlp": 0.01023088,
      "balance_loss_clip": 1.03740501,
      "balance_loss_mlp": 1.01593816,
      "epoch": 0.9710815848012986,
      "flos": 23730166281600.0,
      "grad_norm": 1.621147413992952,
      "language_loss": 0.84582198,
      "learning_rate": 8.731513628827958e-09,
      "loss": 0.8671931,
      "num_input_tokens_seen": 174369200,
      "step": 8076,
      "time_per_iteration": 3.271825075149536
    },
    {
      "auxiliary_loss_clip": 0.01152095,
      "auxiliary_loss_mlp": 0.01023799,
      "balance_loss_clip": 1.04534686,
      "balance_loss_mlp": 1.0165956,
      "epoch": 0.9712018276919377,
      "flos": 23761875012480.0,
      "grad_norm": 3.96040113086316,
      "language_loss": 0.82598138,
      "learning_rate": 8.658955044280825e-09,
      "loss": 0.84774029,
      "num_input_tokens_seen": 174388125,
      "step": 8077,
      "time_per_iteration": 2.47666335105896
    },
    {
      "auxiliary_loss_clip": 0.01150589,
      "auxiliary_loss_mlp": 0.01021192,
      "balance_loss_clip": 1.04697752,
      "balance_loss_mlp": 1.01399779,
      "epoch": 0.9713220705825768,
      "flos": 23330983461120.0,
      "grad_norm": 2.1279930362772497,
      "language_loss": 0.77493244,
      "learning_rate": 8.586698544587268e-09,
      "loss": 0.79665029,
      "num_input_tokens_seen": 174409735,
      "step": 8078,
      "time_per_iteration": 2.4621527194976807
    },
    {
      "auxiliary_loss_clip": 0.01130648,
      "auxiliary_loss_mlp": 0.01025276,
      "balance_loss_clip": 1.04271126,
      "balance_loss_mlp": 1.01736045,
      "epoch": 0.9714423134732159,
      "flos": 22200946611840.0,
      "grad_norm": 1.942505467808045,
      "language_loss": 0.74173582,
      "learning_rate": 8.514744140707853e-09,
      "loss": 0.76329511,
      "num_input_tokens_seen": 174428875,
      "step": 8079,
      "time_per_iteration": 2.49058198928833
    },
    {
      "auxiliary_loss_clip": 0.01161932,
      "auxiliary_loss_mlp": 0.01021949,
      "balance_loss_clip": 1.04512179,
      "balance_loss_mlp": 1.01533628,
      "epoch": 0.971562556363855,
      "flos": 20229917656320.0,
      "grad_norm": 1.6707515383438474,
      "language_loss": 0.76542616,
      "learning_rate": 8.443091843558515e-09,
      "loss": 0.78726494,
      "num_input_tokens_seen": 174447960,
      "step": 8080,
      "time_per_iteration": 2.4033408164978027
    },
    {
      "auxiliary_loss_clip": 0.01131959,
      "auxiliary_loss_mlp": 0.01020976,
      "balance_loss_clip": 1.04291368,
      "balance_loss_mlp": 1.01365054,
      "epoch": 0.9716827992544941,
      "flos": 24970197553920.0,
      "grad_norm": 2.096770904796048,
      "language_loss": 0.64655209,
      "learning_rate": 8.37174166400878e-09,
      "loss": 0.66808146,
      "num_input_tokens_seen": 174463535,
      "step": 8081,
      "time_per_iteration": 2.5080575942993164
    },
    {
      "auxiliary_loss_clip": 0.01164134,
      "auxiliary_loss_mlp": 0.0102404,
      "balance_loss_clip": 1.04750323,
      "balance_loss_mlp": 1.01714945,
      "epoch": 0.9718030421451331,
      "flos": 24681476033280.0,
      "grad_norm": 2.005648810817171,
      "language_loss": 0.84900779,
      "learning_rate": 8.300693612881992e-09,
      "loss": 0.87088954,
      "num_input_tokens_seen": 174483600,
      "step": 8082,
      "time_per_iteration": 2.4283270835876465
    },
    {
      "auxiliary_loss_clip": 0.01148927,
      "auxiliary_loss_mlp": 0.0076156,
      "balance_loss_clip": 1.0455029,
      "balance_loss_mlp": 1.00043702,
      "epoch": 0.9719232850357723,
      "flos": 22090700793600.0,
      "grad_norm": 1.8154833613872918,
      "language_loss": 0.81318843,
      "learning_rate": 8.22994770095664e-09,
      "loss": 0.83229339,
      "num_input_tokens_seen": 174502175,
      "step": 8083,
      "time_per_iteration": 2.4316601753234863
    },
    {
      "auxiliary_loss_clip": 0.0113981,
      "auxiliary_loss_mlp": 0.01023543,
      "balance_loss_clip": 1.04910612,
      "balance_loss_mlp": 1.01608944,
      "epoch": 0.9720435279264114,
      "flos": 23656908493440.0,
      "grad_norm": 2.195695438138768,
      "language_loss": 0.75365877,
      "learning_rate": 8.159503938964585e-09,
      "loss": 0.77529228,
      "num_input_tokens_seen": 174519495,
      "step": 8084,
      "time_per_iteration": 2.486135959625244
    },
    {
      "auxiliary_loss_clip": 0.01115123,
      "auxiliary_loss_mlp": 0.01017586,
      "balance_loss_clip": 1.04169464,
      "balance_loss_mlp": 1.01093674,
      "epoch": 0.9721637708170504,
      "flos": 28365910623360.0,
      "grad_norm": 1.8123352055315165,
      "language_loss": 0.70398134,
      "learning_rate": 8.089362337592164e-09,
      "loss": 0.72530842,
      "num_input_tokens_seen": 174543120,
      "step": 8085,
      "time_per_iteration": 2.5561342239379883
    },
    {
      "auxiliary_loss_clip": 0.0113299,
      "auxiliary_loss_mlp": 0.01025978,
      "balance_loss_clip": 1.04344547,
      "balance_loss_mlp": 1.01916194,
      "epoch": 0.9722840137076896,
      "flos": 29130807767040.0,
      "grad_norm": 1.5550398862774735,
      "language_loss": 0.72033215,
      "learning_rate": 8.019522907479536e-09,
      "loss": 0.74192178,
      "num_input_tokens_seen": 174563480,
      "step": 8086,
      "time_per_iteration": 2.519136428833008
    },
    {
      "auxiliary_loss_clip": 0.01154088,
      "auxiliary_loss_mlp": 0.01026556,
      "balance_loss_clip": 1.04679084,
      "balance_loss_mlp": 1.01915598,
      "epoch": 0.9724042565983286,
      "flos": 19243954258560.0,
      "grad_norm": 2.362611294194622,
      "language_loss": 0.7724455,
      "learning_rate": 7.949985659221558e-09,
      "loss": 0.79425192,
      "num_input_tokens_seen": 174580745,
      "step": 8087,
      "time_per_iteration": 2.426196336746216
    },
    {
      "auxiliary_loss_clip": 0.01140333,
      "auxiliary_loss_mlp": 0.01025646,
      "balance_loss_clip": 1.04531229,
      "balance_loss_mlp": 1.01915216,
      "epoch": 0.9725244994889677,
      "flos": 23039676161280.0,
      "grad_norm": 2.0301704482002187,
      "language_loss": 0.78899974,
      "learning_rate": 7.880750603366904e-09,
      "loss": 0.81065953,
      "num_input_tokens_seen": 174599615,
      "step": 8088,
      "time_per_iteration": 2.466291666030884
    },
    {
      "auxiliary_loss_clip": 0.01130884,
      "auxiliary_loss_mlp": 0.01025933,
      "balance_loss_clip": 1.04219651,
      "balance_loss_mlp": 1.01794887,
      "epoch": 0.9726447423796069,
      "flos": 23367468700800.0,
      "grad_norm": 1.8461772697387901,
      "language_loss": 0.79450363,
      "learning_rate": 7.811817750418282e-09,
      "loss": 0.81607181,
      "num_input_tokens_seen": 174618375,
      "step": 8089,
      "time_per_iteration": 2.5140490531921387
    },
    {
      "auxiliary_loss_clip": 0.0112053,
      "auxiliary_loss_mlp": 0.0102358,
      "balance_loss_clip": 1.04447043,
      "balance_loss_mlp": 1.01606369,
      "epoch": 0.9727649852702459,
      "flos": 26541648639360.0,
      "grad_norm": 1.7699347020096867,
      "language_loss": 0.80101824,
      "learning_rate": 7.743187110833105e-09,
      "loss": 0.82245934,
      "num_input_tokens_seen": 174641135,
      "step": 8090,
      "time_per_iteration": 2.5541646480560303
    },
    {
      "auxiliary_loss_clip": 0.01138747,
      "auxiliary_loss_mlp": 0.01019088,
      "balance_loss_clip": 1.04279661,
      "balance_loss_mlp": 1.01243615,
      "epoch": 0.972885228160885,
      "flos": 20522338277760.0,
      "grad_norm": 1.4327316647404809,
      "language_loss": 0.80602801,
      "learning_rate": 7.674858695022602e-09,
      "loss": 0.82760644,
      "num_input_tokens_seen": 174659490,
      "step": 8091,
      "time_per_iteration": 2.4772720336914062
    },
    {
      "auxiliary_loss_clip": 0.01167419,
      "auxiliary_loss_mlp": 0.01025288,
      "balance_loss_clip": 1.04871655,
      "balance_loss_mlp": 1.0179683,
      "epoch": 0.9730054710515241,
      "flos": 17566064196480.0,
      "grad_norm": 4.171080513387283,
      "language_loss": 0.75368786,
      "learning_rate": 7.606832513351591e-09,
      "loss": 0.77561498,
      "num_input_tokens_seen": 174677440,
      "step": 8092,
      "time_per_iteration": 2.4062042236328125
    },
    {
      "auxiliary_loss_clip": 0.01060855,
      "auxiliary_loss_mlp": 0.00752819,
      "balance_loss_clip": 1.00727057,
      "balance_loss_mlp": 0.99990046,
      "epoch": 0.9731257139421632,
      "flos": 68972010117120.0,
      "grad_norm": 0.8387898086433941,
      "language_loss": 0.63925081,
      "learning_rate": 7.539108576140264e-09,
      "loss": 0.65738755,
      "num_input_tokens_seen": 174741550,
      "step": 8093,
      "time_per_iteration": 3.9284298419952393
    },
    {
      "auxiliary_loss_clip": 0.01108176,
      "auxiliary_loss_mlp": 0.01019434,
      "balance_loss_clip": 1.04169524,
      "balance_loss_mlp": 1.01318407,
      "epoch": 0.9732459568328022,
      "flos": 18478841633280.0,
      "grad_norm": 7.059130983022169,
      "language_loss": 0.70622194,
      "learning_rate": 7.471686893661732e-09,
      "loss": 0.72749805,
      "num_input_tokens_seen": 174759845,
      "step": 8094,
      "time_per_iteration": 3.3531012535095215
    },
    {
      "auxiliary_loss_clip": 0.01133073,
      "auxiliary_loss_mlp": 0.01024277,
      "balance_loss_clip": 1.04347253,
      "balance_loss_mlp": 1.01741314,
      "epoch": 0.9733661997234414,
      "flos": 20883886623360.0,
      "grad_norm": 1.8451418180629455,
      "language_loss": 0.64421296,
      "learning_rate": 7.4045674761442636e-09,
      "loss": 0.6657865,
      "num_input_tokens_seen": 174777175,
      "step": 8095,
      "time_per_iteration": 2.484976291656494
    },
    {
      "auxiliary_loss_clip": 0.01163274,
      "auxiliary_loss_mlp": 0.00761438,
      "balance_loss_clip": 1.0464251,
      "balance_loss_mlp": 1.00037682,
      "epoch": 0.9734864426140805,
      "flos": 23766795175680.0,
      "grad_norm": 2.013372456560244,
      "language_loss": 0.74053681,
      "learning_rate": 7.337750333769488e-09,
      "loss": 0.75978392,
      "num_input_tokens_seen": 174796980,
      "step": 8096,
      "time_per_iteration": 3.1700375080108643
    },
    {
      "auxiliary_loss_clip": 0.01141496,
      "auxiliary_loss_mlp": 0.01023873,
      "balance_loss_clip": 1.04051638,
      "balance_loss_mlp": 1.01596034,
      "epoch": 0.9736066855047195,
      "flos": 35042422176000.0,
      "grad_norm": 1.8499675575263224,
      "language_loss": 0.72811949,
      "learning_rate": 7.2712354766737425e-09,
      "loss": 0.7497732,
      "num_input_tokens_seen": 174817310,
      "step": 8097,
      "time_per_iteration": 2.590702533721924
    },
    {
      "auxiliary_loss_clip": 0.01118651,
      "auxiliary_loss_mlp": 0.01025073,
      "balance_loss_clip": 1.04600024,
      "balance_loss_mlp": 1.01694345,
      "epoch": 0.9737269283953586,
      "flos": 20410620001920.0,
      "grad_norm": 1.5692223122004738,
      "language_loss": 0.80446506,
      "learning_rate": 7.2050229149469565e-09,
      "loss": 0.82590234,
      "num_input_tokens_seen": 174837320,
      "step": 8098,
      "time_per_iteration": 2.4967474937438965
    },
    {
      "auxiliary_loss_clip": 0.01125675,
      "auxiliary_loss_mlp": 0.01024111,
      "balance_loss_clip": 1.04091692,
      "balance_loss_mlp": 1.01717913,
      "epoch": 0.9738471712859977,
      "flos": 28911680847360.0,
      "grad_norm": 1.9796723327975578,
      "language_loss": 0.6377998,
      "learning_rate": 7.139112658633984e-09,
      "loss": 0.6592977,
      "num_input_tokens_seen": 174857470,
      "step": 8099,
      "time_per_iteration": 2.5598456859588623
    },
    {
      "auxiliary_loss_clip": 0.01122661,
      "auxiliary_loss_mlp": 0.01022095,
      "balance_loss_clip": 1.04465377,
      "balance_loss_mlp": 1.01537728,
      "epoch": 0.9739674141766368,
      "flos": 27782326356480.0,
      "grad_norm": 1.9990668795700717,
      "language_loss": 0.70322263,
      "learning_rate": 7.073504717733048e-09,
      "loss": 0.72467023,
      "num_input_tokens_seen": 174877035,
      "step": 8100,
      "time_per_iteration": 2.5454845428466797
    },
    {
      "auxiliary_loss_clip": 0.0101923,
      "auxiliary_loss_mlp": 0.01002455,
      "balance_loss_clip": 1.00994813,
      "balance_loss_mlp": 1.00165641,
      "epoch": 0.9740876570672758,
      "flos": 68863057188480.0,
      "grad_norm": 0.7397545835754016,
      "language_loss": 0.57183361,
      "learning_rate": 7.008199102196855e-09,
      "loss": 0.59205049,
      "num_input_tokens_seen": 174938460,
      "step": 8101,
      "time_per_iteration": 3.064176321029663
    },
    {
      "auxiliary_loss_clip": 0.01034016,
      "auxiliary_loss_mlp": 0.01002128,
      "balance_loss_clip": 1.00816333,
      "balance_loss_mlp": 1.00116813,
      "epoch": 0.974207899957915,
      "flos": 58236622646400.0,
      "grad_norm": 0.8218522176226631,
      "language_loss": 0.58975828,
      "learning_rate": 6.9431958219321464e-09,
      "loss": 0.6101197,
      "num_input_tokens_seen": 174994625,
      "step": 8102,
      "time_per_iteration": 3.7765049934387207
    },
    {
      "auxiliary_loss_clip": 0.01136935,
      "auxiliary_loss_mlp": 0.01024942,
      "balance_loss_clip": 1.04401708,
      "balance_loss_mlp": 1.01757753,
      "epoch": 0.9743281428485541,
      "flos": 22600057605120.0,
      "grad_norm": 1.492971481254201,
      "language_loss": 0.77770996,
      "learning_rate": 6.878494886800146e-09,
      "loss": 0.79932868,
      "num_input_tokens_seen": 175015400,
      "step": 8103,
      "time_per_iteration": 2.4827070236206055
    },
    {
      "auxiliary_loss_clip": 0.01138482,
      "auxiliary_loss_mlp": 0.01021934,
      "balance_loss_clip": 1.04606843,
      "balance_loss_mlp": 1.01478493,
      "epoch": 0.9744483857391931,
      "flos": 20008815488640.0,
      "grad_norm": 2.1230928173914614,
      "language_loss": 0.76486599,
      "learning_rate": 6.814096306615669e-09,
      "loss": 0.78647017,
      "num_input_tokens_seen": 175033540,
      "step": 8104,
      "time_per_iteration": 2.4556937217712402
    },
    {
      "auxiliary_loss_clip": 0.01142051,
      "auxiliary_loss_mlp": 0.01024623,
      "balance_loss_clip": 1.04281402,
      "balance_loss_mlp": 1.01721144,
      "epoch": 0.9745686286298323,
      "flos": 17675268520320.0,
      "grad_norm": 2.080499863953652,
      "language_loss": 0.65418673,
      "learning_rate": 6.750000091148011e-09,
      "loss": 0.67585349,
      "num_input_tokens_seen": 175050835,
      "step": 8105,
      "time_per_iteration": 2.442660331726074
    },
    {
      "auxiliary_loss_clip": 0.01165537,
      "auxiliary_loss_mlp": 0.01024805,
      "balance_loss_clip": 1.04791081,
      "balance_loss_mlp": 1.01741982,
      "epoch": 0.9746888715204713,
      "flos": 29460252332160.0,
      "grad_norm": 3.059365685756993,
      "language_loss": 0.72483301,
      "learning_rate": 6.686206250120729e-09,
      "loss": 0.74673647,
      "num_input_tokens_seen": 175072330,
      "step": 8106,
      "time_per_iteration": 2.4667716026306152
    },
    {
      "auxiliary_loss_clip": 0.01128273,
      "auxiliary_loss_mlp": 0.0101988,
      "balance_loss_clip": 1.04172897,
      "balance_loss_mlp": 1.01302528,
      "epoch": 0.9748091144111104,
      "flos": 18479308510080.0,
      "grad_norm": 2.012738753033083,
      "language_loss": 0.74808598,
      "learning_rate": 6.622714793210749e-09,
      "loss": 0.76956749,
      "num_input_tokens_seen": 175091250,
      "step": 8107,
      "time_per_iteration": 2.486872434616089
    },
    {
      "auxiliary_loss_clip": 0.01165791,
      "auxiliary_loss_mlp": 0.01021233,
      "balance_loss_clip": 1.04698598,
      "balance_loss_mlp": 1.01446748,
      "epoch": 0.9749293573017496,
      "flos": 20665154753280.0,
      "grad_norm": 1.6697969042145406,
      "language_loss": 0.79010618,
      "learning_rate": 6.559525730050364e-09,
      "loss": 0.81197643,
      "num_input_tokens_seen": 175111350,
      "step": 8108,
      "time_per_iteration": 2.4031989574432373
    },
    {
      "auxiliary_loss_clip": 0.01126333,
      "auxiliary_loss_mlp": 0.01021914,
      "balance_loss_clip": 1.04463696,
      "balance_loss_mlp": 1.01506567,
      "epoch": 0.9750496001923886,
      "flos": 18478590238080.0,
      "grad_norm": 2.100066966205352,
      "language_loss": 0.75925088,
      "learning_rate": 6.496639070224574e-09,
      "loss": 0.78073335,
      "num_input_tokens_seen": 175129835,
      "step": 8109,
      "time_per_iteration": 2.4959845542907715
    },
    {
      "auxiliary_loss_clip": 0.01154928,
      "auxiliary_loss_mlp": 0.01022647,
      "balance_loss_clip": 1.04728138,
      "balance_loss_mlp": 1.01635861,
      "epoch": 0.9751698430830277,
      "flos": 19572967860480.0,
      "grad_norm": 2.285181785990845,
      "language_loss": 0.84143889,
      "learning_rate": 6.4340548232739714e-09,
      "loss": 0.86321467,
      "num_input_tokens_seen": 175146035,
      "step": 8110,
      "time_per_iteration": 2.4160165786743164
    },
    {
      "auxiliary_loss_clip": 0.01127199,
      "auxiliary_loss_mlp": 0.01023324,
      "balance_loss_clip": 1.0425216,
      "balance_loss_mlp": 1.01586461,
      "epoch": 0.9752900859736668,
      "flos": 23550325862400.0,
      "grad_norm": 2.7302618905670695,
      "language_loss": 0.79359519,
      "learning_rate": 6.371772998692071e-09,
      "loss": 0.81510043,
      "num_input_tokens_seen": 175165290,
      "step": 8111,
      "time_per_iteration": 2.5457630157470703
    },
    {
      "auxiliary_loss_clip": 0.01128015,
      "auxiliary_loss_mlp": 0.01020181,
      "balance_loss_clip": 1.04192674,
      "balance_loss_mlp": 1.01324296,
      "epoch": 0.9754103288643059,
      "flos": 20303211358080.0,
      "grad_norm": 6.202481461481704,
      "language_loss": 0.64976633,
      "learning_rate": 6.309793605927094e-09,
      "loss": 0.67124832,
      "num_input_tokens_seen": 175183610,
      "step": 8112,
      "time_per_iteration": 2.5083487033843994
    },
    {
      "auxiliary_loss_clip": 0.01141585,
      "auxiliary_loss_mlp": 0.01021796,
      "balance_loss_clip": 1.04339075,
      "balance_loss_mlp": 1.0145781,
      "epoch": 0.975530571754945,
      "flos": 19350680544000.0,
      "grad_norm": 1.6971518476172494,
      "language_loss": 0.80095458,
      "learning_rate": 6.248116654381297e-09,
      "loss": 0.82258844,
      "num_input_tokens_seen": 175202080,
      "step": 8113,
      "time_per_iteration": 2.4608726501464844
    },
    {
      "auxiliary_loss_clip": 0.01137997,
      "auxiliary_loss_mlp": 0.01022572,
      "balance_loss_clip": 1.04163122,
      "balance_loss_mlp": 1.01568747,
      "epoch": 0.9756508146455841,
      "flos": 23583399310080.0,
      "grad_norm": 1.669579803985765,
      "language_loss": 0.72641063,
      "learning_rate": 6.186742153410751e-09,
      "loss": 0.74801636,
      "num_input_tokens_seen": 175221575,
      "step": 8114,
      "time_per_iteration": 2.497535228729248
    },
    {
      "auxiliary_loss_clip": 0.01136944,
      "auxiliary_loss_mlp": 0.01028302,
      "balance_loss_clip": 1.04501879,
      "balance_loss_mlp": 1.02047586,
      "epoch": 0.9757710575362232,
      "flos": 22966921163520.0,
      "grad_norm": 2.371762953044903,
      "language_loss": 0.87399918,
      "learning_rate": 6.125670112326453e-09,
      "loss": 0.89565158,
      "num_input_tokens_seen": 175240835,
      "step": 8115,
      "time_per_iteration": 2.492361068725586
    },
    {
      "auxiliary_loss_clip": 0.01150661,
      "auxiliary_loss_mlp": 0.01023932,
      "balance_loss_clip": 1.04295027,
      "balance_loss_mlp": 1.01676512,
      "epoch": 0.9758913004268622,
      "flos": 27966009530880.0,
      "grad_norm": 1.6151264758519213,
      "language_loss": 0.70200801,
      "learning_rate": 6.064900540392548e-09,
      "loss": 0.72375393,
      "num_input_tokens_seen": 175262930,
      "step": 8116,
      "time_per_iteration": 2.5042977333068848
    },
    {
      "auxiliary_loss_clip": 0.01131119,
      "auxiliary_loss_mlp": 0.0102034,
      "balance_loss_clip": 1.0444541,
      "balance_loss_mlp": 1.0135746,
      "epoch": 0.9760115433175014,
      "flos": 22200156512640.0,
      "grad_norm": 1.9972200295645908,
      "language_loss": 0.78850317,
      "learning_rate": 6.0044334468278835e-09,
      "loss": 0.81001776,
      "num_input_tokens_seen": 175282275,
      "step": 8117,
      "time_per_iteration": 2.4768154621124268
    },
    {
      "auxiliary_loss_clip": 0.01112618,
      "auxiliary_loss_mlp": 0.01027179,
      "balance_loss_clip": 1.04240966,
      "balance_loss_mlp": 1.01954341,
      "epoch": 0.9761317862081405,
      "flos": 26250736389120.0,
      "grad_norm": 1.7051864042213905,
      "language_loss": 0.71447563,
      "learning_rate": 5.944268840805345e-09,
      "loss": 0.73587358,
      "num_input_tokens_seen": 175303020,
      "step": 8118,
      "time_per_iteration": 2.5904552936553955
    },
    {
      "auxiliary_loss_clip": 0.01114685,
      "auxiliary_loss_mlp": 0.01022792,
      "balance_loss_clip": 1.03993487,
      "balance_loss_mlp": 1.01632166,
      "epoch": 0.9762520290987795,
      "flos": 26575440359040.0,
      "grad_norm": 3.337457256799733,
      "language_loss": 0.63808692,
      "learning_rate": 5.88440673145163e-09,
      "loss": 0.65946174,
      "num_input_tokens_seen": 175324070,
      "step": 8119,
      "time_per_iteration": 3.382863759994507
    },
    {
      "auxiliary_loss_clip": 0.01151053,
      "auxiliary_loss_mlp": 0.01024807,
      "balance_loss_clip": 1.04823482,
      "balance_loss_mlp": 1.01774132,
      "epoch": 0.9763722719894187,
      "flos": 18005036307840.0,
      "grad_norm": 2.106207766736534,
      "language_loss": 0.82200015,
      "learning_rate": 5.824847127848142e-09,
      "loss": 0.84375882,
      "num_input_tokens_seen": 175342595,
      "step": 8120,
      "time_per_iteration": 2.4163408279418945
    },
    {
      "auxiliary_loss_clip": 0.01111372,
      "auxiliary_loss_mlp": 0.01020143,
      "balance_loss_clip": 1.04342175,
      "balance_loss_mlp": 1.01313317,
      "epoch": 0.9764925148800577,
      "flos": 22455660931200.0,
      "grad_norm": 2.017126212230986,
      "language_loss": 0.7852121,
      "learning_rate": 5.765590039029433e-09,
      "loss": 0.80652726,
      "num_input_tokens_seen": 175361915,
      "step": 8121,
      "time_per_iteration": 3.365196704864502
    },
    {
      "auxiliary_loss_clip": 0.01163146,
      "auxiliary_loss_mlp": 0.01025189,
      "balance_loss_clip": 1.04720449,
      "balance_loss_mlp": 1.01806903,
      "epoch": 0.9766127577706968,
      "flos": 36757084786560.0,
      "grad_norm": 1.5228147647626102,
      "language_loss": 0.7102589,
      "learning_rate": 5.706635473985422e-09,
      "loss": 0.73214221,
      "num_input_tokens_seen": 175385785,
      "step": 8122,
      "time_per_iteration": 2.569197416305542
    },
    {
      "auxiliary_loss_clip": 0.01147091,
      "auxiliary_loss_mlp": 0.01023305,
      "balance_loss_clip": 1.04369509,
      "balance_loss_mlp": 1.01640034,
      "epoch": 0.976733000661336,
      "flos": 22309971367680.0,
      "grad_norm": 1.778794603266535,
      "language_loss": 0.85213721,
      "learning_rate": 5.6479834416591764e-09,
      "loss": 0.87384117,
      "num_input_tokens_seen": 175405145,
      "step": 8123,
      "time_per_iteration": 3.242032051086426
    },
    {
      "auxiliary_loss_clip": 0.01151754,
      "auxiliary_loss_mlp": 0.00762177,
      "balance_loss_clip": 1.04720449,
      "balance_loss_mlp": 1.00044072,
      "epoch": 0.976853243551975,
      "flos": 25810938264960.0,
      "grad_norm": 4.051246936015088,
      "language_loss": 0.68507946,
      "learning_rate": 5.589633950947803e-09,
      "loss": 0.70421875,
      "num_input_tokens_seen": 175422645,
      "step": 8124,
      "time_per_iteration": 2.4661929607391357
    },
    {
      "auxiliary_loss_clip": 0.01136933,
      "auxiliary_loss_mlp": 0.01027225,
      "balance_loss_clip": 1.04482484,
      "balance_loss_mlp": 1.01977992,
      "epoch": 0.9769734864426141,
      "flos": 21397445326080.0,
      "grad_norm": 2.0128132042892366,
      "language_loss": 0.69590014,
      "learning_rate": 5.5315870107035535e-09,
      "loss": 0.71754169,
      "num_input_tokens_seen": 175440695,
      "step": 8125,
      "time_per_iteration": 2.4700679779052734
    },
    {
      "auxiliary_loss_clip": 0.01136227,
      "auxiliary_loss_mlp": 0.01025401,
      "balance_loss_clip": 1.04820156,
      "balance_loss_mlp": 1.01805735,
      "epoch": 0.9770937293332532,
      "flos": 13990977584640.0,
      "grad_norm": 1.7515072225288109,
      "language_loss": 0.78687161,
      "learning_rate": 5.473842629731607e-09,
      "loss": 0.80848789,
      "num_input_tokens_seen": 175459195,
      "step": 8126,
      "time_per_iteration": 2.467736005783081
    },
    {
      "auxiliary_loss_clip": 0.01144096,
      "auxiliary_loss_mlp": 0.0076178,
      "balance_loss_clip": 1.04396796,
      "balance_loss_mlp": 1.00050998,
      "epoch": 0.9772139722238923,
      "flos": 17931994001280.0,
      "grad_norm": 1.9751408344668755,
      "language_loss": 0.7812764,
      "learning_rate": 5.416400816792066e-09,
      "loss": 0.80033517,
      "num_input_tokens_seen": 175476710,
      "step": 8127,
      "time_per_iteration": 2.4525721073150635
    },
    {
      "auxiliary_loss_clip": 0.01162821,
      "auxiliary_loss_mlp": 0.01020739,
      "balance_loss_clip": 1.04615641,
      "balance_loss_mlp": 1.01363122,
      "epoch": 0.9773342151145313,
      "flos": 20446171488000.0,
      "grad_norm": 2.442607377215319,
      "language_loss": 0.78262478,
      "learning_rate": 5.359261580598407e-09,
      "loss": 0.80446041,
      "num_input_tokens_seen": 175492550,
      "step": 8128,
      "time_per_iteration": 2.3831400871276855
    },
    {
      "auxiliary_loss_clip": 0.01154127,
      "auxiliary_loss_mlp": 0.01023394,
      "balance_loss_clip": 1.04757047,
      "balance_loss_mlp": 1.01586878,
      "epoch": 0.9774544580051704,
      "flos": 11837306949120.0,
      "grad_norm": 2.3653080014836827,
      "language_loss": 0.77689081,
      "learning_rate": 5.302424929819027e-09,
      "loss": 0.798666,
      "num_input_tokens_seen": 175506560,
      "step": 8129,
      "time_per_iteration": 3.1712870597839355
    },
    {
      "auxiliary_loss_clip": 0.01154645,
      "auxiliary_loss_mlp": 0.01024732,
      "balance_loss_clip": 1.04369283,
      "balance_loss_mlp": 1.0176394,
      "epoch": 0.9775747008958096,
      "flos": 13479932833920.0,
      "grad_norm": 2.408888524014105,
      "language_loss": 0.73017269,
      "learning_rate": 5.24589087307592e-09,
      "loss": 0.75196648,
      "num_input_tokens_seen": 175524180,
      "step": 8130,
      "time_per_iteration": 2.410435914993286
    },
    {
      "auxiliary_loss_clip": 0.01165728,
      "auxiliary_loss_mlp": 0.01025072,
      "balance_loss_clip": 1.04668736,
      "balance_loss_mlp": 1.01816738,
      "epoch": 0.9776949437864486,
      "flos": 59532314042880.0,
      "grad_norm": 2.011909747498914,
      "language_loss": 0.64827871,
      "learning_rate": 5.189659418944891e-09,
      "loss": 0.67018676,
      "num_input_tokens_seen": 175554355,
      "step": 8131,
      "time_per_iteration": 2.8076958656311035
    },
    {
      "auxiliary_loss_clip": 0.01165473,
      "auxiliary_loss_mlp": 0.01026967,
      "balance_loss_clip": 1.04889047,
      "balance_loss_mlp": 1.02024364,
      "epoch": 0.9778151866770877,
      "flos": 21178605715200.0,
      "grad_norm": 2.1845730907548653,
      "language_loss": 0.78398108,
      "learning_rate": 5.133730575956674e-09,
      "loss": 0.80590552,
      "num_input_tokens_seen": 175574025,
      "step": 8132,
      "time_per_iteration": 2.3939788341522217
    },
    {
      "auxiliary_loss_clip": 0.01139669,
      "auxiliary_loss_mlp": 0.01019615,
      "balance_loss_clip": 1.04460788,
      "balance_loss_mlp": 1.01264966,
      "epoch": 0.9779354295677268,
      "flos": 20886795624960.0,
      "grad_norm": 1.8914406519125866,
      "language_loss": 0.7174443,
      "learning_rate": 5.0781043525953696e-09,
      "loss": 0.73903716,
      "num_input_tokens_seen": 175592090,
      "step": 8133,
      "time_per_iteration": 2.4702401161193848
    },
    {
      "auxiliary_loss_clip": 0.01131888,
      "auxiliary_loss_mlp": 0.01027151,
      "balance_loss_clip": 1.04608107,
      "balance_loss_mlp": 1.01998973,
      "epoch": 0.9780556724583659,
      "flos": 23440618748160.0,
      "grad_norm": 1.7668006370257898,
      "language_loss": 0.73815346,
      "learning_rate": 5.0227807572995605e-09,
      "loss": 0.75974387,
      "num_input_tokens_seen": 175614065,
      "step": 8134,
      "time_per_iteration": 2.4858009815216064
    },
    {
      "auxiliary_loss_clip": 0.01138725,
      "auxiliary_loss_mlp": 0.01019099,
      "balance_loss_clip": 1.04271317,
      "balance_loss_mlp": 1.01222396,
      "epoch": 0.9781759153490049,
      "flos": 20923244951040.0,
      "grad_norm": 2.150014382063035,
      "language_loss": 0.66863084,
      "learning_rate": 4.967759798461646e-09,
      "loss": 0.69020909,
      "num_input_tokens_seen": 175632410,
      "step": 8135,
      "time_per_iteration": 2.4722399711608887
    },
    {
      "auxiliary_loss_clip": 0.01163006,
      "auxiliary_loss_mlp": 0.01021028,
      "balance_loss_clip": 1.0472796,
      "balance_loss_mlp": 1.01430786,
      "epoch": 0.9782961582396441,
      "flos": 28293191539200.0,
      "grad_norm": 1.9720769277041839,
      "language_loss": 0.74722046,
      "learning_rate": 4.913041484428282e-09,
      "loss": 0.76906079,
      "num_input_tokens_seen": 175652885,
      "step": 8136,
      "time_per_iteration": 2.4538471698760986
    },
    {
      "auxiliary_loss_clip": 0.011533,
      "auxiliary_loss_mlp": 0.01021275,
      "balance_loss_clip": 1.04742908,
      "balance_loss_mlp": 1.01448333,
      "epoch": 0.9784164011302832,
      "flos": 25552955808000.0,
      "grad_norm": 1.8485643289530442,
      "language_loss": 0.74049205,
      "learning_rate": 4.858625823500384e-09,
      "loss": 0.76223785,
      "num_input_tokens_seen": 175670585,
      "step": 8137,
      "time_per_iteration": 2.462751626968384
    },
    {
      "auxiliary_loss_clip": 0.01153439,
      "auxiliary_loss_mlp": 0.01025347,
      "balance_loss_clip": 1.04546785,
      "balance_loss_mlp": 1.01767635,
      "epoch": 0.9785366440209222,
      "flos": 29965945956480.0,
      "grad_norm": 1.8551775818326384,
      "language_loss": 0.73223931,
      "learning_rate": 4.80451282393246e-09,
      "loss": 0.75402713,
      "num_input_tokens_seen": 175690570,
      "step": 8138,
      "time_per_iteration": 2.526431083679199
    },
    {
      "auxiliary_loss_clip": 0.01135768,
      "auxiliary_loss_mlp": 0.01019203,
      "balance_loss_clip": 1.04343545,
      "balance_loss_mlp": 1.01248288,
      "epoch": 0.9786568869115614,
      "flos": 32343591847680.0,
      "grad_norm": 1.8344639634645163,
      "language_loss": 0.6755361,
      "learning_rate": 4.750702493933722e-09,
      "loss": 0.69708586,
      "num_input_tokens_seen": 175710455,
      "step": 8139,
      "time_per_iteration": 2.5411572456359863
    },
    {
      "auxiliary_loss_clip": 0.0113968,
      "auxiliary_loss_mlp": 0.0076186,
      "balance_loss_clip": 1.04781199,
      "balance_loss_mlp": 1.0004015,
      "epoch": 0.9787771298022004,
      "flos": 23331414424320.0,
      "grad_norm": 1.8296538899820265,
      "language_loss": 0.85251313,
      "learning_rate": 4.697194841666974e-09,
      "loss": 0.87152851,
      "num_input_tokens_seen": 175729380,
      "step": 8140,
      "time_per_iteration": 2.482389211654663
    },
    {
      "auxiliary_loss_clip": 0.01153549,
      "auxiliary_loss_mlp": 0.01026866,
      "balance_loss_clip": 1.04599369,
      "balance_loss_mlp": 1.01856875,
      "epoch": 0.9788973726928395,
      "flos": 21468548298240.0,
      "grad_norm": 2.3121388636969593,
      "language_loss": 0.81939673,
      "learning_rate": 4.6439898752492764e-09,
      "loss": 0.84120089,
      "num_input_tokens_seen": 175749520,
      "step": 8141,
      "time_per_iteration": 2.4324891567230225
    },
    {
      "auxiliary_loss_clip": 0.01052235,
      "auxiliary_loss_mlp": 0.00752834,
      "balance_loss_clip": 1.00818765,
      "balance_loss_mlp": 0.99988747,
      "epoch": 0.9790176155834787,
      "flos": 68897459439360.0,
      "grad_norm": 0.747806976167841,
      "language_loss": 0.63677412,
      "learning_rate": 4.591087602751731e-09,
      "loss": 0.65482479,
      "num_input_tokens_seen": 175811380,
      "step": 8142,
      "time_per_iteration": 3.103113889694214
    },
    {
      "auxiliary_loss_clip": 0.01151411,
      "auxiliary_loss_mlp": 0.01025138,
      "balance_loss_clip": 1.04546499,
      "balance_loss_mlp": 1.0182898,
      "epoch": 0.9791378584741177,
      "flos": 21430877909760.0,
      "grad_norm": 1.6713511878043568,
      "language_loss": 0.71668696,
      "learning_rate": 4.538488032199916e-09,
      "loss": 0.73845243,
      "num_input_tokens_seen": 175829480,
      "step": 8143,
      "time_per_iteration": 2.4319615364074707
    },
    {
      "auxiliary_loss_clip": 0.01153742,
      "auxiliary_loss_mlp": 0.01028185,
      "balance_loss_clip": 1.04391646,
      "balance_loss_mlp": 1.02075255,
      "epoch": 0.9792581013647568,
      "flos": 20153032594560.0,
      "grad_norm": 1.9393341764512604,
      "language_loss": 0.68740761,
      "learning_rate": 4.486191171572784e-09,
      "loss": 0.70922685,
      "num_input_tokens_seen": 175846750,
      "step": 8144,
      "time_per_iteration": 2.4169678688049316
    },
    {
      "auxiliary_loss_clip": 0.01153707,
      "auxiliary_loss_mlp": 0.01021738,
      "balance_loss_clip": 1.04774129,
      "balance_loss_mlp": 1.01483262,
      "epoch": 0.9793783442553959,
      "flos": 23728191033600.0,
      "grad_norm": 1.6116398258788225,
      "language_loss": 0.77950466,
      "learning_rate": 4.434197028803766e-09,
      "loss": 0.8012591,
      "num_input_tokens_seen": 175865975,
      "step": 8145,
      "time_per_iteration": 2.491436243057251
    },
    {
      "auxiliary_loss_clip": 0.0112927,
      "auxiliary_loss_mlp": 0.01026079,
      "balance_loss_clip": 1.04270816,
      "balance_loss_mlp": 1.01811552,
      "epoch": 0.979498587146035,
      "flos": 23038742407680.0,
      "grad_norm": 2.120129183388703,
      "language_loss": 0.81977075,
      "learning_rate": 4.3825056117805514e-09,
      "loss": 0.84132421,
      "num_input_tokens_seen": 175881860,
      "step": 8146,
      "time_per_iteration": 3.350921630859375
    },
    {
      "auxiliary_loss_clip": 0.01164236,
      "auxiliary_loss_mlp": 0.01017539,
      "balance_loss_clip": 1.04483521,
      "balance_loss_mlp": 1.01034176,
      "epoch": 0.979618830036674,
      "flos": 14318841951360.0,
      "grad_norm": 3.371658584094465,
      "language_loss": 0.79727459,
      "learning_rate": 4.331116928344425e-09,
      "loss": 0.81909239,
      "num_input_tokens_seen": 175898175,
      "step": 8147,
      "time_per_iteration": 2.37263560295105
    },
    {
      "auxiliary_loss_clip": 0.01140426,
      "auxiliary_loss_mlp": 0.00761766,
      "balance_loss_clip": 1.04245234,
      "balance_loss_mlp": 1.00039959,
      "epoch": 0.9797390729273132,
      "flos": 16727514215040.0,
      "grad_norm": 2.061560084937971,
      "language_loss": 0.62689418,
      "learning_rate": 4.28003098629115e-09,
      "loss": 0.6459161,
      "num_input_tokens_seen": 175914310,
      "step": 8148,
      "time_per_iteration": 3.3335866928100586
    },
    {
      "auxiliary_loss_clip": 0.01120241,
      "auxiliary_loss_mlp": 0.01018853,
      "balance_loss_clip": 1.03794074,
      "balance_loss_mlp": 1.01156902,
      "epoch": 0.9798593158179523,
      "flos": 24532661986560.0,
      "grad_norm": 1.8677481259793705,
      "language_loss": 0.78492945,
      "learning_rate": 4.229247793370305e-09,
      "loss": 0.80632043,
      "num_input_tokens_seen": 175933435,
      "step": 8149,
      "time_per_iteration": 2.5447299480438232
    },
    {
      "auxiliary_loss_clip": 0.01168353,
      "auxiliary_loss_mlp": 0.01023926,
      "balance_loss_clip": 1.04859984,
      "balance_loss_mlp": 1.01657724,
      "epoch": 0.9799795587085913,
      "flos": 27308808339840.0,
      "grad_norm": 1.5736061605973455,
      "language_loss": 0.70442384,
      "learning_rate": 4.178767357285951e-09,
      "loss": 0.72634661,
      "num_input_tokens_seen": 175955065,
      "step": 8150,
      "time_per_iteration": 3.272921323776245
    },
    {
      "auxiliary_loss_clip": 0.01151484,
      "auxiliary_loss_mlp": 0.00761873,
      "balance_loss_clip": 1.04627144,
      "balance_loss_mlp": 1.00048232,
      "epoch": 0.9800998015992305,
      "flos": 26286575184000.0,
      "grad_norm": 1.8535681547593335,
      "language_loss": 0.71334445,
      "learning_rate": 4.128589685695516e-09,
      "loss": 0.73247802,
      "num_input_tokens_seen": 175975490,
      "step": 8151,
      "time_per_iteration": 2.495699167251587
    },
    {
      "auxiliary_loss_clip": 0.01164658,
      "auxiliary_loss_mlp": 0.01023643,
      "balance_loss_clip": 1.04765797,
      "balance_loss_mlp": 1.01717293,
      "epoch": 0.9802200444898695,
      "flos": 16723635546240.0,
      "grad_norm": 1.9045521649027974,
      "language_loss": 0.84778726,
      "learning_rate": 4.078714786211135e-09,
      "loss": 0.86967027,
      "num_input_tokens_seen": 175991340,
      "step": 8152,
      "time_per_iteration": 2.3678603172302246
    },
    {
      "auxiliary_loss_clip": 0.01147414,
      "auxiliary_loss_mlp": 0.01019576,
      "balance_loss_clip": 1.04514623,
      "balance_loss_mlp": 1.01288581,
      "epoch": 0.9803402873805086,
      "flos": 24900459298560.0,
      "grad_norm": 1.6416467066632476,
      "language_loss": 0.76966894,
      "learning_rate": 4.029142666398977e-09,
      "loss": 0.79133886,
      "num_input_tokens_seen": 176011505,
      "step": 8153,
      "time_per_iteration": 2.4981791973114014
    },
    {
      "auxiliary_loss_clip": 0.01163146,
      "auxiliary_loss_mlp": 0.01025707,
      "balance_loss_clip": 1.04716539,
      "balance_loss_mlp": 1.01903391,
      "epoch": 0.9804605302711478,
      "flos": 22564937082240.0,
      "grad_norm": 1.767689089059547,
      "language_loss": 0.80228394,
      "learning_rate": 3.979873333778805e-09,
      "loss": 0.8241725,
      "num_input_tokens_seen": 176029680,
      "step": 8154,
      "time_per_iteration": 2.404353141784668
    },
    {
      "auxiliary_loss_clip": 0.01141301,
      "auxiliary_loss_mlp": 0.01025651,
      "balance_loss_clip": 1.0459156,
      "balance_loss_mlp": 1.01857567,
      "epoch": 0.9805807731617868,
      "flos": 38905368382080.0,
      "grad_norm": 1.803885339464823,
      "language_loss": 0.73662245,
      "learning_rate": 3.930906795824862e-09,
      "loss": 0.75829196,
      "num_input_tokens_seen": 176050355,
      "step": 8155,
      "time_per_iteration": 2.6353776454925537
    },
    {
      "auxiliary_loss_clip": 0.0114837,
      "auxiliary_loss_mlp": 0.01019828,
      "balance_loss_clip": 1.04474437,
      "balance_loss_mlp": 1.0128839,
      "epoch": 0.9807010160524259,
      "flos": 17821999578240.0,
      "grad_norm": 2.0487655471343174,
      "language_loss": 0.76738417,
      "learning_rate": 3.882243059965207e-09,
      "loss": 0.78906608,
      "num_input_tokens_seen": 176068070,
      "step": 8156,
      "time_per_iteration": 3.191441297531128
    },
    {
      "auxiliary_loss_clip": 0.0114346,
      "auxiliary_loss_mlp": 0.01021056,
      "balance_loss_clip": 1.04243064,
      "balance_loss_mlp": 1.01410329,
      "epoch": 0.980821258943065,
      "flos": 13552975140480.0,
      "grad_norm": 2.508938664063899,
      "language_loss": 0.66036391,
      "learning_rate": 3.833882133582156e-09,
      "loss": 0.6820091,
      "num_input_tokens_seen": 176083730,
      "step": 8157,
      "time_per_iteration": 2.411853790283203
    },
    {
      "auxiliary_loss_clip": 0.01152159,
      "auxiliary_loss_mlp": 0.01021613,
      "balance_loss_clip": 1.04498196,
      "balance_loss_mlp": 1.01479387,
      "epoch": 0.9809415018337041,
      "flos": 21689794120320.0,
      "grad_norm": 1.8332722094206069,
      "language_loss": 0.78118157,
      "learning_rate": 3.785824024012285e-09,
      "loss": 0.80291933,
      "num_input_tokens_seen": 176102730,
      "step": 8158,
      "time_per_iteration": 2.4413275718688965
    },
    {
      "auxiliary_loss_clip": 0.01129618,
      "auxiliary_loss_mlp": 0.01024136,
      "balance_loss_clip": 1.0465486,
      "balance_loss_mlp": 1.01722431,
      "epoch": 0.9810617447243432,
      "flos": 23294857357440.0,
      "grad_norm": 1.8396009466613221,
      "language_loss": 0.78398681,
      "learning_rate": 3.738068738545541e-09,
      "loss": 0.80552435,
      "num_input_tokens_seen": 176121815,
      "step": 8159,
      "time_per_iteration": 2.4800519943237305
    },
    {
      "auxiliary_loss_clip": 0.01155248,
      "auxiliary_loss_mlp": 0.01029246,
      "balance_loss_clip": 1.0465827,
      "balance_loss_mlp": 1.02221584,
      "epoch": 0.9811819876149822,
      "flos": 18332038748160.0,
      "grad_norm": 3.4786211851779782,
      "language_loss": 0.7845642,
      "learning_rate": 3.6906162844265733e-09,
      "loss": 0.80640912,
      "num_input_tokens_seen": 176138900,
      "step": 8160,
      "time_per_iteration": 2.401216506958008
    },
    {
      "auxiliary_loss_clip": 0.01130173,
      "auxiliary_loss_mlp": 0.01026881,
      "balance_loss_clip": 1.04174781,
      "balance_loss_mlp": 1.01934052,
      "epoch": 0.9813022305056214,
      "flos": 22601961025920.0,
      "grad_norm": 1.9317683172812767,
      "language_loss": 0.70888704,
      "learning_rate": 3.643466668853845e-09,
      "loss": 0.73045754,
      "num_input_tokens_seen": 176156925,
      "step": 8161,
      "time_per_iteration": 2.465182065963745
    },
    {
      "auxiliary_loss_clip": 0.01138091,
      "auxiliary_loss_mlp": 0.01019916,
      "balance_loss_clip": 1.04395604,
      "balance_loss_mlp": 1.01228404,
      "epoch": 0.9814224733962604,
      "flos": 25413335642880.0,
      "grad_norm": 4.715763493552389,
      "language_loss": 0.75653255,
      "learning_rate": 3.59661989898008e-09,
      "loss": 0.77811265,
      "num_input_tokens_seen": 176177980,
      "step": 8162,
      "time_per_iteration": 2.513110637664795
    },
    {
      "auxiliary_loss_clip": 0.01117367,
      "auxiliary_loss_mlp": 0.01024265,
      "balance_loss_clip": 1.04372442,
      "balance_loss_mlp": 1.01737475,
      "epoch": 0.9815427162868995,
      "flos": 25007185584000.0,
      "grad_norm": 1.6625366758979532,
      "language_loss": 0.76566422,
      "learning_rate": 3.5500759819115934e-09,
      "loss": 0.78708053,
      "num_input_tokens_seen": 176198345,
      "step": 8163,
      "time_per_iteration": 2.5160861015319824
    },
    {
      "auxiliary_loss_clip": 0.01166516,
      "auxiliary_loss_mlp": 0.01021826,
      "balance_loss_clip": 1.04857266,
      "balance_loss_mlp": 1.01467061,
      "epoch": 0.9816629591775387,
      "flos": 20662604887680.0,
      "grad_norm": 2.4644786885744705,
      "language_loss": 0.80745482,
      "learning_rate": 3.5038349247094034e-09,
      "loss": 0.82933831,
      "num_input_tokens_seen": 176215605,
      "step": 8164,
      "time_per_iteration": 2.3838396072387695
    },
    {
      "auxiliary_loss_clip": 0.01134614,
      "auxiliary_loss_mlp": 0.0102223,
      "balance_loss_clip": 1.0405457,
      "balance_loss_mlp": 1.01488686,
      "epoch": 0.9817832020681777,
      "flos": 17712220636800.0,
      "grad_norm": 2.0773076081398094,
      "language_loss": 0.77310395,
      "learning_rate": 3.4578967343878994e-09,
      "loss": 0.79467249,
      "num_input_tokens_seen": 176231810,
      "step": 8165,
      "time_per_iteration": 2.4234228134155273
    },
    {
      "auxiliary_loss_clip": 0.01134869,
      "auxiliary_loss_mlp": 0.01023098,
      "balance_loss_clip": 1.04334927,
      "balance_loss_mlp": 1.01638103,
      "epoch": 0.9819034449588168,
      "flos": 22530032040960.0,
      "grad_norm": 1.8230582597765592,
      "language_loss": 0.80955255,
      "learning_rate": 3.4122614179161733e-09,
      "loss": 0.83113223,
      "num_input_tokens_seen": 176251770,
      "step": 8166,
      "time_per_iteration": 2.4711897373199463
    },
    {
      "auxiliary_loss_clip": 0.01113284,
      "auxiliary_loss_mlp": 0.01021099,
      "balance_loss_clip": 1.04131031,
      "balance_loss_mlp": 1.01410198,
      "epoch": 0.9820236878494559,
      "flos": 20011221699840.0,
      "grad_norm": 1.7171763175084334,
      "language_loss": 0.78283048,
      "learning_rate": 3.36692898221691e-09,
      "loss": 0.8041743,
      "num_input_tokens_seen": 176270135,
      "step": 8167,
      "time_per_iteration": 2.4911482334136963
    },
    {
      "auxiliary_loss_clip": 0.01151389,
      "auxiliary_loss_mlp": 0.01023163,
      "balance_loss_clip": 1.0445013,
      "balance_loss_mlp": 1.01653206,
      "epoch": 0.982143930740095,
      "flos": 18807316531200.0,
      "grad_norm": 1.92690498703883,
      "language_loss": 0.73083019,
      "learning_rate": 3.3218994341668305e-09,
      "loss": 0.75257564,
      "num_input_tokens_seen": 176289065,
      "step": 8168,
      "time_per_iteration": 2.413191318511963
    },
    {
      "auxiliary_loss_clip": 0.01164056,
      "auxiliary_loss_mlp": 0.01021881,
      "balance_loss_clip": 1.04855669,
      "balance_loss_mlp": 1.01537204,
      "epoch": 0.982264173630734,
      "flos": 26578026138240.0,
      "grad_norm": 1.5357757619413042,
      "language_loss": 0.75578237,
      "learning_rate": 3.2771727805971373e-09,
      "loss": 0.77764171,
      "num_input_tokens_seen": 176310450,
      "step": 8169,
      "time_per_iteration": 2.4397590160369873
    },
    {
      "auxiliary_loss_clip": 0.01102818,
      "auxiliary_loss_mlp": 0.01020651,
      "balance_loss_clip": 1.03774548,
      "balance_loss_mlp": 1.0137248,
      "epoch": 0.9823844165213732,
      "flos": 22014462176640.0,
      "grad_norm": 1.8060366368054204,
      "language_loss": 0.77218843,
      "learning_rate": 3.232749028292847e-09,
      "loss": 0.79342312,
      "num_input_tokens_seen": 176327415,
      "step": 8170,
      "time_per_iteration": 2.5204882621765137
    },
    {
      "auxiliary_loss_clip": 0.01165422,
      "auxiliary_loss_mlp": 0.01025064,
      "balance_loss_clip": 1.04566145,
      "balance_loss_mlp": 1.01802504,
      "epoch": 0.9825046594120123,
      "flos": 21908166854400.0,
      "grad_norm": 1.9804211903255802,
      "language_loss": 0.88267243,
      "learning_rate": 3.188628183992792e-09,
      "loss": 0.90457731,
      "num_input_tokens_seen": 176347680,
      "step": 8171,
      "time_per_iteration": 2.4180612564086914
    },
    {
      "auxiliary_loss_clip": 0.01051799,
      "auxiliary_loss_mlp": 0.01000784,
      "balance_loss_clip": 1.00713468,
      "balance_loss_mlp": 0.99993753,
      "epoch": 0.9826249023026513,
      "flos": 59494610718720.0,
      "grad_norm": 0.7387942025385362,
      "language_loss": 0.62611729,
      "learning_rate": 3.1448102543902844e-09,
      "loss": 0.64664316,
      "num_input_tokens_seen": 176411595,
      "step": 8172,
      "time_per_iteration": 3.0106372833251953
    },
    {
      "auxiliary_loss_clip": 0.01129078,
      "auxiliary_loss_mlp": 0.01027565,
      "balance_loss_clip": 1.0434792,
      "balance_loss_mlp": 1.02008486,
      "epoch": 0.9827451451932905,
      "flos": 16071031296000.0,
      "grad_norm": 1.821035821082583,
      "language_loss": 0.67597413,
      "learning_rate": 3.1012952461324515e-09,
      "loss": 0.69754052,
      "num_input_tokens_seen": 176430570,
      "step": 8173,
      "time_per_iteration": 3.326709270477295
    },
    {
      "auxiliary_loss_clip": 0.01148025,
      "auxiliary_loss_mlp": 0.01026059,
      "balance_loss_clip": 1.04706907,
      "balance_loss_mlp": 1.01877487,
      "epoch": 0.9828653880839295,
      "flos": 20262775622400.0,
      "grad_norm": 3.6820413674070247,
      "language_loss": 0.73906165,
      "learning_rate": 3.0580831658204575e-09,
      "loss": 0.76080251,
      "num_input_tokens_seen": 176448150,
      "step": 8174,
      "time_per_iteration": 2.43330454826355
    },
    {
      "auxiliary_loss_clip": 0.01149705,
      "auxiliary_loss_mlp": 0.01021411,
      "balance_loss_clip": 1.047019,
      "balance_loss_mlp": 1.0150249,
      "epoch": 0.9829856309745686,
      "flos": 21616141282560.0,
      "grad_norm": 1.5922564052528934,
      "language_loss": 0.77869344,
      "learning_rate": 3.015174020009281e-09,
      "loss": 0.80040467,
      "num_input_tokens_seen": 176467475,
      "step": 8175,
      "time_per_iteration": 3.263957977294922
    },
    {
      "auxiliary_loss_clip": 0.01126519,
      "auxiliary_loss_mlp": 0.01023605,
      "balance_loss_clip": 1.04298818,
      "balance_loss_mlp": 1.01717329,
      "epoch": 0.9831058738652078,
      "flos": 23764209396480.0,
      "grad_norm": 2.0496633053446307,
      "language_loss": 0.74972546,
      "learning_rate": 2.9725678152086043e-09,
      "loss": 0.77122664,
      "num_input_tokens_seen": 176486045,
      "step": 8176,
      "time_per_iteration": 3.312959909439087
    },
    {
      "auxiliary_loss_clip": 0.01123988,
      "auxiliary_loss_mlp": 0.01022397,
      "balance_loss_clip": 1.0419023,
      "balance_loss_mlp": 1.0152328,
      "epoch": 0.9832261167558468,
      "flos": 11320911072000.0,
      "grad_norm": 2.5278741537232836,
      "language_loss": 0.82737422,
      "learning_rate": 2.930264557881257e-09,
      "loss": 0.84883809,
      "num_input_tokens_seen": 176501230,
      "step": 8177,
      "time_per_iteration": 2.418728828430176
    },
    {
      "auxiliary_loss_clip": 0.0106086,
      "auxiliary_loss_mlp": 0.01000673,
      "balance_loss_clip": 1.0073384,
      "balance_loss_mlp": 0.99982661,
      "epoch": 0.9833463596464859,
      "flos": 60000304343040.0,
      "grad_norm": 0.8349583636822246,
      "language_loss": 0.58210409,
      "learning_rate": 2.8882642544452163e-09,
      "loss": 0.60271943,
      "num_input_tokens_seen": 176565955,
      "step": 8178,
      "time_per_iteration": 3.0307512283325195
    },
    {
      "auxiliary_loss_clip": 0.01125384,
      "auxiliary_loss_mlp": 0.01027846,
      "balance_loss_clip": 1.04159105,
      "balance_loss_mlp": 1.0206666,
      "epoch": 0.983466602537125,
      "flos": 13626699805440.0,
      "grad_norm": 2.2724603447668277,
      "language_loss": 0.74419934,
      "learning_rate": 2.8465669112716083e-09,
      "loss": 0.76573163,
      "num_input_tokens_seen": 176583480,
      "step": 8179,
      "time_per_iteration": 2.4320054054260254
    },
    {
      "auxiliary_loss_clip": 0.01151471,
      "auxiliary_loss_mlp": 0.0076166,
      "balance_loss_clip": 1.04367423,
      "balance_loss_mlp": 1.00044596,
      "epoch": 0.9835868454277641,
      "flos": 22926844563840.0,
      "grad_norm": 1.938651430601561,
      "language_loss": 0.76275772,
      "learning_rate": 2.8051725346858177e-09,
      "loss": 0.78188902,
      "num_input_tokens_seen": 176603740,
      "step": 8180,
      "time_per_iteration": 2.453141927719116
    },
    {
      "auxiliary_loss_clip": 0.01167265,
      "auxiliary_loss_mlp": 0.01025723,
      "balance_loss_clip": 1.04661274,
      "balance_loss_mlp": 1.01874924,
      "epoch": 0.9837070883184031,
      "flos": 27673409341440.0,
      "grad_norm": 2.052482446452464,
      "language_loss": 0.7064414,
      "learning_rate": 2.7640811309674883e-09,
      "loss": 0.72837126,
      "num_input_tokens_seen": 176623240,
      "step": 8181,
      "time_per_iteration": 2.4584624767303467
    },
    {
      "auxiliary_loss_clip": 0.01112655,
      "auxiliary_loss_mlp": 0.01020801,
      "balance_loss_clip": 1.04154074,
      "balance_loss_mlp": 1.01393723,
      "epoch": 0.9838273312090423,
      "flos": 29241951425280.0,
      "grad_norm": 1.6232823629135043,
      "language_loss": 0.80797291,
      "learning_rate": 2.7232927063498557e-09,
      "loss": 0.82930744,
      "num_input_tokens_seen": 176643615,
      "step": 8182,
      "time_per_iteration": 3.3102962970733643
    },
    {
      "auxiliary_loss_clip": 0.01153901,
      "auxiliary_loss_mlp": 0.01021006,
      "balance_loss_clip": 1.04618919,
      "balance_loss_mlp": 1.01408613,
      "epoch": 0.9839475740996814,
      "flos": 40110207304320.0,
      "grad_norm": 6.551770927331249,
      "language_loss": 0.68979466,
      "learning_rate": 2.682807267020859e-09,
      "loss": 0.71154368,
      "num_input_tokens_seen": 176666375,
      "step": 8183,
      "time_per_iteration": 2.5814456939697266
    },
    {
      "auxiliary_loss_clip": 0.01150134,
      "auxiliary_loss_mlp": 0.01022598,
      "balance_loss_clip": 1.04517221,
      "balance_loss_mlp": 1.01510835,
      "epoch": 0.9840678169903204,
      "flos": 24169389788160.0,
      "grad_norm": 1.522742496874214,
      "language_loss": 0.62137973,
      "learning_rate": 2.642624819121808e-09,
      "loss": 0.643107,
      "num_input_tokens_seen": 176686525,
      "step": 8184,
      "time_per_iteration": 2.433030605316162
    },
    {
      "auxiliary_loss_clip": 0.01134684,
      "auxiliary_loss_mlp": 0.01025279,
      "balance_loss_clip": 1.04492819,
      "balance_loss_mlp": 1.0182457,
      "epoch": 0.9841880598809596,
      "flos": 14684484447360.0,
      "grad_norm": 1.9074219501131964,
      "language_loss": 0.61766517,
      "learning_rate": 2.6027453687487154e-09,
      "loss": 0.63926476,
      "num_input_tokens_seen": 176703615,
      "step": 8185,
      "time_per_iteration": 2.453526258468628
    },
    {
      "auxiliary_loss_clip": 0.0113783,
      "auxiliary_loss_mlp": 0.01022139,
      "balance_loss_clip": 1.04499722,
      "balance_loss_mlp": 1.01523411,
      "epoch": 0.9843083027715986,
      "flos": 22344768668160.0,
      "grad_norm": 2.3130294478795603,
      "language_loss": 0.54243672,
      "learning_rate": 2.5631689219509643e-09,
      "loss": 0.56403637,
      "num_input_tokens_seen": 176722295,
      "step": 8186,
      "time_per_iteration": 2.466492176055908
    },
    {
      "auxiliary_loss_clip": 0.01139525,
      "auxiliary_loss_mlp": 0.01021497,
      "balance_loss_clip": 1.04729521,
      "balance_loss_mlp": 1.0150423,
      "epoch": 0.9844285456622377,
      "flos": 21800111765760.0,
      "grad_norm": 1.7542054970354624,
      "language_loss": 0.83476424,
      "learning_rate": 2.523895484732197e-09,
      "loss": 0.8563745,
      "num_input_tokens_seen": 176741750,
      "step": 8187,
      "time_per_iteration": 2.4721720218658447
    },
    {
      "auxiliary_loss_clip": 0.01155893,
      "auxiliary_loss_mlp": 0.01024094,
      "balance_loss_clip": 1.04543436,
      "balance_loss_mlp": 1.01610994,
      "epoch": 0.9845487885528769,
      "flos": 18035380321920.0,
      "grad_norm": 1.9993578250037793,
      "language_loss": 0.74915093,
      "learning_rate": 2.4849250630505357e-09,
      "loss": 0.77095079,
      "num_input_tokens_seen": 176759995,
      "step": 8188,
      "time_per_iteration": 2.4326727390289307
    },
    {
      "auxiliary_loss_clip": 0.0106834,
      "auxiliary_loss_mlp": 0.01025149,
      "balance_loss_clip": 1.03526163,
      "balance_loss_mlp": 1.01820219,
      "epoch": 0.9846690314435159,
      "flos": 25228610974080.0,
      "grad_norm": 1.6861320855085145,
      "language_loss": 0.73389876,
      "learning_rate": 2.4462576628172528e-09,
      "loss": 0.75483364,
      "num_input_tokens_seen": 176778625,
      "step": 8189,
      "time_per_iteration": 2.6228549480438232
    },
    {
      "auxiliary_loss_clip": 0.01149123,
      "auxiliary_loss_mlp": 0.01028448,
      "balance_loss_clip": 1.04652023,
      "balance_loss_mlp": 1.02082777,
      "epoch": 0.984789274334155,
      "flos": 18552171248640.0,
      "grad_norm": 2.8843873994560543,
      "language_loss": 0.74140555,
      "learning_rate": 2.407893289898766e-09,
      "loss": 0.76318121,
      "num_input_tokens_seen": 176797655,
      "step": 8190,
      "time_per_iteration": 2.4206769466400146
    },
    {
      "auxiliary_loss_clip": 0.01113867,
      "auxiliary_loss_mlp": 0.01018266,
      "balance_loss_clip": 1.03964412,
      "balance_loss_mlp": 1.01115489,
      "epoch": 0.984909517224794,
      "flos": 27345437233920.0,
      "grad_norm": 1.7752209905871499,
      "language_loss": 0.84196013,
      "learning_rate": 2.3698319501144202e-09,
      "loss": 0.86328149,
      "num_input_tokens_seen": 176818640,
      "step": 8191,
      "time_per_iteration": 2.5583198070526123
    },
    {
      "auxiliary_loss_clip": 0.01156865,
      "auxiliary_loss_mlp": 0.01025094,
      "balance_loss_clip": 1.0452652,
      "balance_loss_mlp": 1.01747036,
      "epoch": 0.9850297601154332,
      "flos": 18734058743040.0,
      "grad_norm": 1.5630931861897777,
      "language_loss": 0.73526227,
      "learning_rate": 2.3320736492382644e-09,
      "loss": 0.75708187,
      "num_input_tokens_seen": 176837475,
      "step": 8192,
      "time_per_iteration": 2.4372599124908447
    },
    {
      "auxiliary_loss_clip": 0.01162336,
      "auxiliary_loss_mlp": 0.01026282,
      "balance_loss_clip": 1.04683065,
      "balance_loss_mlp": 1.01958275,
      "epoch": 0.9851500030060723,
      "flos": 22308247514880.0,
      "grad_norm": 1.5268476826951913,
      "language_loss": 0.68155867,
      "learning_rate": 2.29461839299816e-09,
      "loss": 0.70344484,
      "num_input_tokens_seen": 176857190,
      "step": 8193,
      "time_per_iteration": 2.4199471473693848
    },
    {
      "auxiliary_loss_clip": 0.01124452,
      "auxiliary_loss_mlp": 0.01017563,
      "balance_loss_clip": 1.0418613,
      "balance_loss_mlp": 1.01095009,
      "epoch": 0.9852702458967113,
      "flos": 26353691746560.0,
      "grad_norm": 1.5149358253887673,
      "language_loss": 0.80068064,
      "learning_rate": 2.257466187076229e-09,
      "loss": 0.82210076,
      "num_input_tokens_seen": 176876395,
      "step": 8194,
      "time_per_iteration": 2.5408670902252197
    },
    {
      "auxiliary_loss_clip": 0.01154064,
      "auxiliary_loss_mlp": 0.00761708,
      "balance_loss_clip": 1.04399788,
      "balance_loss_mlp": 1.00036049,
      "epoch": 0.9853904887873505,
      "flos": 20883599314560.0,
      "grad_norm": 2.049029767752672,
      "language_loss": 0.71050161,
      "learning_rate": 2.2206170371081854e-09,
      "loss": 0.72965932,
      "num_input_tokens_seen": 176894980,
      "step": 8195,
      "time_per_iteration": 2.443204402923584
    },
    {
      "auxiliary_loss_clip": 0.01138112,
      "auxiliary_loss_mlp": 0.0102752,
      "balance_loss_clip": 1.04334724,
      "balance_loss_mlp": 1.01947975,
      "epoch": 0.9855107316779895,
      "flos": 25263444188160.0,
      "grad_norm": 1.6394022493108442,
      "language_loss": 0.85019565,
      "learning_rate": 2.1840709486842247e-09,
      "loss": 0.87185192,
      "num_input_tokens_seen": 176914600,
      "step": 8196,
      "time_per_iteration": 2.506575107574463
    },
    {
      "auxiliary_loss_clip": 0.01127804,
      "auxiliary_loss_mlp": 0.01027435,
      "balance_loss_clip": 1.04220319,
      "balance_loss_mlp": 1.02022243,
      "epoch": 0.9856309745686286,
      "flos": 19062102677760.0,
      "grad_norm": 1.9332385512072892,
      "language_loss": 0.79136419,
      "learning_rate": 2.1478279273481335e-09,
      "loss": 0.81291664,
      "num_input_tokens_seen": 176933085,
      "step": 8197,
      "time_per_iteration": 2.462186098098755
    },
    {
      "auxiliary_loss_clip": 0.01150696,
      "auxiliary_loss_mlp": 0.01025565,
      "balance_loss_clip": 1.04712677,
      "balance_loss_mlp": 1.01867497,
      "epoch": 0.9857512174592677,
      "flos": 34130758060800.0,
      "grad_norm": 2.1952292721048243,
      "language_loss": 0.79617262,
      "learning_rate": 2.1118879785981815e-09,
      "loss": 0.81793523,
      "num_input_tokens_seen": 176953225,
      "step": 8198,
      "time_per_iteration": 2.538031816482544
    },
    {
      "auxiliary_loss_clip": 0.01133297,
      "auxiliary_loss_mlp": 0.01022241,
      "balance_loss_clip": 1.04383445,
      "balance_loss_mlp": 1.01555085,
      "epoch": 0.9858714603499068,
      "flos": 25994693266560.0,
      "grad_norm": 1.5733570963846992,
      "language_loss": 0.79581743,
      "learning_rate": 2.0762511078862288e-09,
      "loss": 0.8173728,
      "num_input_tokens_seen": 176973570,
      "step": 8199,
      "time_per_iteration": 3.349174737930298
    },
    {
      "auxiliary_loss_clip": 0.01142479,
      "auxiliary_loss_mlp": 0.01020954,
      "balance_loss_clip": 1.0434587,
      "balance_loss_mlp": 1.01426649,
      "epoch": 0.9859917032405459,
      "flos": 23696230907520.0,
      "grad_norm": 2.171100552167529,
      "language_loss": 0.65155751,
      "learning_rate": 2.0409173206186183e-09,
      "loss": 0.67319179,
      "num_input_tokens_seen": 176992810,
      "step": 8200,
      "time_per_iteration": 2.4816854000091553
    },
    {
      "auxiliary_loss_clip": 0.01121187,
      "auxiliary_loss_mlp": 0.01023869,
      "balance_loss_clip": 1.04552507,
      "balance_loss_mlp": 1.01708949,
      "epoch": 0.986111946131185,
      "flos": 19938287134080.0,
      "grad_norm": 4.955463796134859,
      "language_loss": 0.87317407,
      "learning_rate": 2.0058866221550617e-09,
      "loss": 0.89462465,
      "num_input_tokens_seen": 177011050,
      "step": 8201,
      "time_per_iteration": 3.305004596710205
    },
    {
      "auxiliary_loss_clip": 0.01163263,
      "auxiliary_loss_mlp": 0.0102051,
      "balance_loss_clip": 1.04446673,
      "balance_loss_mlp": 1.01359296,
      "epoch": 0.9862321890218241,
      "flos": 19828831415040.0,
      "grad_norm": 2.128078414798084,
      "language_loss": 0.75860846,
      "learning_rate": 1.971159017809976e-09,
      "loss": 0.78044623,
      "num_input_tokens_seen": 177029340,
      "step": 8202,
      "time_per_iteration": 2.389552354812622
    },
    {
      "auxiliary_loss_clip": 0.01149284,
      "auxiliary_loss_mlp": 0.01024006,
      "balance_loss_clip": 1.04506397,
      "balance_loss_mlp": 1.01650786,
      "epoch": 0.9863524319124631,
      "flos": 21652051904640.0,
      "grad_norm": 23.582536999610802,
      "language_loss": 0.77674824,
      "learning_rate": 1.93673451285159e-09,
      "loss": 0.79848117,
      "num_input_tokens_seen": 177048390,
      "step": 8203,
      "time_per_iteration": 3.1873929500579834
    },
    {
      "auxiliary_loss_clip": 0.0104363,
      "auxiliary_loss_mlp": 0.0100119,
      "balance_loss_clip": 1.00771856,
      "balance_loss_mlp": 1.00029624,
      "epoch": 0.9864726748031023,
      "flos": 52769977920000.0,
      "grad_norm": 0.7331525055409038,
      "language_loss": 0.56509274,
      "learning_rate": 1.9026131125019495e-09,
      "loss": 0.58554095,
      "num_input_tokens_seen": 177105760,
      "step": 8204,
      "time_per_iteration": 2.98915958404541
    },
    {
      "auxiliary_loss_clip": 0.01146008,
      "auxiliary_loss_mlp": 0.01023233,
      "balance_loss_clip": 1.04641187,
      "balance_loss_mlp": 1.01645613,
      "epoch": 0.9865929176937414,
      "flos": 23364631526400.0,
      "grad_norm": 1.6207546569319238,
      "language_loss": 0.86899006,
      "learning_rate": 1.8687948219371363e-09,
      "loss": 0.89068246,
      "num_input_tokens_seen": 177124985,
      "step": 8205,
      "time_per_iteration": 2.4355862140655518
    },
    {
      "auxiliary_loss_clip": 0.01169421,
      "auxiliary_loss_mlp": 0.01025656,
      "balance_loss_clip": 1.04695034,
      "balance_loss_mlp": 1.01823783,
      "epoch": 0.9867131605843804,
      "flos": 21616679986560.0,
      "grad_norm": 1.9949169528993096,
      "language_loss": 0.88505238,
      "learning_rate": 1.835279646287491e-09,
      "loss": 0.90700316,
      "num_input_tokens_seen": 177142995,
      "step": 8206,
      "time_per_iteration": 2.3916306495666504
    },
    {
      "auxiliary_loss_clip": 0.01157556,
      "auxiliary_loss_mlp": 0.01029037,
      "balance_loss_clip": 1.04673064,
      "balance_loss_mlp": 1.02163994,
      "epoch": 0.9868334034750196,
      "flos": 22271403139200.0,
      "grad_norm": 1.8474755622821888,
      "language_loss": 0.76665807,
      "learning_rate": 1.8020675906371685e-09,
      "loss": 0.78852397,
      "num_input_tokens_seen": 177162390,
      "step": 8207,
      "time_per_iteration": 2.433635950088501
    },
    {
      "auxiliary_loss_clip": 0.01103672,
      "auxiliary_loss_mlp": 0.01022467,
      "balance_loss_clip": 1.03899336,
      "balance_loss_mlp": 1.01545727,
      "epoch": 0.9869536463656586,
      "flos": 25809573548160.0,
      "grad_norm": 2.0798946815339803,
      "language_loss": 0.75366139,
      "learning_rate": 1.7691586600243612e-09,
      "loss": 0.77492273,
      "num_input_tokens_seen": 177181290,
      "step": 8208,
      "time_per_iteration": 2.557999610900879
    },
    {
      "auxiliary_loss_clip": 0.01137628,
      "auxiliary_loss_mlp": 0.01024275,
      "balance_loss_clip": 1.04889417,
      "balance_loss_mlp": 1.01703644,
      "epoch": 0.9870738892562977,
      "flos": 16398500613120.0,
      "grad_norm": 5.2309106437645605,
      "language_loss": 0.86625189,
      "learning_rate": 1.7365528594415202e-09,
      "loss": 0.88787091,
      "num_input_tokens_seen": 177195360,
      "step": 8209,
      "time_per_iteration": 3.213785171508789
    },
    {
      "auxiliary_loss_clip": 0.01153724,
      "auxiliary_loss_mlp": 0.00762035,
      "balance_loss_clip": 1.04468632,
      "balance_loss_mlp": 1.00045824,
      "epoch": 0.9871941321469369,
      "flos": 35481358373760.0,
      "grad_norm": 9.45133272130327,
      "language_loss": 0.67614359,
      "learning_rate": 1.7042501938346888e-09,
      "loss": 0.69530118,
      "num_input_tokens_seen": 177218090,
      "step": 8210,
      "time_per_iteration": 2.5642542839050293
    },
    {
      "auxiliary_loss_clip": 0.01124516,
      "auxiliary_loss_mlp": 0.01021806,
      "balance_loss_clip": 1.03919399,
      "balance_loss_mlp": 1.01470113,
      "epoch": 0.9873143750375759,
      "flos": 21434217874560.0,
      "grad_norm": 1.975584357556349,
      "language_loss": 0.76455963,
      "learning_rate": 1.6722506681043913e-09,
      "loss": 0.7860229,
      "num_input_tokens_seen": 177237050,
      "step": 8211,
      "time_per_iteration": 2.476716995239258
    },
    {
      "auxiliary_loss_clip": 0.01140849,
      "auxiliary_loss_mlp": 0.01027585,
      "balance_loss_clip": 1.04503632,
      "balance_loss_mlp": 1.0207485,
      "epoch": 0.987434617928215,
      "flos": 16326499800960.0,
      "grad_norm": 2.979844503968214,
      "language_loss": 0.68847156,
      "learning_rate": 1.640554287104745e-09,
      "loss": 0.71015596,
      "num_input_tokens_seen": 177255325,
      "step": 8212,
      "time_per_iteration": 2.4656195640563965
    },
    {
      "auxiliary_loss_clip": 0.01123628,
      "auxiliary_loss_mlp": 0.01023534,
      "balance_loss_clip": 1.03853714,
      "balance_loss_mlp": 1.01629519,
      "epoch": 0.9875548608188541,
      "flos": 17851984456320.0,
      "grad_norm": 2.073369511657995,
      "language_loss": 0.79677486,
      "learning_rate": 1.609161055644348e-09,
      "loss": 0.81824642,
      "num_input_tokens_seen": 177271250,
      "step": 8213,
      "time_per_iteration": 2.4907920360565186
    },
    {
      "auxiliary_loss_clip": 0.01158763,
      "auxiliary_loss_mlp": 0.01023699,
      "balance_loss_clip": 1.04591489,
      "balance_loss_mlp": 1.01667762,
      "epoch": 0.9876751037094932,
      "flos": 26132876887680.0,
      "grad_norm": 2.0447305519179375,
      "language_loss": 0.68622923,
      "learning_rate": 1.5780709784849467e-09,
      "loss": 0.70805389,
      "num_input_tokens_seen": 177288270,
      "step": 8214,
      "time_per_iteration": 2.4699149131774902
    },
    {
      "auxiliary_loss_clip": 0.01100066,
      "auxiliary_loss_mlp": 0.01025625,
      "balance_loss_clip": 1.0441165,
      "balance_loss_mlp": 1.0185796,
      "epoch": 0.9877953466001322,
      "flos": 15991344973440.0,
      "grad_norm": 2.0068672353271833,
      "language_loss": 0.82483274,
      "learning_rate": 1.5472840603436565e-09,
      "loss": 0.8460896,
      "num_input_tokens_seen": 177305500,
      "step": 8215,
      "time_per_iteration": 2.543728828430176
    },
    {
      "auxiliary_loss_clip": 0.01141856,
      "auxiliary_loss_mlp": 0.01024319,
      "balance_loss_clip": 1.04618812,
      "balance_loss_mlp": 1.0172441,
      "epoch": 0.9879155894907714,
      "flos": 18806777827200.0,
      "grad_norm": 1.9405801073787239,
      "language_loss": 0.77950031,
      "learning_rate": 1.5168003058900757e-09,
      "loss": 0.80116206,
      "num_input_tokens_seen": 177323500,
      "step": 8216,
      "time_per_iteration": 2.4423890113830566
    },
    {
      "auxiliary_loss_clip": 0.01121892,
      "auxiliary_loss_mlp": 0.01026682,
      "balance_loss_clip": 1.04090095,
      "balance_loss_mlp": 1.01993763,
      "epoch": 0.9880358323814105,
      "flos": 22382044007040.0,
      "grad_norm": 2.0017821861851766,
      "language_loss": 0.9196167,
      "learning_rate": 1.4866197197491715e-09,
      "loss": 0.94110239,
      "num_input_tokens_seen": 177342860,
      "step": 8217,
      "time_per_iteration": 2.5106611251831055
    },
    {
      "auxiliary_loss_clip": 0.01155507,
      "auxiliary_loss_mlp": 0.0076251,
      "balance_loss_clip": 1.04641223,
      "balance_loss_mlp": 1.00047112,
      "epoch": 0.9881560752720495,
      "flos": 15668831733120.0,
      "grad_norm": 2.9338304579527676,
      "language_loss": 0.7857607,
      "learning_rate": 1.4567423064988371e-09,
      "loss": 0.80494094,
      "num_input_tokens_seen": 177360210,
      "step": 8218,
      "time_per_iteration": 2.408687114715576
    },
    {
      "auxiliary_loss_clip": 0.01165686,
      "auxiliary_loss_mlp": 0.01024608,
      "balance_loss_clip": 1.04594207,
      "balance_loss_mlp": 1.01744914,
      "epoch": 0.9882763181626887,
      "flos": 21500113374720.0,
      "grad_norm": 1.8667621797794887,
      "language_loss": 0.77782595,
      "learning_rate": 1.4271680706718913e-09,
      "loss": 0.79972887,
      "num_input_tokens_seen": 177377885,
      "step": 8219,
      "time_per_iteration": 2.394237518310547
    },
    {
      "auxiliary_loss_clip": 0.01155444,
      "auxiliary_loss_mlp": 0.01025989,
      "balance_loss_clip": 1.04795313,
      "balance_loss_mlp": 1.01845527,
      "epoch": 0.9883965610533277,
      "flos": 28034598551040.0,
      "grad_norm": 1.6049216300576692,
      "language_loss": 0.82374293,
      "learning_rate": 1.3978970167543013e-09,
      "loss": 0.84555721,
      "num_input_tokens_seen": 177398065,
      "step": 8220,
      "time_per_iteration": 2.5051913261413574
    },
    {
      "auxiliary_loss_clip": 0.01128108,
      "auxiliary_loss_mlp": 0.01025634,
      "balance_loss_clip": 1.04279625,
      "balance_loss_mlp": 1.01836777,
      "epoch": 0.9885168039439668,
      "flos": 14098601710080.0,
      "grad_norm": 2.4830200864720564,
      "language_loss": 0.7796874,
      "learning_rate": 1.3689291491867372e-09,
      "loss": 0.80122483,
      "num_input_tokens_seen": 177416380,
      "step": 8221,
      "time_per_iteration": 2.443425178527832
    },
    {
      "auxiliary_loss_clip": 0.01165878,
      "auxiliary_loss_mlp": 0.01028487,
      "balance_loss_clip": 1.04621661,
      "balance_loss_mlp": 1.02078021,
      "epoch": 0.988637046834606,
      "flos": 26432013352320.0,
      "grad_norm": 2.1511575434542562,
      "language_loss": 0.73536474,
      "learning_rate": 1.3402644723636836e-09,
      "loss": 0.75730836,
      "num_input_tokens_seen": 177438410,
      "step": 8222,
      "time_per_iteration": 2.483613967895508
    },
    {
      "auxiliary_loss_clip": 0.01131337,
      "auxiliary_loss_mlp": 0.01023972,
      "balance_loss_clip": 1.04553604,
      "balance_loss_mlp": 1.01703703,
      "epoch": 0.988757289725245,
      "flos": 25229113764480.0,
      "grad_norm": 1.9001244244780227,
      "language_loss": 0.83645672,
      "learning_rate": 1.311902990633218e-09,
      "loss": 0.85800976,
      "num_input_tokens_seen": 177457375,
      "step": 8223,
      "time_per_iteration": 2.4874227046966553
    },
    {
      "auxiliary_loss_clip": 0.01130236,
      "auxiliary_loss_mlp": 0.01019633,
      "balance_loss_clip": 1.04057741,
      "balance_loss_mlp": 1.01319551,
      "epoch": 0.9888775326158841,
      "flos": 26359042872960.0,
      "grad_norm": 1.5520314851969768,
      "language_loss": 0.71581125,
      "learning_rate": 1.2838447082978987e-09,
      "loss": 0.73730993,
      "num_input_tokens_seen": 177478530,
      "step": 8224,
      "time_per_iteration": 2.5411577224731445
    },
    {
      "auxiliary_loss_clip": 0.01146979,
      "auxiliary_loss_mlp": 0.01022409,
      "balance_loss_clip": 1.04357994,
      "balance_loss_mlp": 1.01494384,
      "epoch": 0.9889977755065231,
      "flos": 24316120846080.0,
      "grad_norm": 3.907379924692396,
      "language_loss": 0.83664012,
      "learning_rate": 1.2560896296143208e-09,
      "loss": 0.85833395,
      "num_input_tokens_seen": 177496995,
      "step": 8225,
      "time_per_iteration": 2.4394185543060303
    },
    {
      "auxiliary_loss_clip": 0.01165436,
      "auxiliary_loss_mlp": 0.01024812,
      "balance_loss_clip": 1.04679155,
      "balance_loss_mlp": 1.01777577,
      "epoch": 0.9891180183971623,
      "flos": 18951066760320.0,
      "grad_norm": 2.118567691603205,
      "language_loss": 0.82306933,
      "learning_rate": 1.2286377587926722e-09,
      "loss": 0.84497184,
      "num_input_tokens_seen": 177513785,
      "step": 8226,
      "time_per_iteration": 3.2136430740356445
    },
    {
      "auxiliary_loss_clip": 0.01163463,
      "auxiliary_loss_mlp": 0.01022611,
      "balance_loss_clip": 1.04537535,
      "balance_loss_mlp": 1.01529193,
      "epoch": 0.9892382612878013,
      "flos": 26176580760960.0,
      "grad_norm": 1.9185948701520856,
      "language_loss": 0.74751675,
      "learning_rate": 1.2014890999973992e-09,
      "loss": 0.76937747,
      "num_input_tokens_seen": 177530705,
      "step": 8227,
      "time_per_iteration": 2.427504062652588
    },
    {
      "auxiliary_loss_clip": 0.01162274,
      "auxiliary_loss_mlp": 0.01023804,
      "balance_loss_clip": 1.04502368,
      "balance_loss_mlp": 1.01721191,
      "epoch": 0.9893585041784404,
      "flos": 25449605400960.0,
      "grad_norm": 1.5633670665264983,
      "language_loss": 0.78335214,
      "learning_rate": 1.1746436573472073e-09,
      "loss": 0.80521291,
      "num_input_tokens_seen": 177552440,
      "step": 8228,
      "time_per_iteration": 3.2761571407318115
    },
    {
      "auxiliary_loss_clip": 0.01145328,
      "auxiliary_loss_mlp": 0.01024164,
      "balance_loss_clip": 1.04438376,
      "balance_loss_mlp": 1.01721108,
      "epoch": 0.9894787470690796,
      "flos": 20189302352640.0,
      "grad_norm": 1.9098205785811575,
      "language_loss": 0.691966,
      "learning_rate": 1.1481014349141726e-09,
      "loss": 0.71366096,
      "num_input_tokens_seen": 177569660,
      "step": 8229,
      "time_per_iteration": 2.446530818939209
    },
    {
      "auxiliary_loss_clip": 0.01136167,
      "auxiliary_loss_mlp": 0.01024777,
      "balance_loss_clip": 1.0436995,
      "balance_loss_mlp": 1.01718926,
      "epoch": 0.9895989899597186,
      "flos": 24644308435200.0,
      "grad_norm": 1.822870311681976,
      "language_loss": 0.84340888,
      "learning_rate": 1.121862436724852e-09,
      "loss": 0.86501831,
      "num_input_tokens_seen": 177588500,
      "step": 8230,
      "time_per_iteration": 3.2739226818084717
    },
    {
      "auxiliary_loss_clip": 0.01153257,
      "auxiliary_loss_mlp": 0.01026766,
      "balance_loss_clip": 1.04843354,
      "balance_loss_mlp": 1.01961923,
      "epoch": 0.9897192328503577,
      "flos": 21799034357760.0,
      "grad_norm": 1.7222976887879071,
      "language_loss": 0.70558274,
      "learning_rate": 1.0959266667598388e-09,
      "loss": 0.72738296,
      "num_input_tokens_seen": 177607315,
      "step": 8231,
      "time_per_iteration": 2.500082015991211
    },
    {
      "auxiliary_loss_clip": 0.01127318,
      "auxiliary_loss_mlp": 0.01026287,
      "balance_loss_clip": 1.04540443,
      "balance_loss_mlp": 1.01870525,
      "epoch": 0.9898394757409968,
      "flos": 21325229032320.0,
      "grad_norm": 1.9228778199221135,
      "language_loss": 0.74942732,
      "learning_rate": 1.0702941289533196e-09,
      "loss": 0.77096337,
      "num_input_tokens_seen": 177625990,
      "step": 8232,
      "time_per_iteration": 2.558720827102661
    },
    {
      "auxiliary_loss_clip": 0.01120073,
      "auxiliary_loss_mlp": 0.01024423,
      "balance_loss_clip": 1.04184389,
      "balance_loss_mlp": 1.0182476,
      "epoch": 0.9899597186316359,
      "flos": 18545024442240.0,
      "grad_norm": 2.152609881846572,
      "language_loss": 0.88654989,
      "learning_rate": 1.0449648271939615e-09,
      "loss": 0.90799487,
      "num_input_tokens_seen": 177642335,
      "step": 8233,
      "time_per_iteration": 2.496023654937744
    },
    {
      "auxiliary_loss_clip": 0.01116254,
      "auxiliary_loss_mlp": 0.00762187,
      "balance_loss_clip": 1.04606092,
      "balance_loss_mlp": 1.00047374,
      "epoch": 0.990079961522275,
      "flos": 23766723348480.0,
      "grad_norm": 1.4953693311670235,
      "language_loss": 0.72706974,
      "learning_rate": 1.0199387653240243e-09,
      "loss": 0.7458542,
      "num_input_tokens_seen": 177662025,
      "step": 8234,
      "time_per_iteration": 2.562835454940796
    },
    {
      "auxiliary_loss_clip": 0.01131009,
      "auxiliary_loss_mlp": 0.01021564,
      "balance_loss_clip": 1.04393375,
      "balance_loss_mlp": 1.01464093,
      "epoch": 0.9902002044129141,
      "flos": 16399182971520.0,
      "grad_norm": 1.671507168851399,
      "language_loss": 0.70949006,
      "learning_rate": 9.952159471400267e-10,
      "loss": 0.73101586,
      "num_input_tokens_seen": 177679065,
      "step": 8235,
      "time_per_iteration": 3.1519935131073
    },
    {
      "auxiliary_loss_clip": 0.01153063,
      "auxiliary_loss_mlp": 0.00761591,
      "balance_loss_clip": 1.04518795,
      "balance_loss_mlp": 1.0004065,
      "epoch": 0.9903204473035532,
      "flos": 22559657783040.0,
      "grad_norm": 1.8935456654621574,
      "language_loss": 0.84576666,
      "learning_rate": 9.707963763923022e-10,
      "loss": 0.86491323,
      "num_input_tokens_seen": 177698115,
      "step": 8236,
      "time_per_iteration": 2.517108917236328
    },
    {
      "auxiliary_loss_clip": 0.01132951,
      "auxiliary_loss_mlp": 0.01021913,
      "balance_loss_clip": 1.04107857,
      "balance_loss_mlp": 1.01476312,
      "epoch": 0.9904406901941922,
      "flos": 16144001775360.0,
      "grad_norm": 1.9080028037985008,
      "language_loss": 0.79011548,
      "learning_rate": 9.466800567854427e-10,
      "loss": 0.8116641,
      "num_input_tokens_seen": 177716715,
      "step": 8237,
      "time_per_iteration": 2.46001935005188
    },
    {
      "auxiliary_loss_clip": 0.0112425,
      "auxiliary_loss_mlp": 0.0102061,
      "balance_loss_clip": 1.03991556,
      "balance_loss_mlp": 1.01293325,
      "epoch": 0.9905609330848314,
      "flos": 26651499408000.0,
      "grad_norm": 3.4816435553339224,
      "language_loss": 0.68279338,
      "learning_rate": 9.228669919778553e-10,
      "loss": 0.70424193,
      "num_input_tokens_seen": 177735640,
      "step": 8238,
      "time_per_iteration": 2.5612010955810547
    },
    {
      "auxiliary_loss_clip": 0.01131846,
      "auxiliary_loss_mlp": 0.01027475,
      "balance_loss_clip": 1.04183245,
      "balance_loss_mlp": 1.01966715,
      "epoch": 0.9906811759754705,
      "flos": 23111820627840.0,
      "grad_norm": 2.039818090992299,
      "language_loss": 0.79856187,
      "learning_rate": 8.993571855817617e-10,
      "loss": 0.82015508,
      "num_input_tokens_seen": 177754470,
      "step": 8239,
      "time_per_iteration": 2.4819586277008057
    },
    {
      "auxiliary_loss_clip": 0.01151675,
      "auxiliary_loss_mlp": 0.01023849,
      "balance_loss_clip": 1.04525828,
      "balance_loss_mlp": 1.01685715,
      "epoch": 0.9908014188661095,
      "flos": 22090593052800.0,
      "grad_norm": 3.4560413166188786,
      "language_loss": 0.751387,
      "learning_rate": 8.761506411638642e-10,
      "loss": 0.77314222,
      "num_input_tokens_seen": 177773935,
      "step": 8240,
      "time_per_iteration": 2.5067315101623535
    },
    {
      "auxiliary_loss_clip": 0.0113594,
      "auxiliary_loss_mlp": 0.01030578,
      "balance_loss_clip": 1.04533064,
      "balance_loss_mlp": 1.02366042,
      "epoch": 0.9909216617567487,
      "flos": 19242948677760.0,
      "grad_norm": 1.7315299882238075,
      "language_loss": 0.73853493,
      "learning_rate": 8.53247362244236e-10,
      "loss": 0.76020014,
      "num_input_tokens_seen": 177792745,
      "step": 8241,
      "time_per_iteration": 2.50022292137146
    },
    {
      "auxiliary_loss_clip": 0.01138599,
      "auxiliary_loss_mlp": 0.01023076,
      "balance_loss_clip": 1.04534411,
      "balance_loss_mlp": 1.01628959,
      "epoch": 0.9910419046473877,
      "flos": 23621213352960.0,
      "grad_norm": 1.728782356122078,
      "language_loss": 0.68237358,
      "learning_rate": 8.306473522976532e-10,
      "loss": 0.70399034,
      "num_input_tokens_seen": 177812150,
      "step": 8242,
      "time_per_iteration": 2.5199990272521973
    },
    {
      "auxiliary_loss_clip": 0.01165397,
      "auxiliary_loss_mlp": 0.01019958,
      "balance_loss_clip": 1.0478332,
      "balance_loss_mlp": 1.01259971,
      "epoch": 0.9911621475380268,
      "flos": 22711380831360.0,
      "grad_norm": 1.7422668945760342,
      "language_loss": 0.71756887,
      "learning_rate": 8.083506147522623e-10,
      "loss": 0.73942244,
      "num_input_tokens_seen": 177831545,
      "step": 8243,
      "time_per_iteration": 2.4368674755096436
    },
    {
      "auxiliary_loss_clip": 0.01144136,
      "auxiliary_loss_mlp": 0.01026857,
      "balance_loss_clip": 1.04296041,
      "balance_loss_mlp": 1.01975846,
      "epoch": 0.991282390428666,
      "flos": 13516956777600.0,
      "grad_norm": 2.071491621034792,
      "language_loss": 0.853513,
      "learning_rate": 7.863571529906909e-10,
      "loss": 0.87522292,
      "num_input_tokens_seen": 177847130,
      "step": 8244,
      "time_per_iteration": 2.428632974624634
    },
    {
      "auxiliary_loss_clip": 0.01034688,
      "auxiliary_loss_mlp": 0.0100149,
      "balance_loss_clip": 1.00752044,
      "balance_loss_mlp": 1.00058353,
      "epoch": 0.991402633319305,
      "flos": 61830492071040.0,
      "grad_norm": 0.7218832933836399,
      "language_loss": 0.59645915,
      "learning_rate": 7.646669703489372e-10,
      "loss": 0.61682093,
      "num_input_tokens_seen": 177911440,
      "step": 8245,
      "time_per_iteration": 3.1910464763641357
    },
    {
      "auxiliary_loss_clip": 0.01086141,
      "auxiliary_loss_mlp": 0.01021818,
      "balance_loss_clip": 1.04061437,
      "balance_loss_mlp": 1.01485324,
      "epoch": 0.9915228762099441,
      "flos": 18770148933120.0,
      "grad_norm": 1.8423509444893138,
      "language_loss": 0.57068819,
      "learning_rate": 7.432800701177023e-10,
      "loss": 0.59176779,
      "num_input_tokens_seen": 177929440,
      "step": 8246,
      "time_per_iteration": 2.581807851791382
    },
    {
      "auxiliary_loss_clip": 0.01043244,
      "auxiliary_loss_mlp": 0.01001444,
      "balance_loss_clip": 1.00897479,
      "balance_loss_mlp": 1.00056732,
      "epoch": 0.9916431191005832,
      "flos": 65936660244480.0,
      "grad_norm": 0.7936290014642606,
      "language_loss": 0.57828838,
      "learning_rate": 7.221964555415017e-10,
      "loss": 0.59873521,
      "num_input_tokens_seen": 177989100,
      "step": 8247,
      "time_per_iteration": 3.0172791481018066
    },
    {
      "auxiliary_loss_clip": 0.01135196,
      "auxiliary_loss_mlp": 0.01019479,
      "balance_loss_clip": 1.04408693,
      "balance_loss_mlp": 1.01303303,
      "epoch": 0.9917633619912223,
      "flos": 16581573256320.0,
      "grad_norm": 1.7519676958283414,
      "language_loss": 0.75069213,
      "learning_rate": 7.01416129818222e-10,
      "loss": 0.77223885,
      "num_input_tokens_seen": 178006720,
      "step": 8248,
      "time_per_iteration": 2.4533557891845703
    },
    {
      "auxiliary_loss_clip": 0.01129357,
      "auxiliary_loss_mlp": 0.01024999,
      "balance_loss_clip": 1.0436641,
      "balance_loss_mlp": 1.01781344,
      "epoch": 0.9918836048818613,
      "flos": 25411108999680.0,
      "grad_norm": 1.824356976086391,
      "language_loss": 0.58724999,
      "learning_rate": 6.809390961006745e-10,
      "loss": 0.6087935,
      "num_input_tokens_seen": 178026850,
      "step": 8249,
      "time_per_iteration": 2.563532590866089
    },
    {
      "auxiliary_loss_clip": 0.0113507,
      "auxiliary_loss_mlp": 0.01027173,
      "balance_loss_clip": 1.04441929,
      "balance_loss_mlp": 1.02013946,
      "epoch": 0.9920038477725005,
      "flos": 25046867134080.0,
      "grad_norm": 1.8895234176989288,
      "language_loss": 0.68826032,
      "learning_rate": 6.607653574948191e-10,
      "loss": 0.70988274,
      "num_input_tokens_seen": 178047630,
      "step": 8250,
      "time_per_iteration": 2.498539686203003
    },
    {
      "auxiliary_loss_clip": 0.01142,
      "auxiliary_loss_mlp": 0.01024422,
      "balance_loss_clip": 1.04175401,
      "balance_loss_mlp": 1.01758504,
      "epoch": 0.9921240906631396,
      "flos": 21829773421440.0,
      "grad_norm": 2.000025237290295,
      "language_loss": 0.81559461,
      "learning_rate": 6.408949170613187e-10,
      "loss": 0.83725882,
      "num_input_tokens_seen": 178066895,
      "step": 8251,
      "time_per_iteration": 2.4991910457611084
    },
    {
      "auxiliary_loss_clip": 0.01135389,
      "auxiliary_loss_mlp": 0.01024816,
      "balance_loss_clip": 1.04371572,
      "balance_loss_mlp": 1.01703203,
      "epoch": 0.9922443335537786,
      "flos": 24864225454080.0,
      "grad_norm": 2.615890616532026,
      "language_loss": 0.81753099,
      "learning_rate": 6.213277778144288e-10,
      "loss": 0.83913302,
      "num_input_tokens_seen": 178088540,
      "step": 8252,
      "time_per_iteration": 2.5331714153289795
    },
    {
      "auxiliary_loss_clip": 0.0109828,
      "auxiliary_loss_mlp": 0.01027662,
      "balance_loss_clip": 1.03925061,
      "balance_loss_mlp": 1.02021778,
      "epoch": 0.9923645764444178,
      "flos": 21613088626560.0,
      "grad_norm": 2.0226087257790306,
      "language_loss": 0.6694026,
      "learning_rate": 6.020639427224416e-10,
      "loss": 0.69066203,
      "num_input_tokens_seen": 178106185,
      "step": 8253,
      "time_per_iteration": 3.434929847717285
    },
    {
      "auxiliary_loss_clip": 0.01136079,
      "auxiliary_loss_mlp": 0.01021989,
      "balance_loss_clip": 1.04462194,
      "balance_loss_mlp": 1.01495004,
      "epoch": 0.9924848193350568,
      "flos": 25001798544000.0,
      "grad_norm": 2.0996007644077106,
      "language_loss": 0.72443128,
      "learning_rate": 5.831034147076864e-10,
      "loss": 0.74601185,
      "num_input_tokens_seen": 178123435,
      "step": 8254,
      "time_per_iteration": 2.504014730453491
    },
    {
      "auxiliary_loss_clip": 0.01049314,
      "auxiliary_loss_mlp": 0.01000677,
      "balance_loss_clip": 1.00832915,
      "balance_loss_mlp": 0.9997946,
      "epoch": 0.9926050622256959,
      "flos": 68912543151360.0,
      "grad_norm": 0.6875357353503534,
      "language_loss": 0.55787557,
      "learning_rate": 5.644461966463065e-10,
      "loss": 0.57837552,
      "num_input_tokens_seen": 178191045,
      "step": 8255,
      "time_per_iteration": 4.022282361984253
    },
    {
      "auxiliary_loss_clip": 0.01136333,
      "auxiliary_loss_mlp": 0.0102066,
      "balance_loss_clip": 1.0449394,
      "balance_loss_mlp": 1.01431227,
      "epoch": 0.9927253051163349,
      "flos": 20923675914240.0,
      "grad_norm": 1.7225020069692987,
      "language_loss": 0.75699162,
      "learning_rate": 5.460922913687049e-10,
      "loss": 0.77856159,
      "num_input_tokens_seen": 178210135,
      "step": 8256,
      "time_per_iteration": 3.2249083518981934
    },
    {
      "auxiliary_loss_clip": 0.01107072,
      "auxiliary_loss_mlp": 0.00762176,
      "balance_loss_clip": 1.03843856,
      "balance_loss_mlp": 1.00043488,
      "epoch": 0.9928455480069741,
      "flos": 22308211601280.0,
      "grad_norm": 2.286550897856297,
      "language_loss": 0.74978691,
      "learning_rate": 5.280417016593208e-10,
      "loss": 0.76847941,
      "num_input_tokens_seen": 178229925,
      "step": 8257,
      "time_per_iteration": 2.6291749477386475
    },
    {
      "auxiliary_loss_clip": 0.01148363,
      "auxiliary_loss_mlp": 0.00760879,
      "balance_loss_clip": 1.04695773,
      "balance_loss_mlp": 1.00039744,
      "epoch": 0.9929657908976132,
      "flos": 17383889393280.0,
      "grad_norm": 1.732674229424684,
      "language_loss": 0.74804699,
      "learning_rate": 5.102944302559642e-10,
      "loss": 0.76713943,
      "num_input_tokens_seen": 178247420,
      "step": 8258,
      "time_per_iteration": 2.442061424255371
    },
    {
      "auxiliary_loss_clip": 0.01104334,
      "auxiliary_loss_mlp": 0.01026245,
      "balance_loss_clip": 1.04263341,
      "balance_loss_mlp": 1.01882112,
      "epoch": 0.9930860337882522,
      "flos": 22674680110080.0,
      "grad_norm": 3.710278410150471,
      "language_loss": 0.79882371,
      "learning_rate": 4.9285047985137e-10,
      "loss": 0.82012951,
      "num_input_tokens_seen": 178266840,
      "step": 8259,
      "time_per_iteration": 2.6384880542755127
    },
    {
      "auxiliary_loss_clip": 0.01153456,
      "auxiliary_loss_mlp": 0.01028022,
      "balance_loss_clip": 1.04535997,
      "balance_loss_mlp": 1.02107477,
      "epoch": 0.9932062766788914,
      "flos": 28147789284480.0,
      "grad_norm": 1.6430070788512317,
      "language_loss": 0.74637204,
      "learning_rate": 4.757098530916436e-10,
      "loss": 0.76818681,
      "num_input_tokens_seen": 178287285,
      "step": 8260,
      "time_per_iteration": 2.5147809982299805
    },
    {
      "auxiliary_loss_clip": 0.01156336,
      "auxiliary_loss_mlp": 0.01025337,
      "balance_loss_clip": 1.04876614,
      "balance_loss_mlp": 1.0181278,
      "epoch": 0.9933265195695304,
      "flos": 20156659868160.0,
      "grad_norm": 3.782602774906576,
      "language_loss": 0.77083755,
      "learning_rate": 4.5887255257670563e-10,
      "loss": 0.79265428,
      "num_input_tokens_seen": 178304325,
      "step": 8261,
      "time_per_iteration": 2.4801785945892334
    },
    {
      "auxiliary_loss_clip": 0.01166238,
      "auxiliary_loss_mlp": 0.01027377,
      "balance_loss_clip": 1.04711235,
      "balance_loss_mlp": 1.0200752,
      "epoch": 0.9934467624601695,
      "flos": 21362037494400.0,
      "grad_norm": 2.6216314162661725,
      "language_loss": 0.7667163,
      "learning_rate": 4.4233858086117906e-10,
      "loss": 0.78865242,
      "num_input_tokens_seen": 178322850,
      "step": 8262,
      "time_per_iteration": 3.1451010704040527
    },
    {
      "auxiliary_loss_clip": 0.0110727,
      "auxiliary_loss_mlp": 0.01025509,
      "balance_loss_clip": 1.0444746,
      "balance_loss_mlp": 1.01830852,
      "epoch": 0.9935670053508087,
      "flos": 19756040503680.0,
      "grad_norm": 2.265969735330147,
      "language_loss": 0.67954516,
      "learning_rate": 4.261079404528356e-10,
      "loss": 0.70087296,
      "num_input_tokens_seen": 178342330,
      "step": 8263,
      "time_per_iteration": 2.5545506477355957
    },
    {
      "auxiliary_loss_clip": 0.01148542,
      "auxiliary_loss_mlp": 0.01020643,
      "balance_loss_clip": 1.04423273,
      "balance_loss_mlp": 1.01347888,
      "epoch": 0.9936872482414477,
      "flos": 21978838863360.0,
      "grad_norm": 3.2466274623058737,
      "language_loss": 0.69114614,
      "learning_rate": 4.1018063381437205e-10,
      "loss": 0.71283805,
      "num_input_tokens_seen": 178362715,
      "step": 8264,
      "time_per_iteration": 2.4658918380737305
    },
    {
      "auxiliary_loss_clip": 0.01047146,
      "auxiliary_loss_mlp": 0.01002243,
      "balance_loss_clip": 1.01047111,
      "balance_loss_mlp": 1.00147402,
      "epoch": 0.9938074911320868,
      "flos": 69810667839360.0,
      "grad_norm": 0.8778695489876152,
      "language_loss": 0.61184144,
      "learning_rate": 3.9455666336141167e-10,
      "loss": 0.63233531,
      "num_input_tokens_seen": 178426495,
      "step": 8265,
      "time_per_iteration": 3.1018972396850586
    },
    {
      "auxiliary_loss_clip": 0.01164182,
      "auxiliary_loss_mlp": 0.01023043,
      "balance_loss_clip": 1.04696429,
      "balance_loss_mlp": 1.01627243,
      "epoch": 0.9939277340227259,
      "flos": 15084170058240.0,
      "grad_norm": 2.5755021350693985,
      "language_loss": 0.83323324,
      "learning_rate": 3.7923603146450267e-10,
      "loss": 0.85510552,
      "num_input_tokens_seen": 178442555,
      "step": 8266,
      "time_per_iteration": 2.384077787399292
    },
    {
      "auxiliary_loss_clip": 0.01123074,
      "auxiliary_loss_mlp": 0.01024825,
      "balance_loss_clip": 1.03999043,
      "balance_loss_mlp": 1.01764226,
      "epoch": 0.994047976913365,
      "flos": 17712364291200.0,
      "grad_norm": 1.9986799115018,
      "language_loss": 0.81031752,
      "learning_rate": 3.642187404473418e-10,
      "loss": 0.83179653,
      "num_input_tokens_seen": 178460715,
      "step": 8267,
      "time_per_iteration": 2.5039939880371094
    },
    {
      "auxiliary_loss_clip": 0.01150662,
      "auxiliary_loss_mlp": 0.01020048,
      "balance_loss_clip": 1.0439105,
      "balance_loss_mlp": 1.0130173,
      "epoch": 0.994168219804004,
      "flos": 19171558396800.0,
      "grad_norm": 2.4422687537614896,
      "language_loss": 0.86333156,
      "learning_rate": 3.495047925885508e-10,
      "loss": 0.88503867,
      "num_input_tokens_seen": 178479050,
      "step": 8268,
      "time_per_iteration": 2.4145679473876953
    },
    {
      "auxiliary_loss_clip": 0.01135162,
      "auxiliary_loss_mlp": 0.01025372,
      "balance_loss_clip": 1.04222631,
      "balance_loss_mlp": 1.0183835,
      "epoch": 0.9942884626946432,
      "flos": 17851589406720.0,
      "grad_norm": 3.280059971402832,
      "language_loss": 0.82468426,
      "learning_rate": 3.350941901199e-10,
      "loss": 0.84628963,
      "num_input_tokens_seen": 178495970,
      "step": 8269,
      "time_per_iteration": 2.4775407314300537
    },
    {
      "auxiliary_loss_clip": 0.01139904,
      "auxiliary_loss_mlp": 0.01024556,
      "balance_loss_clip": 1.04442763,
      "balance_loss_mlp": 1.01753485,
      "epoch": 0.9944087055852823,
      "flos": 18796578364800.0,
      "grad_norm": 2.8330819347656204,
      "language_loss": 0.83172882,
      "learning_rate": 3.2098693522764066e-10,
      "loss": 0.85337341,
      "num_input_tokens_seen": 178509170,
      "step": 8270,
      "time_per_iteration": 2.444460868835449
    },
    {
      "auxiliary_loss_clip": 0.01143707,
      "auxiliary_loss_mlp": 0.00761732,
      "balance_loss_clip": 1.04585123,
      "balance_loss_mlp": 1.00047684,
      "epoch": 0.9945289484759213,
      "flos": 20996969616000.0,
      "grad_norm": 1.8870604872235957,
      "language_loss": 0.81237614,
      "learning_rate": 3.071830300516165e-10,
      "loss": 0.83143055,
      "num_input_tokens_seen": 178527000,
      "step": 8271,
      "time_per_iteration": 2.5649683475494385
    },
    {
      "auxiliary_loss_clip": 0.01155754,
      "auxiliary_loss_mlp": 0.01028625,
      "balance_loss_clip": 1.04585576,
      "balance_loss_mlp": 1.02126396,
      "epoch": 0.9946491913665605,
      "flos": 14756952136320.0,
      "grad_norm": 2.946529532764987,
      "language_loss": 0.70839703,
      "learning_rate": 2.9368247668615234e-10,
      "loss": 0.73024082,
      "num_input_tokens_seen": 178545590,
      "step": 8272,
      "time_per_iteration": 2.434903621673584
    },
    {
      "auxiliary_loss_clip": 0.01169771,
      "auxiliary_loss_mlp": 0.01026383,
      "balance_loss_clip": 1.04889834,
      "balance_loss_mlp": 1.01915324,
      "epoch": 0.9947694342571995,
      "flos": 12669931186560.0,
      "grad_norm": 2.834484945376903,
      "language_loss": 0.61655366,
      "learning_rate": 2.804852771789434e-10,
      "loss": 0.63851517,
      "num_input_tokens_seen": 178558890,
      "step": 8273,
      "time_per_iteration": 2.3944907188415527
    },
    {
      "auxiliary_loss_clip": 0.01162144,
      "auxiliary_loss_mlp": 0.01023541,
      "balance_loss_clip": 1.04530811,
      "balance_loss_mlp": 1.01653194,
      "epoch": 0.9948896771478386,
      "flos": 18843442634880.0,
      "grad_norm": 1.7999877857802915,
      "language_loss": 0.55636847,
      "learning_rate": 2.675914335321661e-10,
      "loss": 0.57822537,
      "num_input_tokens_seen": 178577645,
      "step": 8274,
      "time_per_iteration": 2.3823635578155518
    },
    {
      "auxiliary_loss_clip": 0.01157135,
      "auxiliary_loss_mlp": 0.01026259,
      "balance_loss_clip": 1.0468781,
      "balance_loss_mlp": 1.01828969,
      "epoch": 0.9950099200384778,
      "flos": 24900207903360.0,
      "grad_norm": 2.2975960007290586,
      "language_loss": 0.79673886,
      "learning_rate": 2.550009477018111e-10,
      "loss": 0.81857276,
      "num_input_tokens_seen": 178596415,
      "step": 8275,
      "time_per_iteration": 2.5290284156799316
    },
    {
      "auxiliary_loss_clip": 0.01137668,
      "auxiliary_loss_mlp": 0.00762007,
      "balance_loss_clip": 1.04595709,
      "balance_loss_mlp": 1.00047469,
      "epoch": 0.9951301629291168,
      "flos": 23733613987200.0,
      "grad_norm": 2.1993405447889796,
      "language_loss": 0.62824106,
      "learning_rate": 2.4271382159790634e-10,
      "loss": 0.64723784,
      "num_input_tokens_seen": 178613845,
      "step": 8276,
      "time_per_iteration": 2.50490665435791
    },
    {
      "auxiliary_loss_clip": 0.01107082,
      "auxiliary_loss_mlp": 0.01029012,
      "balance_loss_clip": 1.04100072,
      "balance_loss_mlp": 1.02220547,
      "epoch": 0.9952504058197559,
      "flos": 22236893147520.0,
      "grad_norm": 1.7755456709311237,
      "language_loss": 0.85643125,
      "learning_rate": 2.3073005708429406e-10,
      "loss": 0.87779218,
      "num_input_tokens_seen": 178633490,
      "step": 8277,
      "time_per_iteration": 2.58384370803833
    },
    {
      "auxiliary_loss_clip": 0.01122953,
      "auxiliary_loss_mlp": 0.01020328,
      "balance_loss_clip": 1.04578829,
      "balance_loss_mlp": 1.01383066,
      "epoch": 0.995370648710395,
      "flos": 21211032718080.0,
      "grad_norm": 1.7508562613524354,
      "language_loss": 0.72135162,
      "learning_rate": 2.190496559788535e-10,
      "loss": 0.74278438,
      "num_input_tokens_seen": 178651775,
      "step": 8278,
      "time_per_iteration": 2.5453786849975586
    },
    {
      "auxiliary_loss_clip": 0.01135067,
      "auxiliary_loss_mlp": 0.01027814,
      "balance_loss_clip": 1.04354143,
      "balance_loss_mlp": 1.02114177,
      "epoch": 0.9954908916010341,
      "flos": 14866731077760.0,
      "grad_norm": 2.746550380378938,
      "language_loss": 0.76418847,
      "learning_rate": 2.0767262005372265e-10,
      "loss": 0.78581727,
      "num_input_tokens_seen": 178669290,
      "step": 8279,
      "time_per_iteration": 3.3816354274749756
    },
    {
      "auxiliary_loss_clip": 0.01129969,
      "auxiliary_loss_mlp": 0.01023569,
      "balance_loss_clip": 1.04281509,
      "balance_loss_mlp": 1.01700974,
      "epoch": 0.9956111344916732,
      "flos": 19208259118080.0,
      "grad_norm": 2.048887141908911,
      "language_loss": 0.74988109,
      "learning_rate": 1.965989510346322e-10,
      "loss": 0.77141643,
      "num_input_tokens_seen": 178688410,
      "step": 8280,
      "time_per_iteration": 2.513462543487549
    },
    {
      "auxiliary_loss_clip": 0.01105228,
      "auxiliary_loss_mlp": 0.01026597,
      "balance_loss_clip": 1.04150772,
      "balance_loss_mlp": 1.01866937,
      "epoch": 0.9957313773823123,
      "flos": 20047060494720.0,
      "grad_norm": 1.9790593245790244,
      "language_loss": 0.71111608,
      "learning_rate": 1.8582865060134955e-10,
      "loss": 0.73243439,
      "num_input_tokens_seen": 178706600,
      "step": 8281,
      "time_per_iteration": 3.2828047275543213
    },
    {
      "auxiliary_loss_clip": 0.01060818,
      "auxiliary_loss_mlp": 0.0100096,
      "balance_loss_clip": 1.0073421,
      "balance_loss_mlp": 1.00010133,
      "epoch": 0.9958516202729514,
      "flos": 57483253768320.0,
      "grad_norm": 0.7859882120812638,
      "language_loss": 0.557284,
      "learning_rate": 1.7536172038790098e-10,
      "loss": 0.57790178,
      "num_input_tokens_seen": 178766910,
      "step": 8282,
      "time_per_iteration": 3.0593149662017822
    },
    {
      "auxiliary_loss_clip": 0.01137493,
      "auxiliary_loss_mlp": 0.01020369,
      "balance_loss_clip": 1.04611886,
      "balance_loss_mlp": 1.01371384,
      "epoch": 0.9959718631635904,
      "flos": 27782900974080.0,
      "grad_norm": 2.007162501087017,
      "language_loss": 0.69224477,
      "learning_rate": 1.651981619819054e-10,
      "loss": 0.71382344,
      "num_input_tokens_seen": 178784060,
      "step": 8283,
      "time_per_iteration": 3.3584096431732178
    },
    {
      "auxiliary_loss_clip": 0.01113905,
      "auxiliary_loss_mlp": 0.01023593,
      "balance_loss_clip": 1.04202104,
      "balance_loss_mlp": 1.0162971,
      "epoch": 0.9960921060542296,
      "flos": 24024095274240.0,
      "grad_norm": 2.3053812037058274,
      "language_loss": 0.71121198,
      "learning_rate": 1.5533797692546257e-10,
      "loss": 0.73258698,
      "num_input_tokens_seen": 178802795,
      "step": 8284,
      "time_per_iteration": 2.545639753341675
    },
    {
      "auxiliary_loss_clip": 0.01147523,
      "auxiliary_loss_mlp": 0.01019912,
      "balance_loss_clip": 1.0439508,
      "balance_loss_mlp": 1.01255369,
      "epoch": 0.9962123489448687,
      "flos": 18697393935360.0,
      "grad_norm": 1.9730891523720424,
      "language_loss": 0.84074718,
      "learning_rate": 1.4578116671404296e-10,
      "loss": 0.86242151,
      "num_input_tokens_seen": 178821075,
      "step": 8285,
      "time_per_iteration": 2.429304599761963
    },
    {
      "auxiliary_loss_clip": 0.01147004,
      "auxiliary_loss_mlp": 0.01028342,
      "balance_loss_clip": 1.04730105,
      "balance_loss_mlp": 1.02129698,
      "epoch": 0.9963325918355077,
      "flos": 20010754823040.0,
      "grad_norm": 2.318298631824008,
      "language_loss": 0.71324831,
      "learning_rate": 1.3652773279759777e-10,
      "loss": 0.7350018,
      "num_input_tokens_seen": 178837725,
      "step": 8286,
      "time_per_iteration": 2.428987503051758
    },
    {
      "auxiliary_loss_clip": 0.01149722,
      "auxiliary_loss_mlp": 0.0102428,
      "balance_loss_clip": 1.04419994,
      "balance_loss_mlp": 1.01704144,
      "epoch": 0.9964528347261468,
      "flos": 33108488991360.0,
      "grad_norm": 2.262626396535188,
      "language_loss": 0.62857044,
      "learning_rate": 1.2757767657989305e-10,
      "loss": 0.65031046,
      "num_input_tokens_seen": 178861515,
      "step": 8287,
      "time_per_iteration": 2.546614408493042
    },
    {
      "auxiliary_loss_clip": 0.01149698,
      "auxiliary_loss_mlp": 0.0101993,
      "balance_loss_clip": 1.04598641,
      "balance_loss_mlp": 1.01333213,
      "epoch": 0.9965730776167859,
      "flos": 23109342589440.0,
      "grad_norm": 1.900974434634456,
      "language_loss": 0.87091804,
      "learning_rate": 1.1893099941850948e-10,
      "loss": 0.89261436,
      "num_input_tokens_seen": 178880410,
      "step": 8288,
      "time_per_iteration": 2.4448535442352295
    },
    {
      "auxiliary_loss_clip": 0.01140624,
      "auxiliary_loss_mlp": 0.01024703,
      "balance_loss_clip": 1.04210591,
      "balance_loss_mlp": 1.01733637,
      "epoch": 0.996693320507425,
      "flos": 22965843755520.0,
      "grad_norm": 1.9484440571806665,
      "language_loss": 0.77200848,
      "learning_rate": 1.105877026252866e-10,
      "loss": 0.79366171,
      "num_input_tokens_seen": 178898740,
      "step": 8289,
      "time_per_iteration": 3.2521162033081055
    },
    {
      "auxiliary_loss_clip": 0.01165879,
      "auxiliary_loss_mlp": 0.0102603,
      "balance_loss_clip": 1.04586184,
      "balance_loss_mlp": 1.01809692,
      "epoch": 0.996813563398064,
      "flos": 13222740476160.0,
      "grad_norm": 1.9755491691134537,
      "language_loss": 0.71674532,
      "learning_rate": 1.0254778746565663e-10,
      "loss": 0.73866439,
      "num_input_tokens_seen": 178914015,
      "step": 8290,
      "time_per_iteration": 2.352820634841919
    },
    {
      "auxiliary_loss_clip": 0.01121647,
      "auxiliary_loss_mlp": 0.01026668,
      "balance_loss_clip": 1.04423678,
      "balance_loss_mlp": 1.01964045,
      "epoch": 0.9969338062887032,
      "flos": 14647855553280.0,
      "grad_norm": 2.1297054866249323,
      "language_loss": 0.7353223,
      "learning_rate": 9.481125515953259e-11,
      "loss": 0.75680548,
      "num_input_tokens_seen": 178932075,
      "step": 8291,
      "time_per_iteration": 2.52404522895813
    },
    {
      "auxiliary_loss_clip": 0.01108891,
      "auxiliary_loss_mlp": 0.01023996,
      "balance_loss_clip": 1.03809643,
      "balance_loss_mlp": 1.01663804,
      "epoch": 0.9970540491793423,
      "flos": 25735741142400.0,
      "grad_norm": 1.607387266526498,
      "language_loss": 0.79856205,
      "learning_rate": 8.737810688064228e-11,
      "loss": 0.81989086,
      "num_input_tokens_seen": 178951910,
      "step": 8292,
      "time_per_iteration": 2.5697712898254395
    },
    {
      "auxiliary_loss_clip": 0.01120101,
      "auxiliary_loss_mlp": 0.01030636,
      "balance_loss_clip": 1.04326797,
      "balance_loss_mlp": 1.02142727,
      "epoch": 0.9971742920699813,
      "flos": 21470236237440.0,
      "grad_norm": 2.0263255259910165,
      "language_loss": 0.79260868,
      "learning_rate": 8.024834375608414e-11,
      "loss": 0.814116,
      "num_input_tokens_seen": 178970500,
      "step": 8293,
      "time_per_iteration": 2.500955820083618
    },
    {
      "auxiliary_loss_clip": 0.01060692,
      "auxiliary_loss_mlp": 0.01000356,
      "balance_loss_clip": 1.00724888,
      "balance_loss_mlp": 0.99950927,
      "epoch": 0.9972945349606205,
      "flos": 72211223629440.0,
      "grad_norm": 0.8280346749141044,
      "language_loss": 0.6286484,
      "learning_rate": 7.342196686788149e-11,
      "loss": 0.64925897,
      "num_input_tokens_seen": 179023665,
      "step": 8294,
      "time_per_iteration": 2.9057698249816895
    },
    {
      "auxiliary_loss_clip": 0.01137953,
      "auxiliary_loss_mlp": 0.01026779,
      "balance_loss_clip": 1.04893553,
      "balance_loss_mlp": 1.01976371,
      "epoch": 0.9974147778512595,
      "flos": 19678293515520.0,
      "grad_norm": 1.9549901843514716,
      "language_loss": 0.69124424,
      "learning_rate": 6.689897725142834e-11,
      "loss": 0.71289158,
      "num_input_tokens_seen": 179043140,
      "step": 8295,
      "time_per_iteration": 2.4749598503112793
    },
    {
      "auxiliary_loss_clip": 0.01138681,
      "auxiliary_loss_mlp": 0.0102738,
      "balance_loss_clip": 1.04306006,
      "balance_loss_mlp": 1.020257,
      "epoch": 0.9975350207418986,
      "flos": 15960821391360.0,
      "grad_norm": 2.2458612826183146,
      "language_loss": 0.89006549,
      "learning_rate": 6.067937589615545e-11,
      "loss": 0.91172612,
      "num_input_tokens_seen": 179061215,
      "step": 8296,
      "time_per_iteration": 2.4623379707336426
    },
    {
      "auxiliary_loss_clip": 0.01044403,
      "auxiliary_loss_mlp": 0.01001086,
      "balance_loss_clip": 1.00851381,
      "balance_loss_mlp": 1.00013244,
      "epoch": 0.9976552636325378,
      "flos": 59961879768960.0,
      "grad_norm": 0.7528782873822731,
      "language_loss": 0.57673573,
      "learning_rate": 5.476316374575241e-11,
      "loss": 0.59719062,
      "num_input_tokens_seen": 179124700,
      "step": 8297,
      "time_per_iteration": 3.042626142501831
    },
    {
      "auxiliary_loss_clip": 0.011653,
      "auxiliary_loss_mlp": 0.01024709,
      "balance_loss_clip": 1.04595232,
      "balance_loss_mlp": 1.01693606,
      "epoch": 0.9977755065231768,
      "flos": 22487872452480.0,
      "grad_norm": 1.8741773739445933,
      "language_loss": 0.7285217,
      "learning_rate": 4.9150341697723476e-11,
      "loss": 0.7504217,
      "num_input_tokens_seen": 179144590,
      "step": 8298,
      "time_per_iteration": 2.4386112689971924
    },
    {
      "auxiliary_loss_clip": 0.01134111,
      "auxiliary_loss_mlp": 0.01029019,
      "balance_loss_clip": 1.04580164,
      "balance_loss_mlp": 1.02175307,
      "epoch": 0.9978957494138159,
      "flos": 26030280666240.0,
      "grad_norm": 1.5686091087729772,
      "language_loss": 0.66557115,
      "learning_rate": 4.384091060338768e-11,
      "loss": 0.68720245,
      "num_input_tokens_seen": 179165060,
      "step": 8299,
      "time_per_iteration": 2.5679140090942383
    },
    {
      "auxiliary_loss_clip": 0.01147877,
      "auxiliary_loss_mlp": 0.01024976,
      "balance_loss_clip": 1.04495907,
      "balance_loss_mlp": 1.01781726,
      "epoch": 0.998015992304455,
      "flos": 22637835734400.0,
      "grad_norm": 2.3049236335459375,
      "language_loss": 0.73332906,
      "learning_rate": 3.883487126810081e-11,
      "loss": 0.75505757,
      "num_input_tokens_seen": 179184320,
      "step": 8300,
      "time_per_iteration": 2.5333354473114014
    },
    {
      "auxiliary_loss_clip": 0.01140805,
      "auxiliary_loss_mlp": 0.01022428,
      "balance_loss_clip": 1.0424962,
      "balance_loss_mlp": 1.01512599,
      "epoch": 0.9981362351950941,
      "flos": 18223444955520.0,
      "grad_norm": 1.7308427971709435,
      "language_loss": 0.79488659,
      "learning_rate": 3.41322244516995e-11,
      "loss": 0.8165189,
      "num_input_tokens_seen": 179202265,
      "step": 8301,
      "time_per_iteration": 2.470193386077881
    },
    {
      "auxiliary_loss_clip": 0.01097787,
      "auxiliary_loss_mlp": 0.0102471,
      "balance_loss_clip": 1.04121625,
      "balance_loss_mlp": 1.0175364,
      "epoch": 0.9982564780857331,
      "flos": 33474095573760.0,
      "grad_norm": 1.6369820164721491,
      "language_loss": 0.63092965,
      "learning_rate": 2.9732970866946925e-11,
      "loss": 0.65215462,
      "num_input_tokens_seen": 179222145,
      "step": 8302,
      "time_per_iteration": 2.629645347595215
    },
    {
      "auxiliary_loss_clip": 0.0110876,
      "auxiliary_loss_mlp": 0.01023539,
      "balance_loss_clip": 1.03783226,
      "balance_loss_mlp": 1.01604652,
      "epoch": 0.9983767209763723,
      "flos": 15523465392000.0,
      "grad_norm": 2.0318882216207843,
      "language_loss": 0.78223205,
      "learning_rate": 2.563711118175327e-11,
      "loss": 0.80355501,
      "num_input_tokens_seen": 179239030,
      "step": 8303,
      "time_per_iteration": 2.491415023803711
    },
    {
      "auxiliary_loss_clip": 0.01119852,
      "auxiliary_loss_mlp": 0.01024339,
      "balance_loss_clip": 1.04409909,
      "balance_loss_mlp": 1.01711535,
      "epoch": 0.9984969638670114,
      "flos": 19974377324160.0,
      "grad_norm": 1.7846632600815349,
      "language_loss": 0.83492184,
      "learning_rate": 2.184464601717728e-11,
      "loss": 0.85636377,
      "num_input_tokens_seen": 179257345,
      "step": 8304,
      "time_per_iteration": 2.490250587463379
    },
    {
      "auxiliary_loss_clip": 0.01156084,
      "auxiliary_loss_mlp": 0.01023918,
      "balance_loss_clip": 1.04848409,
      "balance_loss_mlp": 1.01642227,
      "epoch": 0.9986172067576504,
      "flos": 20375750874240.0,
      "grad_norm": 2.4936858363745684,
      "language_loss": 0.78070009,
      "learning_rate": 1.8355575948758585e-11,
      "loss": 0.80250013,
      "num_input_tokens_seen": 179275330,
      "step": 8305,
      "time_per_iteration": 2.4227540493011475
    },
    {
      "auxiliary_loss_clip": 0.01134825,
      "auxiliary_loss_mlp": 0.01024285,
      "balance_loss_clip": 1.04143059,
      "balance_loss_mlp": 1.01687062,
      "epoch": 0.9987374496482896,
      "flos": 23727903724800.0,
      "grad_norm": 2.092696406343103,
      "language_loss": 0.73691463,
      "learning_rate": 1.5169901505407424e-11,
      "loss": 0.75850576,
      "num_input_tokens_seen": 179292395,
      "step": 8306,
      "time_per_iteration": 3.3225760459899902
    },
    {
      "auxiliary_loss_clip": 0.01135571,
      "auxiliary_loss_mlp": 0.01021153,
      "balance_loss_clip": 1.04548454,
      "balance_loss_mlp": 1.0143795,
      "epoch": 0.9988576925389286,
      "flos": 25044029959680.0,
      "grad_norm": 3.150302437516206,
      "language_loss": 0.74091977,
      "learning_rate": 1.228762317073695e-11,
      "loss": 0.76248705,
      "num_input_tokens_seen": 179311225,
      "step": 8307,
      "time_per_iteration": 2.500269889831543
    },
    {
      "auxiliary_loss_clip": 0.01138507,
      "auxiliary_loss_mlp": 0.01023328,
      "balance_loss_clip": 1.04582012,
      "balance_loss_mlp": 1.01616979,
      "epoch": 0.9989779354295677,
      "flos": 31285627637760.0,
      "grad_norm": 2.6241618388627823,
      "language_loss": 0.79009187,
      "learning_rate": 9.70874138195299e-12,
      "loss": 0.81171024,
      "num_input_tokens_seen": 179333135,
      "step": 8308,
      "time_per_iteration": 3.370873212814331
    },
    {
      "auxiliary_loss_clip": 0.01165362,
      "auxiliary_loss_mlp": 0.01026353,
      "balance_loss_clip": 1.04553401,
      "balance_loss_mlp": 1.01942658,
      "epoch": 0.9990981783202069,
      "flos": 19573398823680.0,
      "grad_norm": 2.2544950448494077,
      "language_loss": 0.74613214,
      "learning_rate": 7.433256530076093e-12,
      "loss": 0.76804924,
      "num_input_tokens_seen": 179353090,
      "step": 8309,
      "time_per_iteration": 2.4445178508758545
    },
    {
      "auxiliary_loss_clip": 0.0111342,
      "auxiliary_loss_mlp": 0.01021633,
      "balance_loss_clip": 1.04081964,
      "balance_loss_mlp": 1.01496065,
      "epoch": 0.9992184212108459,
      "flos": 17199667514880.0,
      "grad_norm": 2.1625161308300807,
      "language_loss": 0.75373125,
      "learning_rate": 5.46116896038562e-12,
      "loss": 0.77508187,
      "num_input_tokens_seen": 179367500,
      "step": 8310,
      "time_per_iteration": 3.2368834018707275
    },
    {
      "auxiliary_loss_clip": 0.0113511,
      "auxiliary_loss_mlp": 0.01026915,
      "balance_loss_clip": 1.04501009,
      "balance_loss_mlp": 1.01980472,
      "epoch": 0.999338664101485,
      "flos": 46497853681920.0,
      "grad_norm": 1.9497580178527667,
      "language_loss": 0.61725223,
      "learning_rate": 3.792478972197699e-12,
      "loss": 0.63887244,
      "num_input_tokens_seen": 179388085,
      "step": 8311,
      "time_per_iteration": 2.6791017055511475
    },
    {
      "auxiliary_loss_clip": 0.01162768,
      "auxiliary_loss_mlp": 0.01019195,
      "balance_loss_clip": 1.04503465,
      "balance_loss_mlp": 1.01223326,
      "epoch": 0.9994589069921241,
      "flos": 15158253859200.0,
      "grad_norm": 3.0613174986916745,
      "language_loss": 0.70076668,
      "learning_rate": 2.4271868181990895e-12,
      "loss": 0.72258627,
      "num_input_tokens_seen": 179405250,
      "step": 8312,
      "time_per_iteration": 2.390583038330078
    },
    {
      "auxiliary_loss_clip": 0.01152469,
      "auxiliary_loss_mlp": 0.01023196,
      "balance_loss_clip": 1.04528689,
      "balance_loss_mlp": 1.01573968,
      "epoch": 0.9995791498827632,
      "flos": 12531460256640.0,
      "grad_norm": 2.1285603506656257,
      "language_loss": 0.81137228,
      "learning_rate": 1.3652927060014973e-12,
      "loss": 0.83312893,
      "num_input_tokens_seen": 179420845,
      "step": 8313,
      "time_per_iteration": 2.4048800468444824
    },
    {
      "auxiliary_loss_clip": 0.01129273,
      "auxiliary_loss_mlp": 0.01027272,
      "balance_loss_clip": 1.04499984,
      "balance_loss_mlp": 1.01942825,
      "epoch": 0.9996993927734023,
      "flos": 19245175320960.0,
      "grad_norm": 2.1174869671760064,
      "language_loss": 0.6384474,
      "learning_rate": 6.067967965872612e-13,
      "loss": 0.6600129,
      "num_input_tokens_seen": 179440455,
      "step": 8314,
      "time_per_iteration": 2.47021222114563
    },
    {
      "auxiliary_loss_clip": 0.01121787,
      "auxiliary_loss_mlp": 0.01026695,
      "balance_loss_clip": 1.04453707,
      "balance_loss_mlp": 1.01979291,
      "epoch": 0.9998196356640414,
      "flos": 62952804518400.0,
      "grad_norm": 3.3841141806213604,
      "language_loss": 0.77099526,
      "learning_rate": 1.5169920497548615e-13,
      "loss": 0.79248011,
      "num_input_tokens_seen": 179465075,
      "step": 8315,
      "time_per_iteration": 3.634526014328003
    },
    {
      "auxiliary_loss_clip": 0.01104206,
      "auxiliary_loss_mlp": 0.01015116,
      "balance_loss_clip": 1.02576852,
      "balance_loss_mlp": 1.01107836,
      "epoch": 0.9999398785546805,
      "flos": 50922375073920.0,
      "grad_norm": 1.0914396496542906,
      "language_loss": 0.54964697,
      "learning_rate": 0.0,
      "loss": 0.57084024,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "time_per_iteration": 3.0176162719726562
    },
    {
      "epoch": 0.9999398785546805,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "total_flos": 6.996749092776837e+17,
      "train_loss": 0.7892607923968014,
      "train_runtime": 23405.2918,
      "train_samples_per_second": 14.213,
      "train_steps_per_second": 0.355
    }
  ],
  "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
}